As a young welder I worked at an outfit that repaired tankers that had been damaged or needed refurbishing. We lengthened frames, sometimes on brand new trucks. We cut the frame 90 degrees and butt spliced the extension piece. We backed it up with plates on the sides and stress relieved the area. That’s all we did, right or wrong and never had a problem.
All very good but why did the reinforcing plates go on the sides (webs) where the stresses are lowest. Stress relieving is great but very unusual on thin sections.
We heated the area evenly with a rosebud and let it cool. I was following the boss’s instructions, 1970, I didn’t know jack about it at the time. Still don’t…
@johna7661 I have heard of that referred to as normalizing, it does exactly what you were told it does. It allows the grain of the steel relax and find the spot it wants to be in. Just be weary of warpage in longer parts.
I am a welder for a Recreational Vehicle manufacturer in the USA. The plant I work at does predominantly motor homes. On some of the class A and Class C units we modify the frames by cutting them in half and adding or removing sections of the frame. When we cut a frame the cut is straight across the frame at a 90 degree angle or straight up and down much like you are advocating here. When the frame is welded back together there is a section of metal that fits on the inside of the "C" channel much like a sister rafter welded in place on the inside of the frame and if the chassis is shortened the two frame halfs are butt welded together. If the chassis is being lengthened than there is a section of steel of the same profile butt welded to the frame halfs on each end. The only reason I could see using an angle cut like you have discussed is if it is a tubular or box section frame and you make the cuts in opposite direction on the inside and outside. But I was also taught on most steel if the weld is done correctly it should be stronger than the metal itself. I've welded a few things together only to have them break in an area adjacent to the weld, but this is on repair work, not new fabrication. Sometimes we welders like to layout the weld and make it more complicated than it has to be if for no other reason to show off our skills and convince those looking at the weld in the future that we knew what we were doing when we did it.
Well, RVs do not have to abide by Federal Motor Carrier standards, but we lengthen semi truck frames and the FMCSA DOES NOT recognize or approve of vertical butt welds. We use the Z cut only.
@@TheRoadhammer379 I'd be very interested to engage with the FMCSA engineers who drafted their regs to understand why they are at odds with the rest of the structural fabrication industry. Or maybe the regs were drafted by unqualfied amateurs like the NZ Hot Rod Association (as related to me by a NZ Engineer).
@MrBlackbutang they get sued if they f up. Was a service manager for Peterbuilt and Kenworth. We were only allowed to follow company procedures. Finished up As Maintenance Manger at 4 of the largest oil field ports in the world, always did what the manufacturer recommended. Never sued , never lost a piece of equipment to poor repair. Lots of shape minds made decisions long before I came along.
As a retired mechanical engineer it was wonderful to be reminded of Mohr’s circle which I studied 45 years ago as part of Statics, which is the realm of Civil Engineers who design and build structures where the dynamic load s are small compared with the static loads. For statically loaded structures the failure mode you considered, plastic deformation is appropriate. However, dynamically loaded structures often fail catastrophically by fatigue cracking. Design and fabrication of joints to be resistant to fatigue cracking is a specialised discipline that is suggest you investigate further, perhaps in your next video. In light duty motor vehicles, particularly the off-road variants, it is not uncommon for fatigue to be the most relevant design failure mode.
Hi, pleased to have you on board. You are preaching to the converted when talking about fatigue (my background - PhD in Fracture Mechanics and more than 4 decades in offshore engineering where everything gets fatigued 24/7). The dudes who promote the / and Z profile joints generally say they are "way stronger" so I decided to refute the argument based on static strength. To keep it simple I didn't get into fatigue and torsion but maybe I should make a second video to cover these too. I think the statistics are that about 90% of all failures in metal structures and components across the board are due to fatigue.
Hello! Interesting discussion. You have a lot more experience than I have so feel free to correct me if I’m wrong. I fail to see how a Z-joint or an angular joint would make a structure more fatigue resistant. In my experience the fatigue cracks start at either a defect or at the point of highest stress concentration. I fail to see how a Z-joint or any other shape would help with that. As pointed out in other comments a longer weld only increases the risk of defects. A correctly performed weld is as strong or stronger than the base material. A longer weld on the side of the member as a z- or angular joint will do nothing to improve the fatigue or yield strength. The only reliable way I know of is to avoid sharp corners, add re-enforcing plates where the stress is highest (top and bottom as shown in this case) and make sure you have a flawless weld. I don’t work as a mechanical engineer so even though I have studied mechanical engineering a long time ago, it’s not my profession. I do however work in the aerospace industry, and I know for a fact that we use riveted butt joints with splice plates to join members because, among other reasons, it’s difficult to guarantee the quality of a welded joint and fatigue cracking is a major concern. I have never seen a z- or angular joint in an airframe.
A significant factor not considered at all in this academic exercise is heat stress & that effect on fatigue already built in with the welding.. nothing quite like a REAL test.. engineers! & their bloody maths.. they need to learn to weld themselves & understand the materials in reality.. double plating is a sticking plaster solution to piss poor work & design
@@Errol.C-nz So you don't need engineers with their bloody maths to design anything. Just leave it to welders to get on with fabricating things and it'll be OK then.
I Stumbled on this video, Its nice to see someone using Mohr's circle in a practical example! I primarily use EC (Euro Code) myself and think the joint configuration has more to do with the allowable fatigue stress. The maximum stresses, as you very nicely calculated and presented, are irrelevant of the joint cut. However the fatigue stress is greatly dependent on the joint used. According to detail categories within the tables of chapter 8 in EN1993-1-9, longitudinal welds in joined members will have a higher Δσ (allowable fatigue stress range) than perpendicular (butt) welds. Using cover plates (fish plates or stretcher plates) further increases the fatigue resistance of the joint. Furthermore, i think the welding requirements for a full section butt weld are often neglected and dictate greatly the final strength of the joint (and for closed sections is nearly impossible to properly implement). The joint configuration to use is for me debatable, but to me the problem is more to do with fatigue resistance rather than strength analysis
The purpose of this video was to debunk the myth touted by the chassis welding dudes that those exotic joint configurations are somehow "stronger" than conventonal butt welds. Given the lack of disagreement in the comments I think that has largely been achieved. You are of course correct that in practice fatigue will govern. Clearly a longitudinal butt weld will have better performance than a transverse butt weld but this isn't going to help much when you need to splice two lengths of beam together. In a new build the manufacturer will locate the weld in a low stress location (ideally at a point of contraflexure) and will have no problems. If the weld has to go into a high stress location then it needs to be very sound and suitably detailed. A double sided butt weld free from defects with modest weld reinforcement of smooth profile will perform almost as well as the parent metal. This can be achieved with a single sided weld but is much harder (all offshore platforms made from tubulars have these welds and they perform well). A practical solution as you say is to use reinforcing plates to reduce the general stress level but these must be suitably detailed too. I hope we agree on this.
G'day Mate, No the fracture was a direct result of severe corrugations over a long period of time & a reluctance of the previous owner to deflate the tyre pressure. I have not had any trouble since purchasing the truck & carrying out the repairs, plus even though we run on 10 X 20" split rims I always reduce our tire pressure from 100 Psi down to 70 Psi to help cushion the vibracions.
@@defendermodsandtravels it may perform as well as the parent material with regards to strength, but not with regards to fatigue. A welded joint increases fatigue since the weld material is a different tensile strength, and there will always be a heat affected zone that partially anneals the material, so there are discontinuities along and near the weld. A welded joint puts a low ductility weld next to a high ductility heat affected zone which is a recipe for fatigue. Fatigue in bending like this would want to follow a vertical line down the chassis rail. Welding vertically gives a crack a path to follow along the highest stress plane. A Z profile gives a hard barrier to crack propagation, and keeps less of the joint within the highest stress plane. It seems to me that fatigue is a convincing argument, especially once you consider that this is for chassis modifications on existing structures where you can't assume perfect material, perfect weld, perfect joint, etc. It doesn't help that as far as I'm aware these welds aren't tested other than visually. All that being said, a simple vertical butt joint is probably perfectly acceptable, especially on a passenger vehicle. Personally, I think fishplating is more important than the actual joint geometry.
You are correct! Look at oil ships construction . Rarely do you see anything other than a vertical or horizontal be weld . ships must flex loaded or unloaded must be loaded and unload due stress on cargo areas. But at Sea they encounter forces that are incredible from multiple directions at the same time. Thank you for the time explaining stress and showing the process of figuring out.❤❤❤❤
Yes I am right. I have been involved in steel framed buildings, bridges, fixed offshore platforms and floating offshore platforms (converted tankers) for decades. I have never seen one of these exotic joint configurations proposed by the chassis welding dudes. They seem to know something the structural engineering industry doesn't. I wait to hear their technical case.
wonderful video. As a long term off-roader, I have read in numerous articles the same as you heard. But in most of these articles, the evidence cited was in heavy competition rigs there were broken frames and the breaks were discovered to be at the frame welds.
You can support his findings but you would be proven wrong. Why does the federal agency that oversees semi truck regulations and inspections not approve of vertical butt welds for semi truck frame splices? Our facility has performed over 5000 frame lengthening splices and all require a Z cut. But hey, you are a welder and math genius, and I perform frame splices in real world day to day work, so I guess hauling 80,000lbs, over hundreds of thousands of miles, withstanding tens of thousands of pounds of torsional frame twist from torque with zero failure is in significant.
@@TheRoadhammer379 Perhaps because they follow some old wives tale and "that's how we've always done it", without coming up with any real engineering standards.
@@TheRoadhammer379 ... the effective use of an equally strong cut and weld is not a counter example. It's an example of what also works - but requires extra cutting and fitting. As a result, you should feel very confident that the method you are using is just as right as any other.
I'm confused, all truck frames/chassis say DO NOT WELD. We would strip the truck down to just a chassis and the run it through the crush until back to speck, but not all were repairable. As far as lengthening they would be done along the rows of multiple holes in the chassis. Two plates one either side (along a vertical plane) and the next hour or so of nipping and then tensioning them down. Over the years I have seen so many different ways of doing this on boxed chassis . What I can't understand is why we don't use a process of drawing back the welded area. A 1943 GMC 6x6 weapons carrier had been modified by the armed forces in 1944. It had been cut 18°from vertical with a fish plate ( stick welded) on the inside and a boxed channel on the inside of the chassis 6" past the fish plate, but the ends had 2" holes and they been cut through the centers. They then used a drawing back chalk with oxy acetylene. I had a lengthy chat with the chap who built it and restored it after it had seen action in Europe. There was quite a storey that went with the modification. The kit was sent from the States and fitted in the UK in under a few hours.
@@TheRoadhammer379 Great to have someone who actually does this chime in. As far as I'm concerned, over built is better than theoretically correct. (but, I'm just a mechanic so my opinion is worthless) I've had to repair far too much stuff that met engineering standards and design protocols which 'should not break until xxxxx hours/miles/Km', but, it does, sometimes within hours of putting in service I'm sure most engineers know they are not infallible and can't think of everything, but, I have met a few who think like surgeons - no one can possibly know any better than they do. ( like the joke, what 's the difference between God and a surgeon? - - - - - - - God doesn't think he's a surgeon)
@@OverThickness I think it was an F1 driver who said "Shut up, I know what I'm doing" when he received a lot of irritating radio messages from his team during a GP. I forget who he was. Yes I qualified as a Chartered Engineer about 45 years ago. Designing a butt joint in a chassis rail isn't the most difficult technical challenge I ever faced.
I'd forgotten all the techy stuff I learned at college and the equations were going over my head a bit, but I was just starting to think "I bet this is a hangover from those 'wood botherers' habits ?" and had a little chuckle when you came to the same conclusion !
@@defendermodsandtravels You only have to look at how the Ironbridge in Telford was built (the first major bridge to be built with cast iron), to see how much influence the chippies had in the joint work. You can also see how over-engineered it was, given their limited understanding of cast iron's performance.
I agree with you. I have extended a truck chassis using a square butt joint. This happened in Queensland, Australia. about 1970. Preheating (warming) , and low hydrogen electrodes used.
Well ! Very interesting indeed. Whilst I have a 50 year history of electronic and electrical engineering, I'm convinced by your analysis of the chassis joints. I would have thought an angle joint would be required. Not any more !. Thanks a lot for the video. Colin ( Wakefield )
I agree completely, while I wasn’t able to follow all your maths , I can agree from nearly 20 years experience on and off of doing chasing repairs that all the straight mig welded but joints iv done have been so far completely trouble free
Square cut seams to be top notch for stationary items. Avation chassis repairs, eg dented portion of tube, are called out for angle cuts and fish plates. This is where some of the idea that angle cuts are better. The problem i see with this approach is that round tube , Square and c channels are different in there bending properties. Thank you for the clear explanation. It will beapplyed when i splice a chevy diesel 4x4 front frame section into my 2wd 1st gen dodge Cummins
I served my apprenticeship in the 70s on Foden trucks, and their inhouse built chassis were legendry. The only inclined joint in the chassis was for flitch plates ends, and these were only bolted in. So I have always butt welded my chassis repairs!
As good a reason as I can think, the test of time is always my consideration and seldom do manufacturers continue to use a flawed design if it results in warranty claims against faulty products !
Hmmm, allow me to chime in here with my experience as a DIY and Professional Engineer now retired. I had a case where an overhead travelling crane was shut down by the DoL and I got a call for rescuing the little fabricator. To cut a long story short, the main beam of the overhead travelling hoist had a perpendicular "square" welded joint in the middle of its 30 ft span! I advised the owner to contact an inspection firm that specialized in structural steel examination and have them do a magnetic particle examination of the weld on all surfaces of the beam, and to evaluate any indications in compliance with the appropriate code. If all was well with this examination, it was to be followed by radiographic examination of the weld to insure that the weld interior was also sound metal (No partial penetration welds here!). Again, any indications were to be evaluated in accordance with the applicable code. Only with this examination and sound results would I certify the load capacity of this hoist. Moral of the story: For a butt joint to be of the same or better strength than the parent metal, the weld must be examined to ensure its quality. And structural welds for buildings, bridges, oil platforms etc. are thusly examined to ensure this required quality. The same cannot be said for DIY work on automotive frames. Here a longer weld with greater cross sectional area is worthwhile to make up for lower quality welding and unknown steel properties. As an aside, LENGTHENING a chassis without due analysis using known values as shown herein, is wrought with its own difficulties.
With the correct filler and proper welding techniques you are right. The reason some area's do this is because poor and improper welding. More overlap gives more weld area to bind it together. Also long term fatigue is probably being considered. In my area it's also common to see reinforcing plates welded on over the joint.
The most important point to remember is the overall strenght depends on the base materiel. In an old rusted frame, this can be a concern. As a precaution, it might be interesting to double up the material. So adding a backing plate might be more useful than the type of joint.
If someone wants to add a doubler plate then fine but it sticks in my craw when I see huge plates in the area of lowest stress (the web) and no reinforcement on the flanges.
@@defendermodsandtravelsonly problem is if you plate the flange, how do you attach it to existing rail? You can't weld or drill the flange. Kenworth is known for have just ONE bolt in the lower flange up front for the spring hanger and eventually the flange cracks and goes up through the web. Plus if you plate the flange now no brackets will line up and re-drilling may bring the new holes close to the flange. I'm no expert but that's just my 2 cents but you're right most all the stress is on the flange. I think the only safe way really would be if you need more strengh do a double frame like in dump trucks but it must be run most of the length or else it'll cause a crack where it ends due to flexing
Early in my apprenticeship 70s there wasn't much general metal engineering knowledge amongst mechanical engineers i could call on . Kenworth had chassis repair configerative data scenarios in their workshop manuals , all of which had angle welded joints with in and out gusset plates . All truck manufacturer's stated do not weld top and bottom flanges. As an Aussie i wrote BHP and asked their engineers opinion quite comprehensively butt joint with specific rods and welding process All of the chassis i welded were butt welded and crack repairs were x-ray checked for welding defects.
Used to work in an auto frame shop. Repairs of a rusted or broken frame usually require patch pieces, and these in sometimes very difficult locations. There also may be unmeasurable consequences to the welding heat of the strength of the parent metal. Consequently, the experts I observed (I am not one for this kind of work) used a strategy of smaller patches and much weld bead to achieve multiple connections of surfaces and, of course, best penetration of the weld, along with a distribution of load among several weld joints. Additional considerations arise from the particular channel cross section of the frame, its thickness, and accessibility at the point of repair. I also would suppose that the kinds of loads on a vehicle frame differ from those in a building.
I appreciate your video on frame fabrication. When I join frame rails I use angle cuts because of the length of the welds which are far longer than square cuts - call it covering the possibility of a hidden flaw in my welding execution. Thanks
Excellent presentation. Thank you for your A+ effort. Most OEM's recommend that it the joint be a clean, straight butt joint, cut perpendicular to the length of frame rail. And most OEM's will provide an acceptable "splice zone(s)". Then again, the OEM may not unless you are a vehicle up fitter.
Hey sorry if this was already answerd but I could not find it below. I'm only a novice in metal work but I've know that when you weld you pump heat into the metal and this changes the metalstructure next to the weld more or less depending on how good you are with heatmanagment. So the theory I´ve heard for this is to spread out the heated section over a longer strech. So when you go down verticly you have only a small part which was effected by the heat and not one large strech from top to bottom. I also know that normaly you just weld verticly from top to bottom on chasies, but this weld and the changed metal structure should be already calculeted into the strength by construction. The new cut and extension of the chasies ist not and you if you don´t extend it by a new longer end, but with a middle section you have two welds which chould worsen this effect. Normaly the extension are also not at the end but right under some of the most loadbarring sections because you want to extend the truck bed for example, this would also worsen the effect. I'm not a welder and never have done this or had the need to do, but from the logic side this made some sense, especialy if the welder puts a lot of heat into the metal.
Dear Felix 1st) Please go back to school and learn good English spelling (or use a dictionary). 2nd) No, with this type of joint, the weld is normally done vertical up. 3rd) Don't overcomplicate this, your "logic" is based on incomplete information.
@@tandemwings4733 if we can set aside our ugly prejudices for a moment and avoid judging someone on their spelling ability alone, he makes a very valid point. which is that unless you are a professional fabricator working in controlled conditions and subject to safety inspections, all bets are off regarding quality and safety of work carried out ie. some amateur in their garage playing around with a vehicle which they will then use on public roads. what felix is exhibiting is healthy use of intuition.
The welding electrode is designed to have a smaller crystalline structure than the parent metal so strength wise the bottom to top weld when done correctly allows the electrode to penetrate further into the chassis or parent metal to create a stronger weld. We were taught any structural weld that was vertical had to be bottom to top for that reason Top to bottom is fine on non structural welds
I will agree with the creator of this video, however, Felix makes a great point regarding weld quality. I, myself have made these alterations and repairs with minimal technical education in this field. I said minimal, not zero. I have been welding for 25+ years and tonight is an example of my level of continued adult education. I watched this video in an interest to steal some free education, whether it was good or bad (I thought it was terrific BTW). I will say that in my knowledge of and opinion of some professional welders and a a lot of the non professional welders, spreading that weld joint on angles has an added benefit in cases of sup-perfect welds that I and others might make. Expert welders who are, experts 100% of the time in control atmosphere conditions can get away with these 90 degree cuts as the engineer has correctly stated. But can the engineer himself make the quality weld needed or should he make the angular weld and plate it??? It might be more difficult to fabricate and cut angularly but not by much in my opinion. One last thing, your a jack a$$ for the comments about grammar. Please comment about mine…..
I think this comes from the days of wrought iron construction which was at first carried out as if it was wood using mortice and tenon type joints and wedges, later on when rivets were used they still used much the same as wrought iron has a grain just like wood, so two beams would be riveted together using fish plates which looked like two fish tails back to back. The idea here was that the greatest stress raiser area was in the neutral zone, this practice is still used on Lorries where chassis sections are joined or reinforced and riveted or bolted together, as to whether the term fish plate is due to its shape or not I have no idea but fish plates were used to join railway lines before other large iron or steel construction took off. I have extended lorry chassis in the past with the plug and plate method where the extension is then bolted to the vertical plates, also I have used the overlap and reinforce method with fish plates and in this case used traditional shape fish plates as it looks nice. I have never hear of the angled splice though.
You are correct in your calculations and the result of the forces involved if you loaded a fresh chassis to failure. So a 90 degree welded joints with full penetration and ground flat with no imperfections should show almost no real change in performance from the tons of load to fail It is vibration induced fatigue, the changes in the steels grain sizes of the welded joint is the reason for the inclined joint as you pointed out the load will induce a failure at 90 degrees so a tear begining at a flange will want to take the shortest route across the member in the loads plane. If a weld is there,,, it will not crack in the center of the weld but at the interface between the fine grained filler metal and the enlarged grains in the heat affected zone parent metal. Now vehical manufactures have the opportunity to test there chassis for harmonic hot spots generated by real world use. And they can put 90 degrees joints if required away from these areas if they land to close to these areas so it is crack propagation the inclined joint is trying to defeat in a one off application and bridging across a potential harmonic hot spot..
You're forgetting something......... Bridges and boats using butt connections, flex and vibrate all the time too......Also remember, just SAYING it, doesn't make it true. The man asked you to present your case with the source to back it up. Where's your source?
Excellent presentation. I, too, believe the idea of angle splice joints originated from using scarf joints to join wood members in structures. For wood spars in aircraft, this is still common practice. Wood and metal are disimilar materials, as are the bonding mechanics. In the end, the calculations say it all. Thanks for sharing your knowledge.
I was thinking the same thing that the idea originated from scarf joints for wood. Which I happened to use to fix the rotten ends of my front porch joists a couple of years back.
As the man who started all this materials testing said "Facts, Not Opinions" . Your observation around the reason this is so made think of the first Iron bridge. Yes it was cast, but look at the jointing method. It's all woodwork joints. Great video that follows that principle.
I’ve welded loads of truck & semi trailers in a fab shop in Australia , we would fold them in a large 250t press , then Weld them up , ALL straight joints. Mainly Kenworth and Mack .
All I know is the two main chassis rails on my 1976 VW type 2 have one vertical welded joint each. …still connected. When facing a problem, those of us who don’t have the technical expertise, look at the possible solutions and take our best guess at what would be stronger, in this case. And I suspect the solution that gets selected the most becomes THE solution. Great video!
You Sir, are a very watchable guy. I of course could not follow the equations exactly, but was able to follow the context. The end conclusion, transition from wood to steel construction. 👍🏻
As an engineer, I entirely agree from a stress analysis point of view. I suspect that the guys advocating for splice style joints are doing so more because (in a "working in your shed" context) 1) the longer joint offers more opportunities to clamp the two sides to ensure good alignment, which would not so much make it stronger than a properly executed full penetration butt weld, but reduce the risk of misalignment between the two sides that would introduce various problems with stress concentrations and load eccentricity etc that would weaken it, and 2) a longer weld at a gentle angle is closer to a horizontal position weld (and is therefore easier for an amateur to get a good weld) than the vertical up or vertical down alternative for welding the beam web. Obviously this isn't a concern in a factory for a properly trained welder, but for the guy in his shed working in an awkward position it may be easier, and therefore it may reduce the risk of a poor weld, and again this would potentially reduce risk of failure from introduced defects. Not so much making it stronger as reducing risk of making it weaker. Of course, I am not advocating for back yard DIYers to attempt chassis mods; I can't think of a more dangerous field of endeavour for an MBA (mediocre but ambitious) welder... but people do try it, and in that backyard world, the splice joint might in fact have a slightly lower chance of failure and hence be part of the accepted wisdom for a reason. But (to repeat for emphasis) obviously I would infinitely prefer the DIY guy to use some common-sense and bring in a qualified welder for the frame splice welding at least.
I would point out yet again that any failure is likely to emanate from the top or bottom flange where the highest stresses will be (unless there are attachments to the web with high local stresses). If you can get the flange welds right then the web butt shouldn't be a problem. I think the reasoning is simply that if the weld is longer then it must be stronger, which is a seductive if erroneous argument.
Nice one good to know, thank you. I'm not an expert, just an industrial designer who has worked several years in metal fabricating (designing) the only reason I can think of for staggered joints is that in some cases it helped for alignment, but then again as you say not for the final strength of the joint. I agree, weird joint shapes - scarfing, etc. - are a left over from timber joints, which were in the old days not even glued but pegged (trenails).
Very interesting presentation. I am wondering if you have perhaps made any assumptions in your analysis which were not apparent or included in your calculations. The one I am thinking of is the tensile and shear strength of a welded joint versus pristine metal. Given that you cannot account for the quality of the weld itself in your calculations, is it possible that an angled interface offers a greater factor of safety just because the surface area of the weld is greater and could mitigate less than optimal weld penetration etc? Thank you.
Thank you for clarifying. In this case what would you suggest is the best option to mitigate less than optimal welds? I think I am a great tig welder but without xrays or cutting the joint open there is no way to be sure of integrity. @@warrenarthur5629
The AISC code (American Institute for Steel Construction) states that for a weld without defects in structural steel one may assume the weld is as strong as the parent material. This obviously doesn't apply to defective welds or to all types of metals and alloys but it covers the cases we are looking at. The AISC code is the most widely used in the world and few people would question their recommendations.
I really like your videos and the effort put into counter some of the diy "engineers" on youtube! Only reason I find for angled welds would be as a design feature to prevent stress concentration in the chassis from plates welded on to strengthen the frame.
At work, I'm building a massively overbuilt towing rig chassis for my boss out of salvaged pieces of different vehicles starting with a 79 Ford L700 frame. We grafted an engine crossmember, springs, and 4wd axle beams from a Gen 9 92-96 super duty, it's going to relieve a 6.2 Ls/Lt (whatever it is) with a 6L80 behind it. The frame will have a 73 F-100 body dropped onto it. We have found a tongue and groove pattern would be an appropriate pattern if we needed to shorten the frame. Obviously, with this type of build, we have areas where there is no real stress plus, the frame is not your typical frame. It's easily twice as thick as a typical passenger vehicle. This will be our second hand built towing rig. The first one, we learned critical lessons. The second one, we are eliminating those problems and making it far more capable.
You are correct about the timber relations to people's way of thinking when it comes to welding metallic joints, A good weld produces a homogeneous joint which is as good as the original material, I used backing plates inside the rectangular section of the chassis and rose welded them in addition to the butt weld. Never had any problem. When the local inspector/examiner saw the joints, he just smiled and said, "Well , it is never too late to learn from others!"
You clearly did a very good welding job. In industry we use the necessary welds, as defined in the design codes, but without extra reinforcement etc. If you do that your profit is gone.
Last year I was asked to do the welding for an inserted extension on a 40-ton low loader trailer (lowboy for you lot on the other side of the Pacific) and I told them I just wanted vertical (90°) cuts in the (24-inch) beams. They questioned me, but I insisted it would be suitable, plus a hell of a lot less work than joining on the 45. She's still working fine..!!!
Congratulations, you violated FMCSA regulations on cutting and repairing a semi truck or trailer chassis. Unbelievable. Our company lengthens semi truck frames and the Z is the only approved method by the Federal DOT. Who did you do the repair for so I can avoid that death trap
This subject was brought to my attention when I proposed to splice the front chassis section of one vehicle to the rear of another, a Toyota Landcruiser FJ 45. After some research (before Internet) , the understanding I arrived at, which may well be incorrect, was that there should be no vertical butt welded splice joints. As the Landcruiser uses a double C section riveted design, i chose to cut the inner and outer width a 30 centimetre stagger to create a slip joint, using the rivet holes as reference to obtain alignment. Fish plates and bolts were used to bridge the vertical cuts, welded on the horizontal. When i sold the vehicle, I felt obligated to disclose what had been done. The buyer said he had noticed but said he worked at a chassis shop, adding it was an unconventional method but had no problem with the quality of workmanship and paid the asking price.
Just starting out as a welder having completed 2 years of schooling for the job (Southern Technical College, Orlando FL.) I think the difference between welds in chassis manufacture and modification is in the heat affected zone (or lack thereof.) In manufacture, the whole chassis (including all welded joints) will be heat-treated to make the strength of the whole chassis uniform, regardless of weld location. No residual stresses or heat-softened areas remain in a properly manufactured chassis. Most mom-and-pop modification shops (at least here in Florida? I have very little hands-on experience in the industry, take this assumption with a grain of salt) tend not to have access to or knowledge of proper heat-treatment capabilities. Spreading the resulting softened base material across multiple planes reduces the chance of a bend, I imagine. Putting all the heat affected zone into a single plane would definitely reduce the strength of the joint in that plane, especially if the heat treatment of the alloy is annealed drastically by welding (such as in heat treated aluminum alloys.) 12:55, "yielding is governed by stress in top and bottom flange." If the welding process weakens the base material, the separating of those weak points (and butressing by opposing unaltered flanges) will strengthen the joint.
Where angled joints come into there own is in the repair of metallic or non-metallic components using adhesive, braze or solder. That may have found its way into the mindset of repairers. Typically, there used to be a common apprentice workshop lesson involving cutting a 10mm mild steel square bar into two pieces. The bar was cut with an angled joint to provide a joint surface of three times the area of the cross-section of the bar (10x30) and it was then brazed back together. Attempts to make it fail in bending resulted in failure of the steel and not the brazing. It is a pretty convincing lesson and can be usefully applied in many situations. However, it has no relevance to the chassis welding task considered here.
Great info. I would say that with some alloys where the HAZ might be a problem that you could make a joint such that the stressed flanges are V shaped like so... (looking down on the flange) Left member flange]]]]]] >>[[[[[[[right member flange ( hard to draw) Or maybe (looking down on the flange) Left member flange]]]]]]//[[[[[[[right member flange (flipping the scarf on it's side to increase the welded areas length) So that at no point on the stressed flange is the entire weld subjected @ 90deg to the stress at one point and one time. It would also distribute the loads over a longer welded area reducing the effect of any defect.
This is very interesting because I have wondered about this myself. I have usually replaced a rear crossmember with a straight cut at 90degrees. None that I know of have failed. There are things to consider which will affect stresses on a chassis and that is the condition of the parent material at the point where the cut is made. If a chassis rail is cut where corrosion is present, then the wall thickness is reduced and so the stress resistance before and after welding will be less because of the parent metal thickness and Heat Affected Zone (HAZ). You are most certainly not an idiot by any means, I agree with your findings because I am not educated high enough to challenge your calculations and theory. All the best for 2024 young man!
If the parent metal is corroded you do of course move or profile the weld to avoid it. I hope I am not an idiot although I am not infallible, just like everyone else. There was an abusive commenter, very ignorant, who said I was such a fool that he pitied me. I am happy to have an exchange of opinions but I don't take abuse. He has been blocked from this channel.
Early iron structures were built using wooden-structure techniques. (This tendency for practices that made sense in older technologies, even when the original need has completely disappeared could be called Technological Cheshire Cats; the cat has disapeared, but the grin remains. There's a similar phenomenon in language; Diesel engined steamships set sail; that's two generations in one phrase. Scarph joints in wood offer a larger surfaces for the application of glue than buttj oints.
Regarding timber joinery, there are also more than one wood peg or big nail scerw in them to take upp linear tension in the joint. This is used in in the ones that has "hooks " also One "simple" joint that can take bending forces on itself is This variant of the half half split joint . (Note the angled tungs) this joint can be used in boat bow /sterns also, a wodden peg is then put in the joining planes to prevent leaks. Sometimes the joining surface is at an angle but the butt ends should be done in the same way on does too. _____________________ /_________ _____________ /________ In modern timber houses you sometimes butt joint with a joining plate put in sawn groove that goes into both the joining members. This is often only for joining logs in a solid wall , with overlap in the logs above under . This can be achieved with "tung and grove" also. _____________________ ___|___ ___________ |_________ When we want linear tension to be absorbed we use lark tail Joining , often to prevent walls from buckling out join inner walls to outside walls or to prevent floor beams from creeping out of there mouting holes in the timber wall. The beams often hold the walls together to prevent buckling out in between the gavel gavel walls from the weit of the roof , this can happen if the roof is supported by top and side beams logs only and roof joists laid in lateral to them. (Sorry if some terms are not corect/ hard to understand , I am not American / English, English is not my native language. )
I've put in a comment above in relation to zig zag bracing in long span floor joists which would prevent twisting. The Japanese seem to be masters of timber joints that seem impossible to assemble !
You are producing great videos, and I like the way you are challenging "established" methods and principles, when they are clearly wrong. I think it is far better to think about each and every thing you do from first-principles, rather than blindly following convention; 20years in Automotive design and development has taught me that much. In the case of this specific analysis, the unstated but clear conclusion is that repair of chassis longitudinals must be done in a way that allows the top and bottom faces to be repaired robustly - not as-per many Series LR chassis I have seen where a repair has been made with the rear body-tub in-situ, and the welding has been very much at arm's length, blind, and at the wrong torch-angle. One chassis had a top-face joint that was nothing but a tangle of MiG-wire projection-welded either side of the repair!
Weld the top and bottom flanges correctly and chassis will do the job. And no, you simply cannot weld the top flange of a chassis rail if the tub is in place, period. I cringe when I see dudes spending a huge amount of time reinforcing the webs, claiming to be "doing the job right", when that isn't where the critical stresses are. That may earn them a load of extra subscribers but it isn't based on fact. I don't actually set out to challenge anyone. I just approach things from first principles, as an engineer, and if that conflicts with others' opinions I'm sorry but I can't do anything about it.
I actually had to weld a frame recently. Before I go further I am a engineer so I follow your math. Your using ideals thoughout your calculations. What I found fatigue cracking by most of the mounts. This involved stress relief and overly of better material to transfer the load to good material. But excellent analysis.
The time taken to evaluate the stress levels in the chassis beam are demonstrated very well. What might be missed is by applying a building structure weld code to an automotive chassis, in my opinion. Welds, or more accurately, the junction of weld and the heat affected zone (HAZ) are sensitive and need to be addressed properly. If you examine BS7608 and its associated S-N curves of the myriad welded joints, you'll see that the strength of the welded joint does not approach the strength of the parent material. In fact, for most of the included materials, a plot of the test coupon results reveals a common stress level that a welded joint should not exceed. If acknowledgement of BS7608 is applied, one would use the tensile stress level in bending or the tensile stress in the moment near the spring perch to use as a guide for creating a joint that minimizes these stresses and keeps them below that for infinite life as seen in the S-N curve. If the stress levels in the welds on the top and bottom surfaces of the beam show risk of not meeting infinite life, then proper doublers* need to be applied in order to add cross section to the high stress junctions and to create a tapered weld that doesn't run 90* to the top or bottom of the beam. To stretch a chassis by using vertical "bologna" cuts and just welding them back together is a recipe for later sadness, in my experience. *You can see several examples in the attached link on how to address tensile stresses in welds on the top of a chassis beam that undergoes loads of several G's in normal operation. Of course, known load cases and significant amounts of FEA and results interpretation led to this successful chassis design by our team. patents.google.com/patent/US10507870B2/en?inventor=chang&assignee=CNH
I will reply to the fatigue question in a subsequent video. The dudes who promote the weird weld details claim they are "way stronger" so I chose to refute it based on static strength. Yes SN curves with any sort of weld are worse than unwelded steel however this depends on the type of detail. Simple butt welds give the best performance of any type of welded joint and the fact is that if a butt weld is well made and has a good profile it will perform excellently in fatigue. My field is offshore structures which are largely made of short lengths of rolled tubulars (or cans) welded together with single sided butt welds. They are subject to fatigue loading 24/7 for their design lives of typically 30 years - orders of magnitude worse than any vehicle chassis. Cases of failure at the butt welds are almost unheard of. If there are fatigue cracks they are invariably at the joints where the stress concentration factors are commonly in the range 5-10. Be careful about doublers; they reduce the general stress level but increase the SCF and put you onto a lower S-N curve.
You do a Scarf joint if it's wood, and even then it's a bit more complicated at the ends. See the videos on boat building, oak framing. brilliant analysis.
The amount of time one could put into a discussion on rights and wrongs is crazy. Being a welder/fabricator/Welding engineering technologist/ instructor and general jack of all trades I for one agree with your points. However I do have a couple of thoughts on this, not necessarily agreeing or disagreeing with them. I think (not know) from a lot of discussions I have had over the years that by angling the cut and weld a lot of people tend to believe they are accounting for any metallurgical issues that may arise from welding on a frame that may have been heat treated to get the mechanical properties required in the frame when manufactured. Some of this being that they simply do not know how their weld has affected the base metal and not knowing how to compensate in any other way for perceived potential problems. Sometimes I have seen it done simply for accommodating bolts holes into the equation to help with alignment issues( ie. making it easier to align and measure). I for one do not believe in cheek plates. If a weld has been done with the proper procedures and filler metal then the properties of the filler metals should be higher than the base metal. The structural ductility in the heat affected zone is the question. In every weld proper preheat, interpass temperatures, and post weld treatment is probably the biggest thing to think about but is generally unknown or not thought about by the average garage mechanic. I welded a crack in an excavator (hitachi 450) boom one night in an open field on a mound of dirt 20 feet in the air at minus 35 degrees C. One inch plate cracked across the top and 3/4 inch plate cracked all the way down one side and all by myself. Preheat consisted of a tiger torch trying to keep this warm then 1/8" 9018 electrodes in a properly prepped open root weld joint. I could preheat all I wanted (not enough) then weld and I could still take my glove off and touch the weld as soon as i finished running the bead. Just a huge metal heat sink. There was no scenario in my mind that this wasn't going to break as soon as they started working with it. Can you imagine the crystalline structure of the heat affected zone. I did not put on cheek plates as what was the point to adding stress risers outside of the broken area. Did my best and 4 years later it was still working in a rock quarry with an operator trying to beat it to death.
Although I can design welds I am not a welder myself and I have great respect for those who can do it. You have clearly put a lot of thought into your trade as well as practising it. We need more like you around.
@@defendermodsandtravels Thank you I appreciate your thoughts. I take great pride in my workmanship and I truly love what I do. I should point out however that when doing frame rail repairs or extending the rails on a transport truck I still prefer to do an inclined joint. This is partly for ease of alignment and the fact that we are not able to easily test for any changes in the grain structure in the heat affected zone of the weld. The weld itself will seldom fail but the original frame next to the weld is the issue. By having that inclined cut we can often have a bolted member such as a spring hanger attached to both sections. On something like a frame, which is a dynamically loaded structure and as such is in constant movement and vibration going down the road, the inclined angle would put the stress as you calculated crossing the weld zone and not running alongside the heat affected zone. The vertical(web) in this case is really not the issue as any cracking that might occur would most likely start at each end of the weld on the flange. An easy experiment to try is to take a piece of material and put a radius on the edge (ground lengthwise to the flat bar) and bend it and then bend a piece with a sharp corner (grinding marks across the flat bar). The sharp corner will crack and cracks the will propagate across the flange. Most failures that I have witnessed in my experience happen when the weld start/stop areas have not been properly ground and blended with the base metal. Engineering aside there are so variables to consider if you truly want to look at every scenario. One little variable can change an outcome. Too much heat in the weld zone or somebody cooling their weld with water can drastically change that grain structure for example. Unfortunately this can not be taken in by those calculations. Knowledge of the theory and the practical are both required for a safe and efficient outcome.
@@defendermodsandtravels in my previous post I should mention that these are my thoughts on heavy truck frames not on passenger vehicles. I have seen a lot of welds that should not be on the roads. I love these discussions and often look forward to a differing point of view. As well as educating and being educated by someone with a differing view. Too many people these days take some of these differing points of view too critically and fail to look at the big picture. I look forward to learning everyday. I have fun knowing there is more than one way to get aa job done.
totaly agree with you mate ive always said if they can do a factory joint vertically there is no reason i cant . i have had the same thoughts as you for most of my life and have come to the conclusion that they just do these fancy double jogged joints and the 45 degree joints to try and look fancy and be special in front of the client along with some silver tongue story how there way is better . its just a fashion statement . i am a big fan of blending my welds on the radius of the chasis though makes me feel good
I'm afraid to say that those who make scarf joints in steel chasses seem to do it mainly out of ignorance. I think they then convince themselves that these joints are "stronger" and make a virtue of it to their clients. I have engaged with one or two of these dudes to suggest they consider that these joints might be unnecessary but find a blank wall of denial and some abuse aimed in my direction.
@@husq2100 In which recognised code or standard have you seen the requirement for a box section to be welded along the webs and the flanges but not at the corners? Have you ever seen a box member chassis (or any other structure) where the butt welds didn't extend around the full circumference? I am afraid that the residual stresses to which you refer are irrelevant and your advice is incorrect. Sorry.
i used to work at a land rover reconditioners and did a lot of work for land rover and army, all lengthened or strengthened chassis were butt welded and had a a plate welded top and bottom full length between spring mounts
I work in the automotive design industry and I have to admit that I'm guilty of using scarf joints in chassis design. Our justification was that, in the case of weld material being weaker than the base material (which is common when using higher strength materials, up to and exceeding 100KSI), the longer weld would be stronger than a shorter weld. I often used this technique because I was taught it was effective but I never actually did the math. Thanks for exploring the subject, I'll will do some more research.
It's a fundamental tenet of structural design that the weld metal (and the HAZ) is at least as strong as the base metal. You can find suitable welding consumables for any grade of structural steel. Although the video looked at the static strength (because most welding guys state that their fancy joints are "way stronger) in reality failure is likely to be fatigue emanating in the top / bottom flanges. This is particularly true for higher strength steels which will be designed to take higher working stresses. Remember that the fatigue damage is approximately proportional to the cube of the stress range. Good luck.
Very interesting. Thanks. On a tangent, when in the UK and dealing with, say, Land Rovers, a lengthening or shortening of the chassis ( however achieved) strictly results in a cancellation of the existing registration, whether historic or not, and a requirement to be reregistered after having passed the rather onerous biva test. Putting that out there just in case anyone thoughf they can just do as they please without possible ramifications.
@@defendermodsandtravels interesting question! Like for like repairs used to be OK as you'd expect, but lately there seems to be some evidence of this coming into question. Bear I mind dvla and vosa have been making up the 'rules' as they wish. Requests for policy documents have been avoided. Heard of a recent pre registration of an old car that went to ridiculous lengths over originality. Seems repairs could be caught up in that, particularly monocoques. There's changes coming down the line, possibly directed at the suppliers of parts and services in the modification field, ie cut off the supply. Precisely what's driving ( no pun) this is uncertain to me, probably just some motivated tree hugging jobsworth? Let's not talk ev conversions!
That seems rather strange that you would consult leading experts on stationary buildings when trying to make an argument about why you should not avoid a abrupt point of failure along the heat affected zone of a weld on a vehicle chassis that is going to be twisted and torqued repeatedly as it drives down the road. This is especially true for heat treated steel frames used on heavy trucks. That's not even the same material that they use in structural steel construction. Even with a perfect weld you are still going to have a heat affected zone on either side of the weld where the parent metal is affected in its strength. On paper this doesn't seem to make any difference at all but the main difference in theory and practice Is that in theory there is no difference. In reality truck frames are rarely treated so kindly has to not be overloaded. And when they are the stresses are concentrated on certain points in the frame and it is best to try to distribute these points as far as possible as to not create a stress riser where it is more likely that a crack will begin. The point of the diagonal cut is not so that it has the same strength that it originally did. It is to minimize the effect of the stress riser that you have created by making the joint in the first place. On a light duty vehicle like a defender, it probably doesn't make any difference. But when you're talking about a log truck that's traveling down the highway next to your family. Do you really want to take a chance on that or do you want to give it the best chance possible to avoid unnecessary deaths?
Excellent decription of findings, sir. I am a spray painter, and used to work in Alberta where truck mounted drill rigs were made to order. They were very heavy, some not road legal. When they shortened frames on transport trucks, they cut the frame on a 45% or so angle, and this was beveled and welded together. After that another rectangular shaped length would be added to the outside, AND DRILLED AND BOLTED OVER WELDED AREA(overlapping side to side). This was called fish plating. 🇨🇦🤔
That's a battleship mod in my book. All they needed to do was a competent butt weld with a bit of reinforcement (I.e. the weld cap proud of the surrounding base metal) and it would have been full strength. Easy to do on open sections where one has access to both sides of the weld, harder but doable for single sided welds. There's an interesting point about bolted fish plates. Unless one makes proper friction grip connections they will have limited effect. If the bolts are acting in shear (especially in single shear) there will be movement and the stress will be taken by the weld instead of being shared; if this starts to crack then the whole joint will start to fail. It sounds as if the company you refer to didn't have proper engineers on their staff.
@@defendermodsandtravels perhaps there was a fatigue issue where the adjoining base metal would crack as the weld was stronger and stiffer. doing what they did may have spread the stress more.
I believe you are correct that it doesn't matter if you use a vertical weld joint in the web of the frame. The problem, as I see it, is the welds you make in the frame flanges. These need to be reinforced or you may well end up with cracks forming in the HAZ of the welds as a result of fatigue loads. After all, automotive frames are continually flexing over the open road and for trucks in particular that are heavily loaded, the stresses in the frame flanges can be quite high and when you superimpose fatigue (flexing) loads as well you may eventually end up with cracks forming in the flange welds. The solution is to weld reinforcing plates over the flange welds. Ideally, the reinforcing plates should be 4 to 6 inches long and be skip welded to the flange (in tension) along the longitudinal axis of the plates. Most importantly, DO NOT weld the ends of the reinforcing plates. Using doublers or fish plates over the web welds really isn't necessary.
One thing you seem to have forgotten is torsional stress that a chassis undergoes. As someone who was studying mechanical engineering and doing a lot of truck chassis work, i know from experience that with a conventional chassis shape of top and bottom flange with one web , and no matter how many inner channels that you have to do a 45deg cut on the web and depending on what the truck will be doing quite often a 45deg cut on top and bottom. I had to do this with many 6 wheel drive Isuzu trucks that were going from Brisbane to Papua New Guinea for mining exploration. They had a nest of 3 channels . Cow of a job they were. In the end though it comes down to whether the person welding is capable of 100% penetration with no inclusions. Generally I would grind out then needle gun each weld, usually doing one side then the other to prevent overheating the steel. The needle gun does assist in stress relief.
And if I am going to look at torsion should i also look at section warping and associated stresses? I haven't forgotten them but I must keep the concepts simple so people can follow them otherwise the only people who will watch it are qualified engineers. I take it you are familiar with the fundamental difference in torsional stresses in open and closed sections? How am I going to explain that to viewers in a short video? The fact is that if the as-designed member resists applied torsion then it will continue to do so with a competent butt joint.
At the Louisville heavy truck plant, Ford welds frame sections of trucks with an overlap of frame rails. One frame section slides inside the other. The weld joint is shaped like a C to eliminate sharp corners and to reduce the odds of cracking under load. I would just copy what Ford does on OEM frames on heavy duty trucks. edit The frames are heat treated as a unit after the weld process.
@@jondor654 The front clip is one assembly, and the rear clip is a different assembly. These 2 are similar for all the models and a center section is added to build the specific wheelbase / model. Seems like the overlap was around a foot on each end but I didn't measure it. It was not over a foot and not less than 6 inches for sure. These are all super duty frames so they are definitely over built. The "C" was at the ends of each of the front and rear sub assembly parts. The center section slid into these on each end. It is all placed and clamped into a fixture by robots and welded by a mig robot. The critical piece of information is that after welding, the whole frame is heat treated.
I am no engineer by any means but from those that I have seen do this they gave some explanation that at the time made sense to me at least. More weld surface and to help prevent a potential crack to keep going and would be stopped by the change in direction. I don't think it had anything to do with the joint itself being stronger.
That isn't reality. If you have a crack starting from the highest stressed location (top or bottom flange) it will then propagate through the member. It won't follow the weld, it'll be equally happy to go through parent metal. A change of direction of weld won't stop it. If you watch the videos published by the chassis welding guys they invariably say that the fancy joints are "way stronger". I have never seen anyone justify their weld details by arguing about crack propagation.
I agree with everything you say regarding the joint and how to weld it back together. I do however, have one comment and that is crack propagation. I used to work on some highly loaded beams that flexed depending on the way the operators loaded or used them. Once a crack is initiated in the beam the crack would travel in a straight line typically from the bottom flange to the top flange in a simply supported beam where the load hung between equidistant lifting points. The idea of angling the joints comes from the fact that it’s harder for cracks to change direction and usually take the least resistive path of a straight line. I agree the cracks have to start somewhere but the case where you are making a field repair and you add scab plates on the repaired crack we usually cut the ends of the scab plates with a large radius or a forward then backward angle. This type of repair tended to be long lasting as compared to scab plates that had plain vertical welds which tended to easily propagate cracks.
Very good explanation on the subject. The only thing I can add to this is that when I was studying, we were only allowed to load the welds to 60% of the allowed stress of the beam material (I think I remember this correctly) I agree on the timberframe explanation as no-one has been able to weld timber, not that I know of.
I don't think you remember it quite correctly. A full strength butt weld is just that. However a single sided butt weld, where full penetration cannot be guaranteed, must operate at reduced stresses (5/8 was specified in BS449).
Great video, 2 things: 1) I ran some of the same calculations you did, from classical structure theory, when I was building a loading ramp. Put a 200 factory of safety in it, based on those calculations, and the truck didn’t even get 10% up the ramp before it buckled 😂. See all those calculations are only valid until the section changes… A truck frame, a rain gutter, a cardboard box… all will buckle not tear. Not that it adds any merit to the fish plate (angled laminations that look like the mouth of a fish). 2) look at how frames typically fail, they crack around holes and at mounting points from fatigue. Again, beam theory goes right out the window, other than to pick where to compromise strength by joining or putting holes in something. That’s why OEMs (and regulatory agents) use rules of thumb along side classical theory. 🎉🎉 Anyways, I commend the effort and I throughly enjoyed the debate, especially in the comments. Your video allowed 99% of the buz to be cut 😊
Sorry but you did your calculations incorrectly. There are two sorts of buckling to consider on a beam: 1. Lateral torsional buckling which depends on the distance between restraint points. Closed sections (boxes or tubes) aren't prone to this. 2. Local buckling of the plates making up the section (sometimes referred to as crippling). Structural sections are usually proportioned so this won't happen but if one makes sections out of sheet metal this can be a major factor. If your mod had been designed by a structural engineer he we would have checked for this and added bracing if necessary. This is standard. You are correct that in a vehicle chassis most failures are caused by fatigue at stress raisers. This will happen whether or not a scarf joint or a straight butt has been used. Incidentally, beam theory doesn't "go out the window". It will give the fluctuating stresses which when used with an SN curve will give a fatigue life estimate. Finally I dispute your assertion that vehicle structures are designed by rule of thumb. Vehicle manufacturers will use detailed finite element models backed up by extensive testing in order to optimise the design and cut costs. I hope this clarifies.
@@defendermodsandtravels Great response, thanks for the depth and consideration. For sure the calculations were the wrong ones, and that's why the ramp failed like a tin can; good thing it was all just for fun 😂. 2) Crippling, cool. I'll have to read up on it if I ever pick that project back up. I'm under the impression that buckling is by very definition a rule of thumb, as is fatigue cycle quantity - these are aggregated from statistics and from experience, not like it is going to fail at exactly one million cycles. FEA like you say provides insight into design but despite best efforts every auto model has it's unforeseen failure points, and they work to fix them as early as possible with test & warrantee data, then feed that back into the knowledge base. Very much rule of thumb & best practice. Again, the debate & knowledge sharing is invaluable. Thank you for setting this up!
@@owlclaw So you had two open sections (I or C) as your loading ramp? If you had put in a bit of cross bracing between the two top flanges it would have been fine. The bracing could be from L section with say 20% the area of one of the flanges arranged in a V pattern if you can visualise what I am saying. You also want one or two braces from the bottom flange to the top bracing to stop that flange from twisting sideways. If you had done that you would have been able to mobilise the full strength of the ramp members.
@defendermodsandtravels Back to the carpenters, recalling the Victorians were up on this by adding strings of top to bottom braces in the centre of long spans of floor joists to stop them from twisting. You'd think that the combination of floorboards above and laths below would make this measure somewhat redundant however?
A couple things that I have run into: Crane structures with horizontal I-beams may require a stepped weld seam if the load was high and the total length of weld was short due to a beam with a small height. In the US, an engineer would have to calculate the requirements, but I am not that kind of engineer. The other thing I have seen is for aircraft frames made of round tube: splices are angle cut and a sleeve is added, utilizing plug welds and full seam weld. I am wondering if stresses in multiple directions, including tension, are the drivers of such design? Edited after some related reading: it seems to be generally agreed that welded joints fail in shear more often than tension. Your calcs based on the dimensions don't seem to include the presence of the weld (unless I missed it)
I suspect the angled cut is a carryover from the days of wooden structural members such as frames and airplane wing spars. Wood is generally joined with a scarf joint to maximize the surface area available for gluing.
The angular or complicated joint looks pretty . Tradition may be the main reason for this type of joint, that is this is the way it has always been done where you worked or learned your trade. I just think it looks better.
Wow, some real engineering analysis. I like everything in the analysis but take exception to the fundamental assumption of uniform loading along the frame span where in actuality we have four point loads, two on the front hangers and two on the back. It will precipitate different stress values but the analysis will be the same.
It makes little difference in fact. With a UDL the bending moment diagram will be a parabola whereas with point loads it will consist of a series of straight lines between the loading points but with a similar max value. I was hardly going to bother to do a detailed survey to measure the individual body and engine / transmission mounting points, and then to estimate the point load at each location. A lot of work for a small improvement in accuracy (although you would do that if you were doing a real design). You will note that I calculated the equivalent UDLs to give the correct axle loads so they can't be far wrong. I take it you aren't an engineer because any engineer would have made the same approximation as me.
The reason is the same reason bicycle lugs are scalloped: if you change the stiffness of the frame ABRUPTLY, then you create a "stress riser" as the frame flexes in use, setting yourself for fatigue failure at the discontinuity The bridge you referenced earlier has a far more static load. SEE ALSO: DeHavilland Comet and square windows.
I am well aware of stress raisers and fatigue failures. I am not sure what your point is though. If you are referring to butt welds there will be a notch stress at the weld toe if the weld hasn't been profiled. I didn't consider the case of welded attachments.
Very interesting video (makes me nostalgic over my past as a welder). If not mentioned yet it seems to be easier to get the two parts straight to each other with an alongated fit rather than a cut fit (like a pair of light beacons used in navigation). Elongatet fit only for measuring purpose(?)
@@defendermodsandtravels As he says in the video, if it were wood the need of transfering shear strength to the other part would need an elongated joint. It's not necessary when welding metal.
First I'd like to say that this video is great. Congratulations!!! I agree with you, the direction of the cut should not influence the results. In Eurocode EN 1993-1-8, that deals with joints calculation, just mention that full penetration welds have the same resistance as the weakest material in the joint. Exactly like you said. A 45 degree (more like a zig zag) weld can be used as a crack arrester for fatigue purposes in other situations. But it's clearly not the case in this video. Very nice approach!
Sorry but welds aren't good crack arrestors, a fatigue crack will go right through if the weld is in the way (it takes the path which releases most strain energy). This is my specialist field. Happy to debate it if you have any doubts.
You can find factory approved frame splice methods on instructions for installing custom work bodies. It's interesting reading but as always weld quality is everything. My welding instructor butt spliced farm truck frames to lengthen day cab trucks with 7018 with zero problems but the beads he laid with stick looked like robot welds. He also normalized with a torch.
Very educational, there is still a Mercedes structural guide for chassis modification that requires or suggests the angled joint. It would be interesting to find out why they recommend it.
Yes it would. Out of interest I looked at the chasses of a number of trucks and semi-trailers, incliding Mercedes, and all of the butts were straight cut. I wull have to take your word for it about the manual. I find it odd that there's a factory manual on chassis mods.
From the trucking viewpoint. Vocational truck frames are modified all the time when new for various jobs. The U.S manufacturers do provide guidelines for what can and shouldn't be done when welding on frames. The manuals will claim you shouldn't weld frames and then tell you how.
@@marioreali5925read more closely, the manufacturer clearly states "do not weld or drill on frame FLANGES". The flange being the top and bottom of the C channel. In the United States, the Federal DOT ONLY approves of the Z cut but this guy is an expert on torsional twisting force in the tens of thousands of pounds of torque on semi truck frames, oh wait he isn't.
Good video. I think the scarf style joint could actually be worse than a straight butt joint. Surly the forces don’t like the welds changing at 90 degrees… I wonder if there has been some confusion with the welding joint of the chassis rail and the welding of fish plates, or doublers on to chassis rails?
As I tried to explain in the video the member will fail (if overloaded) by bending, and all that really matters is the strength of the top and bottom flanges. If you want to make it stronger add some material to the flanges, not to the web. Don't think of lines of force following the line of the weld - it doesn't work like that. There will be a stress field which is in equilibrium with the external loads and this will flow through whatever material is there (either the base metal or the weld metal). It doesn't follow the weld.
@@defendermodsandtravels I didn’t mean the line of force would change due to the joint shape. I ment the forces flowing through a change in direction of the weld. Fish plates or doublers are added for many reasons other than bending of the chassis in the vertical plane. You might be attaching link mounts or a cross member to the webs….
I extended my caravan chassis rails by 1 metre. The two rails were 50mm x 100mm x 4mm RHS. I used a butt joint weld and then welded angled fishplates either side and only welded the top and bottom edges of the plates, leaving the side faces ‘angled at 45 degrees’ free from welding. I hope this is OK because we used to extend truck chassis this way years ago.
Here in the Alberta oilfield many of the bigger heavy trucks used for moving drilling rigs had the chassis lengthen. Nowadays it is always a straight or 90 degree joint and has been for about 30 years now. But years ago I was always told by some of the older men I worked with that it had to be an angled joint.
Very interestng to hear that. The older guys will have had it fixed in their minds that an angled weld is better without being able to explain why (except perhaps with a lot of arm waving). Conditions where you live are particularly harsh because of the low temperatures. Thanks for the comment.
@@defendermodsandtravelsoh stop with the condescending bullshit. You neither live in Canada or the united states, so you are oblivious to the federal regulations on frame splices. How many comments do I have to reply to make you understand that you are not an expert in this field. Aside from the arm waving, please explain why federal regulations only approve the Z cut. Also, please explain why numerous states in the US WILL NOT PASS A MOTOR VEHICLE FOR ANNUAL SAFETY INSPECTION with a vertical frame weld repair...oh because that repair is suseptible to cracking.
@@TheRoadhammer379 I made it clear in the video that I know nothing about local regulations on chassis welding however I'd like to have a conversation with those who drafted them. You are very quick to assume that I nothing about engineering in North America. I have an engineering degree from Washington University; I am a professional member of ASME and ASCE; the design codes I have always worked to are API, AISC and AWS; I have worked on projects in the USA, principally in Houston. The comment I would make is that there's a disconnect between practice in the Civil, Structural and Energy industries on the one side and those involved in chassis repairs on the other.
This was an outstanding presentation with some very in-depth analysis. However, something you might consider is fatigue failure. Auto chassis are subjected to all manner of impact forces and vibration from rotating, and possibly out of balance, elements. Architectural constructions are not subjected to much of this except low frequency wind loads. However, a high-rise apartment building near a supersonic test sight would certainly expose the building to square wave events. A Fourier analysis of these waves will disclose that something in the design will likely go into resonance and possibly fail by the members exceeding their elastic limit. It might be instructive to analyze wrecked automobiles that had square cut modifications to try to determine if the square cut was a contributing factor. In addition, an in-dept look at Formula one racing cars and inquire about the protocols they employ for builds and the numerous modifications.
Clearly fatigue will most likely be the cause of any failure Dynamics associated with the suspension will be a factor (which is why I applied a simple DAF to the static load however the dynamics associated with the body and chassis will be minor. The guys who propose those exotic joint profiles claim they are "way stronger" so I refuted that argument based on static strength.
@@defendermodsandtravels I loved your very in-depth analysis ...we don't see much of that on TH-cam from people who actually work on cars. I hope Elon considers fatigue issues for his interplanetary missions where vibration damping i is most likely totally absent due to the absence of air. I have visions of some of the proposed structures flying apart from resonant vibrations. I sleep well at night because I have a three wheel car !!!
Welds shrink as they cool leaving the area right alongside the weld in stress. That’s why weld repairs usually fail at the edge of the weld, it’s rare for decent welds to actually in the weld bead. The idea of the more complicated joints is with the longer welds there’s more area that has to crack before failure. As a maintenance welder with over 50 years’ experience I like to fishplate my weld jobs on frames to avoid any future failures. I use plates at least 3 times longer than the height of the frame rails and attach the fishplates with only horizontal welds. The mistake that I’ve seen several times is welders will run vertical welds at the ends of the fishplate, figuring it’s stronger but in actuality they are only causing 2 more vertical stressed areas in the frame. Vertical welds on frames should be avoided if possible, especially on highly stressed truck frames.
Practical application speaks louder than mathematical speculations. Many times over I've corrected their "ideal" , to deem it actually safe to use . Overbuild for safeties sake , less failures . If a person took all this time to do mathematical equations, they would be employed long in the competitive work force .
Good stuff that got me looking. In America the Trailer Body Builders organization, GM Upfitters, and some tech articles from the SAE pretty much say the same thing. My issue is his example for Mor's is waay to close to the spring mount. But most of us hopefully will take it for what it was intended
He is not suggesting that the splice should be at the spring mount. He is doing a hypothetical study to find the maximum stresses in the beam and the maximum hogging moment is at the spring mount. This is obvious from the video.
Might get an extended half for my d2 after looking at the technical bits of this 😂 not sure if i rewlly wanna blob it together just Infront of the shocks anyway but this backs it up
The international harvester service manual instructs how to repair the heat treated c channel frame of their 1960s -1970s pickups. They instruct you to: 1) use only an arc welder, not a gas welder as it reduces the amount of heat you put into the metal (Although there are in fact ways to control that heat). 2) when replacing a damaged piece of frame cut the frame at an angle. 3) reinforce the repair with angle stock of a minimum specified thickness and a minimum specified length. 4) cut 1" holes in the angle stock and weld it around the perimeter of those holes to the frame. If you do a single vertical weld you will create a vertical anealed section. A strip of soft metal in the heat affected zone. But there is a difference in a joint from the factory. The welds in the frame are done and THEN the frame is heat treated, and thus the welds are as strong as the frame metal. (on my IH there ARE no (or at least very few) welds, its all riveted) I can however demonstrate that the argument that joint configuration in a heat treated frame does nothing for strength is wrong with a simple thought experiment. This isnt to say that i will prove the differences are enough to worry about, or that they are not enough. That is another matter. But i can establish that its not irrelevant under all configurations. The argument does not hold. The thought experiment goes like this: Consider a leaf from a 4" wide automotive leaf spring. Its a piece lf heat treated steel. Its very diffficult to break. Make a straight cut through the center of the spring perpendicular to the length. Then weld it back together. Now take an identical spring, cut it in half parallel to its length. Then weld it back together. Now take these two springs , support the ends and load them up in the middle. We dont need a computer similation to tell us which will fail first. The one cut perpendicular to the length will fail in the heat affected zone where the heat treatment of the spring is destroyed. There is a section in the center where, all across there is NO properly heat treated steel. The other spring is weakened somewhat for the same reasons, but it still has heat treated metal along its entire length. The second example still has metal that acts as a spring along its entire length while the first example has a section in the center where its just no longer a spring. Depending on how it cooled and what filler was used its going to be hard and brittile or soft and malleable, either breaking or folding. But the second example is extreme and not at all representative of a frame repair. (Although perhaps not as bad as one might think as it could represent the longitudinal section of a z shaped cut) So consider a diagonal cut from one corner of the spring to its opposite corner. That does not suddenly make it just as weak as the one that was compromised in the middle. It still has the property that there is good heat treated metal along the entire length. Now repeat the experiment by making a diagonal cut just forward of one corner, to just shy of the opposite corner. And again, just forward of where you started the last cut to just forward of where you ended. I posit that as you change the angle of the cut (and consequently also change the length) the properties of the cut and rewelded spring will smoothly transition from those of the one where the spring was cut and welded corner to corner to those of the spring that was cut perpendicular. But what of all the math? And the official structural engineering guidelines? How is it that those very smart people say with such cerrainty that the weld does not depend on that angle? The answer is, they didnt quite say that. They said that it doesnt depend on the angle, but only the strength of the underlying material. There is an implicit assumption baked into this statement that the strength of the weld is the same as that of the underlying material. This is true if your welding mild steel structural beams with mild steel filler. This is absolutely NOT true when your welding heat treated steel. The strength of the steel in the heat affected zone is going to be quite different. But wasnt i talking about a spring and isnt that quite different than a frame? Well, not really. Everything is really a spring until you load it to the point you make a permanent change to it, at which point, whether it be a leaf spring or a structural beam, it's no longer acts like a spring. So if your vehicle has a mild steel frame, them I agree. The weld should be as strong as the underlying metal if you use the right filler and the right technique, regardless of the angle of the cut. But if heat treated steel, its a different ball game.
This is 100%. Great job. Ive always understood that welding is as strong as base metal (excepting i think weld is harder?) But essentially i think youve got it, this is carpentry joinery making its way into welding. At least in this case it does no harm.
Inclined and Z joints are appropriate for wooden glue-ups where the glue line length needs to be long enough and/or a mechanical "trick" like a scarf joint needs to be produced. Metal is different. The assumption is that the welded joint approximates any other part of the metal in terms of strength.
the only reason i would not square cut is when i have a cross member in the way in 2 places, where i have to go in between them, or to miss a gusset for structural support. the other reason might be to use a piece of metal that would be too short if you have to cut it :) but then.... if in doubt reinforce it hey> yip.. over engineerd how many times do we see it in DIY jobs..i bet 100 % of the time. thank you for a entertaining video.
People assume you need slanted or stepped joints in welded steel because you you need them when gluing other materials such as wood or plastic, where the glue may not be as strong as the base material.
I’ve seen 140’ of suspended 24” seem-less .375WT pipe that was welded in the 40s by butt welds dangle in the open ditch for 3 weeks without failure. It was supposed to be supposed by skids but it was a river bed crossing.
Very interesting. I just built a car ramp from short pieces and welded them with an angled cut. My instinct was that it would be better to spread sheer and compression forces over a longer weld, than a shorter weld. I think in a perfect weld, well executed with no welding flaws, the metal of the joint is as strong as parent metal or even stronger, if the weld is proud of the parent metal. I'm wondering if, due to possible welding flaws, it would be better to spread the forces on the welds along a greater distance. I don't know, but could the axiom about mitered cuts be found in weld failure analysis and not in "ideal" mathematics?
Very interesting analysis. I do however think that unless you grossly overload the vehicle, chassis failure will be caused by fatigue and dynamic loading. Under such a failure mode the quality of the weld is vitally important. Manufacturers usually have their weld procedures and quality checks in place hence you only see vertical welds. A poor vertical weld together with fatigue failure can result in catastrophic failure. The longer weld may first give an indication of failure. A modified chassis weld will more than likely have many inclusions worsening the fatigue strength.
Something like 90% of failures in metal structures are due to fatigue which emanate in the zone of highest fluctuating stress usually initiating at a defect or stress raiser. For most bending members this will be in one of the flanges. Once the crack has started it'll just grow at an accelerating rate until fast fracture takes place. Angling the web weld at 45 deg will make no difference.
Okay, novice here, and I'd have to say the advantage to an angle weld is a longer weld length, which can compensate for margins of error in workmanship, i.e. sloppy cuts, sloppy welding, inadequate alignment. This theory seems to have been confirmed by other comments from experienced and informed persons.
A method to avoid the issues of poor welding during work on the chassis. I read that Mercedes absolutely forbid drilling hole on their sprinter chassis to stop failures at those points.
The "z" cut or the long angle cuts are more for alignment from how it was explained to me. It's easier to line the parts up straight with the longer sides.
Very interesting. I just did this to the wifes truck and I cut the splice at an unknown angle. Part of my logic was that of increased area of weld and I added plates to the lower flange figuring that tensile load would likely be the biggest factor. Another factor was that the angle made the welding easier as it gave more room to work, although the box was off the cab was still in place and close to the splice location. You show several examples in industry and I'm afraid I have to respond to that with "what a silly little engineer man". 😂 That's the bean counters not wanting to spend any labour time creating scrap. John
Your logic is quite wrong. Adding more weld metal in areas of low stress achieves nothing, particularly when a well made weld is at least as strong as the parent metal. If however it makes you feel good that you "did the job right" that's OK. You mention that I show a few examples from industry of straight cut butt welds. In 40 years I have never seen a joint which wasn't like that. There must be one somewhere but I didn't see it. From the silly little engineer.
Please don't think I'm suggesting you're wrong, you clearly showed I wasn't right. For me it was mostly about working room and manufacturers holes. It is good to see the numbers though. I have a few heavy commercial trucks that we run at 65000 kilos, double frames, square cuts and no welds, everything is riveted or bolted. Our accountant has come up with some great ideas over the years on how to "do it better"😂 Cheers and thanks, John Neither an engineer a bean counter.
you do understand the bean counters have to operate within safety parameters dictated by the 'silly engineers' in order to meet safety requirements? they can shave costs here and there, but if a bridge (for example) is to be approved for public use then it has to be safe. there are plenty of examples of what happens when engineers & designers are ignored.
As a young welder I worked at an outfit that repaired tankers that had been damaged or needed refurbishing. We lengthened frames, sometimes on brand new trucks. We cut the frame 90 degrees and butt spliced the extension piece. We backed it up with plates on the sides and stress relieved the area. That’s all we did, right or wrong and never had a problem.
All very good but why did the reinforcing plates go on the sides (webs) where the stresses are lowest.
Stress relieving is great but very unusual on thin sections.
How did you stress relieve?
We heated the area evenly with a rosebud and let it cool. I was following the boss’s instructions, 1970, I didn’t know jack about it at the time. Still don’t…
@johna7661 I have heard of that referred to as normalizing, it does exactly what you were told it does. It allows the grain of the steel relax and find the spot it wants to be in. Just be weary of warpage in longer parts.
Spot on in my opinion.
I am a welder for a Recreational Vehicle manufacturer in the USA. The plant I work at does predominantly motor homes. On some of the class A and Class C units we modify the frames by cutting them in half and adding or removing sections of the frame. When we cut a frame the cut is straight across the frame at a 90 degree angle or straight up and down much like you are advocating here. When the frame is welded back together there is a section of metal that fits on the inside of the "C" channel much like a sister rafter welded in place on the inside of the frame and if the chassis is shortened the two frame halfs are butt welded together. If the chassis is being lengthened than there is a section of steel of the same profile butt welded to the frame halfs on each end. The only reason I could see using an angle cut like you have discussed is if it is a tubular or box section frame and you make the cuts in opposite direction on the inside and outside. But I was also taught on most steel if the weld is done correctly it should be stronger than the metal itself. I've welded a few things together only to have them break in an area adjacent to the weld, but this is on repair work, not new fabrication. Sometimes we welders like to layout the weld and make it more complicated than it has to be if for no other reason to show off our skills and convince those looking at the weld in the future that we knew what we were doing when we did it.
Well, RVs do not have to abide by Federal Motor Carrier standards, but we lengthen semi truck frames and the FMCSA DOES NOT recognize or approve of vertical butt welds. We use the Z cut only.
@@TheRoadhammer379 I'd be very interested to engage with the FMCSA engineers who drafted their regs to understand why they are at odds with the rest of the structural fabrication industry. Or maybe the regs were drafted by unqualfied amateurs like the NZ Hot Rod Association (as related to me by a NZ Engineer).
All Class 8 truck manufactures have bulletins on how to repair their frames. Great place to learn what they recommend.
No one knows more about it then the manufacturer.? Because they cause their own headache’s some care others do not now it’s just your problem. ❤
@MrBlackbutang they get sued if they f up. Was a service manager for Peterbuilt and Kenworth. We were only allowed to follow company procedures. Finished up As Maintenance Manger at 4 of the largest oil field ports in the world, always did what the manufacturer recommended. Never sued , never lost a piece of equipment to poor repair. Lots of shape minds made decisions long before I came along.
As a retired mechanical engineer it was wonderful to be reminded of Mohr’s circle which I studied 45 years ago as part of Statics, which is the realm of Civil Engineers who design and build structures where the dynamic load s are small compared with the static loads.
For statically loaded structures the failure mode you considered, plastic deformation is appropriate. However, dynamically loaded structures often fail catastrophically by fatigue cracking.
Design and fabrication of joints to be resistant to fatigue cracking is a specialised discipline that is suggest you investigate further, perhaps in your next video.
In light duty motor vehicles, particularly the off-road variants, it is not uncommon for fatigue to be the most relevant design failure mode.
Hi, pleased to have you on board. You are preaching to the converted when talking about fatigue (my background - PhD in Fracture Mechanics and more than 4 decades in offshore engineering where everything gets fatigued 24/7).
The dudes who promote the / and Z profile joints generally say they are "way stronger" so I decided to refute the argument based on static strength. To keep it simple I didn't get into fatigue and torsion but maybe I should make a second video to cover these too.
I think the statistics are that about 90% of all failures in metal structures and components across the board are due to fatigue.
Hello! Interesting discussion. You have a lot more experience than I have so feel free to correct me if I’m wrong. I fail to see how a Z-joint or an angular joint would make a structure more fatigue resistant. In my experience the fatigue cracks start at either a defect or at the point of highest stress concentration. I fail to see how a Z-joint or any other shape would help with that. As pointed out in other comments a longer weld only increases the risk of defects. A correctly performed weld is as strong or stronger than the base material. A longer weld on the side of the member as a z- or angular joint will do nothing to improve the fatigue or yield strength.
The only reliable way I know of is to avoid sharp corners, add re-enforcing plates where the stress is highest (top and bottom as shown in this case) and make sure you have a flawless weld.
I don’t work as a mechanical engineer so even though I have studied mechanical engineering a long time ago, it’s not my profession. I do however work in the aerospace industry, and I know for a fact that we use riveted butt joints with splice plates to join members because, among other reasons, it’s difficult to guarantee the quality of a welded joint and fatigue cracking is a major concern.
I have never seen a z- or angular joint in an airframe.
@@tomasbengtsson5157 Agree with everything you say.
A significant factor not considered at all in this academic exercise is heat stress & that effect on fatigue already built in with the welding.. nothing quite like a REAL test.. engineers! & their bloody maths.. they need to learn to weld themselves & understand the materials in reality.. double plating is a sticking plaster solution to piss poor work & design
@@Errol.C-nz So you don't need engineers with their bloody maths to design anything. Just leave it to welders to get on with fabricating things and it'll be OK then.
I Stumbled on this video, Its nice to see someone using Mohr's circle in a practical example! I primarily use EC (Euro Code) myself and think the joint configuration has more to do with the allowable fatigue stress. The maximum stresses, as you very nicely calculated and presented, are irrelevant of the joint cut. However the fatigue stress is greatly dependent on the joint used. According to detail categories within the tables of chapter 8 in EN1993-1-9, longitudinal welds in joined members will have a higher Δσ (allowable fatigue stress range) than perpendicular (butt) welds. Using cover plates (fish plates or stretcher plates) further increases the fatigue resistance of the joint. Furthermore, i think the welding requirements for a full section butt weld are often neglected and dictate greatly the final strength of the joint (and for closed sections is nearly impossible to properly implement). The joint configuration to use is for me debatable, but to me the problem is more to do with fatigue resistance rather than strength analysis
The purpose of this video was to debunk the myth touted by the chassis welding dudes that those exotic joint configurations are somehow "stronger" than conventonal butt welds. Given the lack of disagreement in the comments I think that has largely been achieved.
You are of course correct that in practice fatigue will govern. Clearly a longitudinal butt weld will have better performance than a transverse butt weld but this isn't going to help much when you need to splice two lengths of beam together.
In a new build the manufacturer will locate the weld in a low stress location (ideally at a point of contraflexure) and will have no problems.
If the weld has to go into a high stress location then it needs to be very sound and suitably detailed. A double sided butt weld free from defects with modest weld reinforcement of smooth profile will perform almost as well as the parent metal. This can be achieved with a single sided weld but is much harder (all offshore platforms made from tubulars have these welds and they perform well). A practical solution as you say is to use reinforcing plates to reduce the general stress level but these must be suitably detailed too.
I hope we agree on this.
Mike should respond to this one.
Brilliant answer and one of the few people that have mentioned fatigue. BS5400 is another good reference on log/log curves for classes of weld.
G'day Mate, No the fracture was a direct result of severe corrugations over a long period of time & a reluctance of the previous owner to deflate the tyre pressure. I have not had any trouble since purchasing the truck & carrying out the repairs, plus even though we run on 10 X 20" split rims I always reduce our tire pressure from 100 Psi down to 70 Psi to help cushion the vibracions.
@@defendermodsandtravels it may perform as well as the parent material with regards to strength, but not with regards to fatigue. A welded joint increases fatigue since the weld material is a different tensile strength, and there will always be a heat affected zone that partially anneals the material, so there are discontinuities along and near the weld. A welded joint puts a low ductility weld next to a high ductility heat affected zone which is a recipe for fatigue.
Fatigue in bending like this would want to follow a vertical line down the chassis rail. Welding vertically gives a crack a path to follow along the highest stress plane. A Z profile gives a hard barrier to crack propagation, and keeps less of the joint within the highest stress plane.
It seems to me that fatigue is a convincing argument, especially once you consider that this is for chassis modifications on existing structures where you can't assume perfect material, perfect weld, perfect joint, etc. It doesn't help that as far as I'm aware these welds aren't tested other than visually.
All that being said, a simple vertical butt joint is probably perfectly acceptable, especially on a passenger vehicle. Personally, I think fishplating is more important than the actual joint geometry.
You are correct! Look at oil ships construction . Rarely do you see anything other than a vertical or horizontal be weld . ships must flex loaded or unloaded must be loaded and unload due stress on cargo areas. But at Sea they encounter forces that are incredible from multiple directions at the same time. Thank you for the time explaining stress and showing the process of figuring out.❤❤❤❤
Yes I am right. I have been involved in steel framed buildings, bridges, fixed offshore platforms and floating offshore platforms (converted tankers) for decades. I have never seen one of these exotic joint configurations proposed by the chassis welding dudes. They seem to know something the structural engineering industry doesn't. I wait to hear their technical case.
wonderful video. As a long term off-roader, I have read in numerous articles the same as you heard. But in most of these articles, the evidence cited was in heavy competition rigs there were broken frames and the breaks were discovered to be at the frame welds.
Well done. As a welder/fabricator (also with a math degree) I fully support your findings. I also appreciate your simplified direct presentation.
You can support his findings but you would be proven wrong. Why does the federal agency that oversees semi truck regulations and inspections not approve of vertical butt welds for semi truck frame splices? Our facility has performed over 5000 frame lengthening splices and all require a Z cut. But hey, you are a welder and math genius, and I perform frame splices in real world day to day work, so I guess hauling 80,000lbs, over hundreds of thousands of miles, withstanding tens of thousands of pounds of torsional frame twist from torque with zero failure is in significant.
@@TheRoadhammer379 Perhaps because they follow some old wives tale and "that's how we've always done it", without coming up with any real engineering standards.
@@TheRoadhammer379 ... the effective use of an equally strong cut and weld is not a counter example. It's an example of what also works - but requires extra cutting and fitting.
As a result, you should feel very confident that the method you are using is just as right as any other.
I'm confused, all truck frames/chassis say DO NOT WELD. We would strip the truck down to just a chassis and the run it through the crush until back to speck, but not all were repairable. As far as lengthening they would be done along the rows of multiple holes in the chassis. Two plates one either side (along a vertical plane) and the next hour or so of nipping and then tensioning them down. Over the years I have seen so many different ways of doing this on boxed chassis . What I can't understand is why we don't use a process of drawing back the welded area. A 1943 GMC 6x6 weapons carrier had been modified by the armed forces in 1944. It had been cut 18°from vertical with a fish plate ( stick welded) on the inside and a boxed channel on the inside of the chassis 6" past the fish plate, but the ends had 2" holes and they been cut through the centers. They then used a drawing back chalk with oxy acetylene. I had a lengthy chat with the chap who built it and restored it after it had seen action in Europe. There was quite a storey that went with the modification. The kit was sent from the States and fitted in the UK in under a few hours.
@@TheRoadhammer379
Great to have someone who actually does this chime in.
As far as I'm concerned, over built is better than theoretically correct. (but, I'm just a mechanic so my opinion is worthless)
I've had to repair far too much stuff that met engineering standards and design protocols which 'should not break until xxxxx hours/miles/Km', but, it does, sometimes within hours of putting in service
I'm sure most engineers know they are not infallible and can't think of everything, but, I have met a few who think like surgeons - no one can possibly know any better than they do. ( like the joke, what 's the difference between God and a surgeon? - - - - - - - God doesn't think he's a surgeon)
This is the most eloquent, informative and educated "Shut up, I know what I am doing" I have had the pleasure of bearing witness to in some time.
@@OverThickness I think it was an F1 driver who said "Shut up, I know what I'm doing" when he received a lot of irritating radio messages from his team during a GP. I forget who he was.
Yes I qualified as a Chartered Engineer about 45 years ago. Designing a butt joint in a chassis rail isn't the most difficult technical challenge I ever faced.
I'd forgotten all the techy stuff I learned at college and the equations were going over my head a bit, but I was just starting to think "I bet this is a hangover from those 'wood botherers' habits ?" and had a little chuckle when you came to the same conclusion !
Yup so it is. A carpenters' joint which got adopted by the chassis welding dudes.
@@defendermodsandtravels You only have to look at how the Ironbridge in Telford was built (the first major bridge to be built with cast iron), to see how much influence the chippies had in the joint work. You can also see how over-engineered it was, given their limited understanding of cast iron's performance.
I agree with you. I have extended a truck chassis using a square butt joint. This happened in Queensland, Australia. about 1970. Preheating (warming) , and low hydrogen electrodes used.
Its more a case of quality of work. I've seen some shocking welds on structural items.
Well ! Very interesting indeed. Whilst I have a 50 year history of electronic and electrical engineering, I'm convinced by your analysis of the chassis joints. I would have thought an angle joint would be required. Not any more !. Thanks a lot for the video. Colin ( Wakefield )
I agree completely, while I wasn’t able to follow all your maths , I can agree from nearly 20 years experience on and off of doing chasing repairs that all the straight mig welded but joints iv done have been so far completely trouble free
Square cut seams to be top notch for stationary items. Avation chassis repairs, eg dented portion of tube, are called out for angle cuts and fish plates.
This is where some of the idea that angle cuts are better. The problem i see with this approach is that round tube , Square and c channels are different in there bending properties.
Thank you for the clear explanation. It will beapplyed when i splice a chevy diesel 4x4 front frame section into my 2wd 1st gen dodge Cummins
I served my apprenticeship in the 70s on Foden trucks, and their inhouse built chassis were legendry. The only inclined joint in the chassis was for flitch plates ends, and these were only bolted in. So I have always butt welded my chassis repairs!
As good a reason as I can think, the test of time is always my consideration and seldom do manufacturers continue to use a flawed design if it results in warranty claims against faulty products !
Hmmm, allow me to chime in here with my experience as a DIY and Professional Engineer now retired.
I had a case where an overhead travelling crane was shut down by the DoL and I got a call for rescuing the little fabricator.
To cut a long story short, the main beam of the overhead travelling hoist had a perpendicular "square" welded joint in the middle of its 30 ft span!
I advised the owner to contact an inspection firm that specialized in structural steel examination and have them do a magnetic particle examination of the weld on all surfaces of the beam, and to evaluate any indications in compliance with the appropriate code.
If all was well with this examination, it was to be followed by radiographic examination of the weld to insure that the weld interior was also sound metal (No partial penetration welds here!). Again, any indications were to be evaluated in accordance with the applicable code.
Only with this examination and sound results would I certify the load capacity of this hoist.
Moral of the story: For a butt joint to be of the same or better strength than the parent metal, the weld must be examined to ensure its quality.
And structural welds for buildings, bridges, oil platforms etc. are thusly examined to ensure this required quality.
The same cannot be said for DIY work on automotive frames. Here a longer weld with greater cross sectional area is worthwhile to make up for lower quality welding and unknown steel properties.
As an aside, LENGTHENING a chassis without due analysis using known values as shown herein, is wrought with its own difficulties.
With the correct filler and proper welding techniques you are right. The reason some area's do this is because poor and improper welding. More overlap gives more weld area to bind it together. Also long term fatigue is probably being considered.
In my area it's also common to see reinforcing plates welded on over the joint.
if you don't remove the chassis it can be difficult to weld vertically. the overlapping splice gives a greater horizontal area to weld on.
The most important point to remember is the overall strenght depends on the base materiel. In an old rusted frame, this can be a concern. As a precaution, it might be interesting to double up the material. So adding a backing plate might be more useful than the type of joint.
If someone wants to add a doubler plate then fine but it sticks in my craw when I see huge plates in the area of lowest stress (the web) and no reinforcement on the flanges.
@@defendermodsandtravels Yes, the diamond shape backing plate makes no sense. It should be more like an hourglass!
@@defendermodsandtravelsonly problem is if you plate the flange, how do you attach it to existing rail? You can't weld or drill the flange. Kenworth is known for have just ONE bolt in the lower flange up front for the spring hanger and eventually the flange cracks and goes up through the web. Plus if you plate the flange now no brackets will line up and re-drilling may bring the new holes close to the flange. I'm no expert but that's just my 2 cents but you're right most all the stress is on the flange. I think the only safe way really would be if you need more strengh do a double frame like in dump trucks but it must be run most of the length or else it'll cause a crack where it ends due to flexing
Early in my apprenticeship 70s there wasn't much general metal engineering knowledge amongst mechanical engineers i could call on . Kenworth had chassis repair configerative data scenarios in their workshop manuals , all of which had angle welded joints with in and out gusset plates . All truck manufacturer's stated do not weld top and bottom flanges. As an Aussie i wrote BHP and asked their engineers opinion quite comprehensively butt joint with specific rods and welding process All of the chassis i welded were butt welded and crack repairs were x-ray checked for welding defects.
Used to work in an auto frame shop. Repairs of a rusted or broken frame usually require patch pieces, and these in sometimes very difficult locations. There also may be unmeasurable consequences to the welding heat of the strength of the parent metal. Consequently, the experts I observed (I am not one for this kind of work) used a strategy of smaller patches and much weld bead to achieve multiple connections of surfaces and, of course, best penetration of the weld, along with a distribution of load among several weld joints. Additional considerations arise from the particular channel cross section of the frame, its thickness, and accessibility at the point of repair. I also would suppose that the kinds of loads on a vehicle frame differ from those in a building.
I appreciate your video on frame fabrication. When I join frame rails I use angle cuts because of the length of the welds which are far longer than square cuts - call it covering the possibility of a hidden flaw in my welding execution. Thanks
Excellent presentation. Thank you for your A+ effort. Most OEM's recommend that it the joint be a clean, straight butt joint, cut perpendicular to the length of frame rail. And most OEM's will provide an acceptable "splice zone(s)". Then again, the OEM may not unless you are a vehicle up fitter.
Hey sorry if this was already answerd but I could not find it below.
I'm only a novice in metal work but I've know that when you weld you pump heat into the metal and this changes the metalstructure next to the weld more or less depending on how good you are with heatmanagment.
So the theory I´ve heard for this is to spread out the heated section over a longer strech. So when you go down verticly you have only a small part which was effected by the heat and not one large strech from top to bottom.
I also know that normaly you just weld verticly from top to bottom on chasies, but this weld and the changed metal structure should be already calculeted into the strength by construction. The new cut and extension of the chasies ist not and you if you don´t extend it by a new longer end, but with a middle section you have two welds which chould worsen this effect. Normaly the extension are also not at the end but right under some of the most loadbarring sections because you want to extend the truck bed for example, this would also worsen the effect.
I'm not a welder and never have done this or had the need to do, but from the logic side this made some sense, especialy if the welder puts a lot of heat into the metal.
Dear Felix
1st) Please go back to school and learn good English spelling (or use a dictionary).
2nd) No, with this type of joint, the weld is normally done vertical up.
3rd) Don't overcomplicate this, your "logic" is based on incomplete information.
@@tandemwings4733 if we can set aside our ugly prejudices for a moment and avoid judging someone on their spelling ability alone, he makes a very valid point. which is that unless you are a professional fabricator working in controlled conditions and subject to safety inspections, all bets are off regarding quality and safety of work carried out ie. some amateur in their garage playing around with a vehicle which they will then use on public roads. what felix is exhibiting is healthy use of intuition.
The welding electrode is designed to have a smaller crystalline structure than the parent metal so strength wise the bottom to top weld when done correctly allows the electrode to penetrate further into the chassis or parent metal to create a stronger weld. We were taught any structural weld that was vertical had to be bottom to top for that reason
Top to bottom is fine on non structural welds
I will agree with the creator of this video, however, Felix makes a great point regarding weld quality. I, myself have made these alterations and repairs with minimal technical education in this field. I said minimal, not zero. I have been welding for 25+ years and tonight is an example of my level of continued adult education. I watched this video in an interest to steal some free education, whether it was good or bad (I thought it was terrific BTW). I will say that in my knowledge of and opinion of some professional welders and a a lot of the non professional welders, spreading that weld joint on angles has an added benefit in cases of sup-perfect welds that I and others might make. Expert welders who are, experts 100% of the time in control atmosphere conditions can get away with these 90 degree cuts as the engineer has correctly stated. But can the engineer himself make the quality weld needed or should he make the angular weld and plate it??? It might be more difficult to fabricate and cut angularly but not by much in my opinion. One last thing, your a jack a$$ for the comments about grammar. Please comment about mine…..
I think this comes from the days of wrought iron construction which was at first carried out as if it was wood using mortice and tenon type joints and wedges, later on when rivets were used they still used much the same as wrought iron has a grain just like wood, so two beams would be riveted together using fish plates which looked like two fish tails back to back. The idea here was that the greatest stress raiser area was in the neutral zone, this practice is still used on Lorries where chassis sections are joined or reinforced and riveted or bolted together, as to whether the term fish plate is due to its shape or not I have no idea but fish plates were used to join railway lines before other large iron or steel construction took off. I have extended lorry chassis in the past with the plug and plate method where the extension is then bolted to the vertical plates, also I have used the overlap and reinforce method with fish plates and in this case used traditional shape fish plates as it looks nice. I have never hear of the angled splice though.
You are correct in your calculations and the result of the forces involved if you loaded a fresh chassis to failure. So a 90 degree welded joints with full penetration and ground flat with no imperfections should show almost no real change in performance from the tons of load to fail
It is vibration induced fatigue, the changes in the steels grain sizes of the welded joint is the reason for the inclined joint as you pointed out the load will induce a failure at 90 degrees so a tear begining at a flange will want to take the shortest route across the member in the loads plane. If a weld is there,,, it will not crack in the center of the weld but at the interface between the fine grained filler metal and the enlarged grains in the heat affected zone parent metal.
Now vehical manufactures have the opportunity to test there chassis for harmonic hot spots generated by real world use. And they can put 90 degrees joints if required away from these areas if they land to close to these areas so it is crack propagation the inclined joint is trying to defeat in a one off application and bridging across a potential harmonic hot spot..
This!
You're forgetting something......... Bridges and boats using butt connections, flex and vibrate all the time too......Also remember, just SAYING it, doesn't make it true. The man asked you to present your case with the source to back it up. Where's your source?
Excellent presentation. I, too, believe the idea of angle splice joints originated from using scarf joints to join wood members in structures. For wood spars in aircraft, this is still common practice. Wood and metal are disimilar materials, as are the bonding mechanics. In the end, the calculations say it all. Thanks for sharing your knowledge.
I was thinking the same thing that the idea originated from scarf joints for wood. Which I happened to use to fix the rotten ends of my front porch joists a couple of years back.
As the man who started all this materials testing said "Facts, Not Opinions" .
Your observation around the reason this is so made think of the first Iron bridge. Yes it was cast, but look at the jointing method. It's all woodwork joints.
Great video that follows that principle.
I’ve welded loads of truck & semi trailers in a fab shop in Australia , we would fold them in a large 250t press , then Weld them up , ALL straight joints. Mainly Kenworth and Mack .
@@dalelyons4366 Of course no problems.
All I know is the two main chassis rails on my 1976 VW type 2 have one vertical welded joint each. …still connected.
When facing a problem, those of us who don’t have the technical expertise, look at the possible solutions and take our best guess at what would be stronger, in this case. And I suspect the solution that gets selected the most becomes THE solution.
Great video!
Every structural iron job I have been on utilized splice plates. 0:47 It’s crazy seeing all these projects without them.
You Sir, are a very watchable guy. I of course could not follow the equations exactly, but was able to follow the context. The end conclusion, transition from wood to steel construction. 👍🏻
As an engineer, I entirely agree from a stress analysis point of view.
I suspect that the guys advocating for splice style joints are doing so more because (in a "working in your shed" context)
1) the longer joint offers more opportunities to clamp the two sides to ensure good alignment, which would not so much make it stronger than a properly executed full penetration butt weld, but reduce the risk of misalignment between the two sides that would introduce various problems with stress concentrations and load eccentricity etc that would weaken it,
and
2) a longer weld at a gentle angle is closer to a horizontal position weld (and is therefore easier for an amateur to get a good weld) than the vertical up or vertical down alternative for welding the beam web. Obviously this isn't a concern in a factory for a properly trained welder, but for the guy in his shed working in an awkward position it may be easier, and therefore it may reduce the risk of a poor weld, and again this would potentially reduce risk of failure from introduced defects. Not so much making it stronger as reducing risk of making it weaker.
Of course, I am not advocating for back yard DIYers to attempt chassis mods; I can't think of a more dangerous field of endeavour for an MBA (mediocre but ambitious) welder... but people do try it, and in that backyard world, the splice joint might in fact have a slightly lower chance of failure and hence be part of the accepted wisdom for a reason. But (to repeat for emphasis) obviously I would infinitely prefer the DIY guy to use some common-sense and bring in a qualified welder for the frame splice welding at least.
I would point out yet again that any failure is likely to emanate from the top or bottom flange where the highest stresses will be (unless there are attachments to the web with high local stresses). If you can get the flange welds right then the web butt shouldn't be a problem.
I think the reasoning is simply that if the weld is longer then it must be stronger, which is a seductive if erroneous argument.
Nice one good to know, thank you. I'm not an expert, just an industrial designer who has worked several years in metal fabricating (designing) the only reason I can think of for staggered joints is that in some cases it helped for alignment, but then again as you say not for the final strength of the joint. I agree, weird joint shapes - scarfing, etc. - are a left over from timber joints, which were in the old days not even glued but pegged (trenails).
Very interesting presentation. I am wondering if you have perhaps made any assumptions in your analysis which were not apparent or included in your calculations. The one I am thinking of is the tensile and shear strength of a welded joint versus pristine metal. Given that you cannot account for the quality of the weld itself in your calculations, is it possible that an angled interface offers a greater factor of safety just because the surface area of the weld is greater and could mitigate less than optimal weld penetration etc? Thank you.
In the bending moment sections there is no increase in weld surface area, only in the shear sections, which are not the failure point in such members.
Thank you for clarifying. In this case what would you suggest is the best option to mitigate less than optimal welds? I think I am a great tig welder but without xrays or cutting the joint open there is no way to be sure of integrity. @@warrenarthur5629
The AISC code (American Institute for Steel Construction) states that for a weld without defects in structural steel one may assume the weld is as strong as the parent material. This obviously doesn't apply to defective welds or to all types of metals and alloys but it covers the cases we are looking at.
The AISC code is the most widely used in the world and few people would question their recommendations.
I really like your videos and the effort put into counter some of the diy "engineers" on youtube! Only reason I find for angled welds would be as a design feature to prevent stress concentration in the chassis from plates welded on to strengthen the frame.
Audun, there's little reason ever to add cheek plates to the (side) of a chassis member.
At work, I'm building a massively overbuilt towing rig chassis for my boss out of salvaged pieces of different vehicles starting with a 79 Ford L700 frame. We grafted an engine crossmember, springs, and 4wd axle beams from a Gen 9 92-96 super duty, it's going to relieve a 6.2 Ls/Lt (whatever it is) with a 6L80 behind it. The frame will have a 73 F-100 body dropped onto it. We have found a tongue and groove pattern would be an appropriate pattern if we needed to shorten the frame. Obviously, with this type of build, we have areas where there is no real stress plus, the frame is not your typical frame. It's easily twice as thick as a typical passenger vehicle. This will be our second hand built towing rig. The first one, we learned critical lessons. The second one, we are eliminating those problems and making it far more capable.
You are correct about the timber relations to people's way of thinking when it comes to welding metallic joints,
A good weld produces a homogeneous joint which is as good as the original material,
I used backing plates inside the rectangular section of the chassis and rose welded them in addition to the butt weld. Never had any problem. When the local inspector/examiner saw the joints, he just smiled and said, "Well , it is never too late to learn from others!"
You clearly did a very good welding job. In industry we use the necessary welds, as defined in the design codes, but without extra reinforcement etc. If you do that your profit is gone.
Last year I was asked to do the welding for an inserted extension on a 40-ton low loader trailer (lowboy for you lot on the other side of the Pacific) and I told them I just wanted vertical (90°) cuts in the (24-inch) beams. They questioned me, but I insisted it would be suitable, plus a hell of a lot less work than joining on the 45.
She's still working fine..!!!
Congratulations, you violated FMCSA regulations on cutting and repairing a semi truck or trailer chassis. Unbelievable. Our company lengthens semi truck frames and the Z is the only approved method by the Federal DOT. Who did you do the repair for so I can avoid that death trap
@@TheRoadhammer379
You don't read very well do you..??
This subject was brought to my attention when I proposed to splice the front chassis section of one vehicle to the rear of another, a Toyota Landcruiser FJ 45. After some research (before Internet) , the understanding I arrived at, which may well be incorrect, was that there should be no vertical butt welded splice joints.
As the Landcruiser uses a double C section riveted design, i chose to cut the inner and outer width a 30 centimetre stagger to create a slip joint, using the rivet holes as reference to obtain alignment. Fish plates and bolts were used to bridge the vertical cuts, welded on the horizontal.
When i sold the vehicle, I felt obligated to disclose what had been done. The buyer said he had noticed but said he worked at a chassis shop, adding it was an unconventional method but had no problem with the quality of workmanship and paid the asking price.
Just starting out as a welder having completed 2 years of schooling for the job (Southern Technical College, Orlando FL.)
I think the difference between welds in chassis manufacture and modification is in the heat affected zone (or lack thereof.)
In manufacture, the whole chassis (including all welded joints) will be heat-treated to make the strength of the whole chassis uniform, regardless of weld location. No residual stresses or heat-softened areas remain in a properly manufactured chassis.
Most mom-and-pop modification shops (at least here in Florida? I have very little hands-on experience in the industry, take this assumption with a grain of salt) tend not to have access to or knowledge of proper heat-treatment capabilities. Spreading the resulting softened base material across multiple planes reduces the chance of a bend, I imagine.
Putting all the heat affected zone into a single plane would definitely reduce the strength of the joint in that plane, especially if the heat treatment of the alloy is annealed drastically by welding (such as in heat treated aluminum alloys.)
12:55, "yielding is governed by stress in top and bottom flange." If the welding process weakens the base material, the separating of those weak points (and butressing by opposing unaltered flanges) will strengthen the joint.
Where angled joints come into there own is in the repair of metallic or non-metallic components using adhesive, braze or solder. That may have found its way into the mindset of repairers. Typically, there used to be a common apprentice workshop lesson involving cutting a 10mm mild steel square bar into two pieces. The bar was cut with an angled joint to provide a joint surface of three times the area of the cross-section of the bar (10x30) and it was then brazed back together. Attempts to make it fail in bending resulted in failure of the steel and not the brazing. It is a pretty convincing lesson and can be usefully applied in many situations. However, it has no relevance to the chassis welding task considered here.
Great info. I would say that with some alloys where the HAZ might be a problem that you could make a joint such that the stressed flanges are V shaped like so... (looking down on the flange) Left member flange]]]]]] >>[[[[[[[right member flange ( hard to draw)
Or maybe (looking down on the flange) Left member flange]]]]]]//[[[[[[[right member flange (flipping the scarf on it's side to increase the welded areas length)
So that at no point on the stressed flange is the entire weld subjected @ 90deg to the stress at one point and one time. It would also distribute the loads over a longer welded area reducing the effect of any defect.
This is very interesting because I have wondered about this myself. I have usually replaced a rear crossmember with a straight cut at 90degrees. None that I know of have failed.
There are things to consider which will affect stresses on a chassis and that is the condition of the parent material at the point where the cut is made. If a chassis rail is cut where corrosion is present, then the wall thickness is reduced and so the stress resistance before and after welding will be less because of the parent metal thickness and Heat Affected Zone (HAZ).
You are most certainly not an idiot by any means, I agree with your findings because I am not educated high enough to challenge your calculations and theory. All the best for 2024 young man!
If the parent metal is corroded you do of course move or profile the weld to avoid it.
I hope I am not an idiot although I am not infallible, just like everyone else. There was an abusive commenter, very ignorant, who said I was such a fool that he pitied me. I am happy to have an exchange of opinions but I don't take abuse. He has been blocked from this channel.
Early iron structures were built using wooden-structure techniques. (This tendency for practices that made sense in older technologies, even when the original need has completely disappeared could be called Technological Cheshire Cats; the cat has disapeared, but the grin remains. There's a similar phenomenon in language; Diesel engined steamships set sail; that's two generations in one phrase. Scarph joints in wood offer a larger surfaces for the application of glue than buttj oints.
Regarding timber joinery, there are also more than one wood peg or big nail scerw in them to take upp linear tension in the joint. This is used in in the ones that has "hooks " also
One "simple" joint that can take bending forces on itself is
This variant of the half half split joint . (Note the angled tungs) this joint can be used in boat bow /sterns also, a wodden peg is then put in the joining planes to prevent leaks.
Sometimes the joining surface is at an angle but the butt ends should be done in the same way on does too.
_____________________
/_________
_____________ /________
In modern timber houses you sometimes butt joint with a joining plate put in sawn groove that goes into both the joining members.
This is often only for joining logs in a solid wall , with overlap in the logs above under . This can be achieved with "tung and grove" also.
_____________________
___|___
___________ |_________
When we want linear tension to be absorbed we use lark tail
Joining , often to prevent walls from buckling out join inner walls to outside walls or to prevent floor beams from creeping out of there mouting holes in the timber wall.
The beams often hold the walls together to prevent buckling out in between the gavel gavel walls from the weit of the roof , this can happen if the roof is supported by top and side beams logs only and roof joists laid in lateral to them.
(Sorry if some terms are not corect/ hard to understand ,
I am not American / English,
English is not my native language. )
I've put in a comment above in relation to zig zag bracing in long span floor joists which would prevent twisting. The Japanese seem to be masters of timber joints that seem impossible to assemble !
You are producing great videos, and I like the way you are challenging "established" methods and principles, when they are clearly wrong. I think it is far better to think about each and every thing you do from first-principles, rather than blindly following convention; 20years in Automotive design and development has taught me that much. In the case of this specific analysis, the unstated but clear conclusion is that repair of chassis longitudinals must be done in a way that allows the top and bottom faces to be repaired robustly - not as-per many Series LR chassis I have seen where a repair has been made with the rear body-tub in-situ, and the welding has been very much at arm's length, blind, and at the wrong torch-angle. One chassis had a top-face joint that was nothing but a tangle of MiG-wire projection-welded either side of the repair!
Weld the top and bottom flanges correctly and chassis will do the job. And no, you simply cannot weld the top flange of a chassis rail if the tub is in place, period.
I cringe when I see dudes spending a huge amount of time reinforcing the webs, claiming to be "doing the job right", when that isn't where the critical stresses are. That may earn them a load of extra subscribers but it isn't based on fact.
I don't actually set out to challenge anyone. I just approach things from first principles, as an engineer, and if that conflicts with others' opinions I'm sorry but I can't do anything about it.
I actually had to weld a frame recently. Before I go further I am a engineer so I follow your math. Your using ideals thoughout your calculations. What I found fatigue cracking by most of the mounts. This involved stress relief and overly of better material to transfer the load to good material. But excellent analysis.
hi, I only watched this because I had never seen this 45 angled joint anywhere for any reason. So glad you presented facts showing this to be BS
The time taken to evaluate the stress levels in the chassis beam are demonstrated very well. What might be missed is by applying a building structure weld code to an automotive chassis, in my opinion. Welds, or more accurately, the junction of weld and the heat affected zone (HAZ) are sensitive and need to be addressed properly. If you examine BS7608 and its associated S-N curves of the myriad welded joints, you'll see that the strength of the welded joint does not approach the strength of the parent material. In fact, for most of the included materials, a plot of the test coupon results reveals a common stress level that a welded joint should not exceed. If acknowledgement of BS7608 is applied, one would use the tensile stress level in bending or the tensile stress in the moment near the spring perch to use as a guide for creating a joint that minimizes these stresses and keeps them below that for infinite life as seen in the S-N curve. If the stress levels in the welds on the top and bottom surfaces of the beam show risk of not meeting infinite life, then proper doublers* need to be applied in order to add cross section to the high stress junctions and to create a tapered weld that doesn't run 90* to the top or bottom of the beam. To stretch a chassis by using vertical "bologna" cuts and just welding them back together is a recipe for later sadness, in my experience.
*You can see several examples in the attached link on how to address tensile stresses in welds on the top of a chassis beam that undergoes loads of several G's in normal operation. Of course, known load cases and significant amounts of FEA and results interpretation led to this successful chassis design by our team.
patents.google.com/patent/US10507870B2/en?inventor=chang&assignee=CNH
I will reply to the fatigue question in a subsequent video. The dudes who promote the weird weld details claim they are "way stronger" so I chose to refute it based on static strength.
Yes SN curves with any sort of weld are worse than unwelded steel however this depends on the type of detail. Simple butt welds give the best performance of any type of welded joint and the fact is that if a butt weld is well made and has a good profile it will perform excellently in fatigue.
My field is offshore structures which are largely made of short lengths of rolled tubulars (or cans) welded together with single sided butt welds. They are subject to fatigue loading 24/7 for their design lives of typically 30 years - orders of magnitude worse than any vehicle chassis. Cases of failure at the butt welds are almost unheard of. If there are fatigue cracks they are invariably at the joints where the stress concentration factors are commonly in the range 5-10.
Be careful about doublers; they reduce the general stress level but increase the SCF and put you onto a lower S-N curve.
You do a Scarf joint if it's wood, and even then it's a bit more complicated at the ends. See the videos on boat building, oak framing. brilliant analysis.
The amount of time one could put into a discussion on rights and wrongs is crazy. Being a welder/fabricator/Welding engineering technologist/ instructor and general jack of all trades I for one agree with your points. However I do have a couple of thoughts on this, not necessarily agreeing or disagreeing with them. I think (not know) from a lot of discussions I have had over the years that by angling the cut and weld a lot of people tend to believe they are accounting for any metallurgical issues that may arise from welding on a frame that may have been heat treated to get the mechanical properties required in the frame when manufactured. Some of this being that they simply do not know how their weld has affected the base metal and not knowing how to compensate in any other way for perceived potential problems. Sometimes I have seen it done simply for accommodating bolts holes into the equation to help with alignment issues( ie. making it easier to align and measure). I for one do not believe in cheek plates. If a weld has been done with the proper procedures and filler metal then the properties of the filler metals should be higher than the base metal. The structural ductility in the heat affected zone is the question. In every weld proper preheat, interpass temperatures, and post weld treatment is probably the biggest thing to think about but is generally unknown or not thought about by the average garage mechanic. I welded a crack in an excavator (hitachi 450) boom one night in an open field on a mound of dirt 20 feet in the air at minus 35 degrees C. One inch plate cracked across the top and 3/4 inch plate cracked all the way down one side and all by myself. Preheat consisted of a tiger torch trying to keep this warm then 1/8" 9018 electrodes in a properly prepped open root weld joint. I could preheat all I wanted (not enough) then weld and I could still take my glove off and touch the weld as soon as i finished running the bead. Just a huge metal heat sink. There was no scenario in my mind that this wasn't going to break as soon as they started working with it. Can you imagine the crystalline structure of the heat affected zone. I did not put on cheek plates as what was the point to adding stress risers outside of the broken area.
Did my best and 4 years later it was still working in a rock quarry with an operator trying to beat it to death.
Although I can design welds I am not a welder myself and I have great respect for those who can do it. You have clearly put a lot of thought into your trade as well as practising it. We need more like you around.
@@defendermodsandtravels Thank you I appreciate your thoughts. I take great pride in my workmanship and I truly love what I do. I should point out however that when doing frame rail repairs or extending the rails on a transport truck I still prefer to do an inclined joint. This is partly for ease of alignment and the fact that we are not able to easily test for any changes in the grain structure in the heat affected zone of the weld. The weld itself will seldom fail but the original frame next to the weld is the issue. By having that inclined cut we can often have a bolted member such as a spring hanger attached to both sections. On something like a frame, which is a dynamically loaded structure and as such is in constant movement and vibration going down the road, the inclined angle would put the stress as you calculated crossing the weld zone and not running alongside the heat affected zone. The vertical(web) in this case is really not the issue as any cracking that might occur would most likely start at each end of the weld on the flange. An easy experiment to try is to take a piece of material and put a radius on the edge (ground lengthwise to the flat bar) and bend it and then bend a piece with a sharp corner (grinding marks across the flat bar). The sharp corner will crack and cracks the will propagate across the flange. Most failures that I have witnessed in my experience happen when the weld start/stop areas have not been properly ground and blended with the base metal. Engineering aside there are so variables to consider if you truly want to look at every scenario. One little variable can change an outcome. Too much heat in the weld zone or somebody cooling their weld with water can drastically change that grain structure for example. Unfortunately this can not be taken in by those calculations. Knowledge of the theory and the practical are both required for a safe and efficient outcome.
@@defendermodsandtravels in my previous post I should mention that these are my thoughts on heavy truck frames not on passenger vehicles. I have seen a lot of welds that should not be on the roads. I love these discussions and often look forward to a differing point of view. As well as educating and being educated by someone with a differing view. Too many people these days take some of these differing points of view too critically and fail to look at the big picture. I look forward to learning everyday. I have fun knowing there is more than one way to get aa job done.
@@davidkelly9218 We all learn from each other, agreed.
totaly agree with you mate ive always said if they can do a factory joint vertically there is no reason i cant . i have had the same thoughts as you for most of my life and have come to the conclusion that they just do these fancy double jogged joints and the 45 degree joints to try and look fancy and be special in front of the client along with some silver tongue story how there way is better . its just a fashion statement . i am a big fan of blending my welds on the radius of the chasis though makes me feel good
I'm afraid to say that those who make scarf joints in steel chasses seem to do it mainly out of ignorance. I think they then convince themselves that these joints are "stronger" and make a virtue of it to their clients. I have engaged with one or two of these dudes to suggest they consider that these joints might be unnecessary but find a blank wall of denial and some abuse aimed in my direction.
You should avoid welding to the radius corners of the chassis rails as stresses are already trapped here in the manufacturing process.
@@husq2100 In which recognised code or standard have you seen the requirement for a box section to be welded along the webs and the flanges but not at the corners? Have you ever seen a box member chassis (or any other structure) where the butt welds didn't extend around the full circumference? I am afraid that the residual stresses to which you refer are irrelevant and your advice is incorrect. Sorry.
i used to work at a land rover reconditioners and did a lot of work for land rover and army, all lengthened or strengthened chassis were butt welded and had a a plate welded top and bottom full length between spring mounts
I work in the automotive design industry and I have to admit that I'm guilty of using scarf joints in chassis design. Our justification was that, in the case of weld material being weaker than the base material (which is common when using higher strength materials, up to and exceeding 100KSI), the longer weld would be stronger than a shorter weld. I often used this technique because I was taught it was effective but I never actually did the math. Thanks for exploring the subject, I'll will do some more research.
It's a fundamental tenet of structural design that the weld metal (and the HAZ) is at least as strong as the base metal. You can find suitable welding consumables for any grade of structural steel.
Although the video looked at the static strength (because most welding guys state that their fancy joints are "way stronger) in reality failure is likely to be fatigue emanating in the top / bottom flanges. This is particularly true for higher strength steels which will be designed to take higher working stresses. Remember that the fatigue damage is approximately proportional to the cube of the stress range.
Good luck.
I have done it both ways as described because that is what I was taught. I thank you for this as I have always wondered why.
Which way was quicker and less work?
Very interesting. Thanks. On a tangent, when in the UK and dealing with, say, Land Rovers, a lengthening or shortening of the chassis ( however achieved) strictly results in a cancellation of the existing registration, whether historic or not, and a requirement to be reregistered after having passed the rather onerous biva test. Putting that out there just in case anyone thoughf they can just do as they please without possible ramifications.
if you tell them
Thanks for pointing that out. Would this apply if you had a corroded section and inserted a new section of the same length?
@@defendermodsandtravels interesting question! Like for like repairs used to be OK as you'd expect, but lately there seems to be some evidence of this coming into question. Bear I mind dvla and vosa have been making up the 'rules' as they wish. Requests for policy documents have been avoided. Heard of a recent pre registration of an old car that went to ridiculous lengths over originality. Seems repairs could be caught up in that, particularly monocoques. There's changes coming down the line, possibly directed at the suppliers of parts and services in the modification field, ie cut off the supply. Precisely what's driving ( no pun) this is uncertain to me, probably just some motivated tree hugging jobsworth? Let's not talk ev conversions!
That seems rather strange that you would consult leading experts on stationary buildings when trying to make an argument about why you should not avoid a abrupt point of failure along the heat affected zone of a weld on a vehicle chassis that is going to be twisted and torqued repeatedly as it drives down the road. This is especially true for heat treated steel frames used on heavy trucks. That's not even the same material that they use in structural steel construction. Even with a perfect weld you are still going to have a heat affected zone on either side of the weld where the parent metal is affected in its strength. On paper this doesn't seem to make any difference at all but the main difference in theory and practice Is that in theory there is no difference. In reality truck frames are rarely treated so kindly has to not be overloaded. And when they are the stresses are concentrated on certain points in the frame and it is best to try to distribute these points as far as possible as to not create a stress riser where it is more likely that a crack will begin. The point of the diagonal cut is not so that it has the same strength that it originally did. It is to minimize the effect of the stress riser that you have created by making the joint in the first place. On a light duty vehicle like a defender, it probably doesn't make any difference. But when you're talking about a log truck that's traveling down the highway next to your family. Do you really want to take a chance on that or do you want to give it the best chance possible to avoid unnecessary deaths?
Excellent decription of findings, sir.
I am a spray painter, and used to work in Alberta where truck mounted drill rigs were made to order. They were very heavy, some not road legal.
When they shortened frames on transport trucks, they cut the frame on a 45% or so angle, and this was beveled and welded together. After that another rectangular shaped length would be added to the outside, AND DRILLED AND BOLTED OVER WELDED AREA(overlapping side to side).
This was called fish plating.
🇨🇦🤔
That's a battleship mod in my book.
All they needed to do was a competent butt weld with a bit of reinforcement (I.e. the weld cap proud of the surrounding base metal) and it would have been full strength. Easy to do on open sections where one has access to both sides of the weld, harder but doable for single sided welds.
There's an interesting point about bolted fish plates. Unless one makes proper friction grip connections they will have limited effect. If the bolts are acting in shear (especially in single shear) there will be movement and the stress will be taken by the weld instead of being shared; if this starts to crack then the whole joint will start to fail.
It sounds as if the company you refer to didn't have proper engineers on their staff.
@@defendermodsandtravels perhaps there was a fatigue issue where the adjoining base metal would crack as the weld was stronger and stiffer. doing what they did may have spread the stress more.
@@ronblack7870 No given that any fatigue crack is likely to emanate from the flange welds.
I believe you are correct that it doesn't matter if you use a vertical weld joint in the web of the frame. The problem, as I see it, is the welds you make in the frame flanges. These need to be reinforced or you may well end up with cracks forming in the HAZ of the welds as a result of fatigue loads. After all, automotive frames are continually flexing over the open road and for trucks in particular that are heavily loaded, the stresses in the frame flanges can be quite high and when you superimpose fatigue (flexing) loads as well you may eventually end up with cracks forming in the flange welds. The solution is to weld reinforcing plates over the flange welds. Ideally, the reinforcing plates should be 4 to 6 inches long and be skip welded to the flange (in tension) along the longitudinal axis of the plates. Most importantly, DO NOT weld the ends of the reinforcing plates. Using doublers or fish plates over the web welds really isn't necessary.
One thing you seem to have forgotten is torsional stress that a chassis undergoes.
As someone who was studying mechanical engineering and doing a lot of truck chassis work, i know from experience that with a conventional chassis shape of top and bottom flange with one web , and no matter how many inner channels that you have to do a 45deg cut on the web and depending on what the truck will be doing quite often a 45deg cut on top and bottom.
I had to do this with many 6 wheel drive Isuzu trucks that were going from Brisbane to Papua New Guinea for mining exploration.
They had a nest of 3 channels . Cow of a job they were.
In the end though it comes down to whether the person welding is capable of 100% penetration with no inclusions.
Generally I would grind out then needle gun each weld, usually doing one side then the other to prevent overheating the steel.
The needle gun does assist in stress relief.
And if I am going to look at torsion should i also look at section warping and associated stresses? I haven't forgotten them but I must keep the concepts simple so people can follow them otherwise the only people who will watch it are qualified engineers.
I take it you are familiar with the fundamental difference in torsional stresses in open and closed sections? How am I going to explain that to viewers in a short video?
The fact is that if the as-designed member resists applied torsion then it will continue to do so with a competent butt joint.
At the Louisville heavy truck plant, Ford welds frame sections of trucks with an overlap of frame rails. One frame section slides inside the other. The weld joint is shaped like a C to eliminate sharp corners and to reduce the odds of cracking under load. I would just copy what Ford does on OEM frames on heavy duty trucks.
edit
The frames are heat treated as a unit after the weld process.
Interesting , belt and braces Insurance , was the C profile on both webs and if you recall what length was the overlap ?. Thanks .
@@jondor654 The front clip is one assembly, and the rear clip is a different assembly. These 2 are similar for all the models and a center section is added to build the specific wheelbase / model. Seems like the overlap was around a foot on each end but I didn't measure it. It was not over a foot and not less than 6 inches for sure.
These are all super duty frames so they are definitely over built.
The "C" was at the ends of each of the front and rear sub assembly parts. The center section slid into these on each end. It is all placed and clamped into a fixture by robots and welded by a mig robot.
The critical piece of information is that after welding, the whole frame is heat treated.
I think it's a hangover from glued joints in woodwork. Sorry commented before the end of the video. I completely agree with you
I am no engineer by any means but from those that I have seen do this they gave some explanation that at the time made sense to me at least. More weld surface and to help prevent a potential crack to keep going and would be stopped by the change in direction. I don't think it had anything to do with the joint itself being stronger.
That isn't reality. If you have a crack starting from the highest stressed location (top or bottom flange) it will then propagate through the member. It won't follow the weld, it'll be equally happy to go through parent metal. A change of direction of weld won't stop it.
If you watch the videos published by the chassis welding guys they invariably say that the fancy joints are "way stronger". I have never seen anyone justify their weld details by arguing about crack propagation.
Well I've seen many welders on big equipment say it but I guess because you haven't I must be wrong.
Thanks for correcting me.
I agree with everything you say regarding the joint and how to weld it back together. I do however, have one comment and that is crack propagation. I used to work on some highly loaded beams that flexed depending on the way the operators loaded or used them. Once a crack is initiated in the beam the crack would travel in a straight line typically from the bottom flange to the top flange in a simply supported beam where the load hung between equidistant lifting points. The idea of angling the joints comes from the fact that it’s harder for cracks to change direction and usually take the least resistive path of a straight line.
I agree the cracks have to start somewhere but the case where you are making a field repair and you add scab plates on the repaired crack we usually cut the ends of the scab plates with a large radius or a forward then backward angle. This type of repair tended to be long lasting as compared to scab plates that had plain vertical welds which tended to easily propagate cracks.
That's what I do
It seems likely this was born of the old adage, "If you can't tie a knot, tie a lot."
great video.
Very good explanation on the subject. The only thing I can add to this is that when I was studying, we were only allowed to load the welds to 60% of the allowed stress of the beam material (I think I remember this correctly) I agree on the timberframe explanation as no-one has been able to weld timber, not that I know of.
I don't think you remember it quite correctly. A full strength butt weld is just that. However a single sided butt weld, where full penetration cannot be guaranteed, must operate at reduced stresses (5/8 was specified in BS449).
I belive you are correct. Thank you Sir@@defendermodsandtravels
Woodcraft rods...
Great video, 2 things: 1) I ran some of the same calculations you did, from classical structure theory, when I was building a loading ramp. Put a 200 factory of safety in it, based on those calculations, and the truck didn’t even get 10% up the ramp before it buckled 😂. See all those calculations are only valid until the section changes… A truck frame, a rain gutter, a cardboard box… all will buckle not tear. Not that it adds any merit to the fish plate (angled laminations that look like the mouth of a fish). 2) look at how frames typically fail, they crack around holes and at mounting points from fatigue. Again, beam theory goes right out the window, other than to pick where to compromise strength by joining or putting holes in something. That’s why OEMs (and regulatory agents) use rules of thumb along side classical theory. 🎉🎉 Anyways, I commend the effort and I throughly enjoyed the debate, especially in the comments. Your video allowed 99% of the buz to be cut 😊
Sorry but you did your calculations incorrectly. There are two sorts of buckling to consider on a beam:
1. Lateral torsional buckling which depends on the distance between restraint points. Closed sections (boxes or tubes) aren't prone to this.
2. Local buckling of the plates making up the section (sometimes referred to as crippling). Structural sections are usually proportioned so this won't happen but if one makes sections out of sheet metal this can be a major factor.
If your mod had been designed by a structural engineer he we would have checked for this and added bracing if necessary. This is standard.
You are correct that in a vehicle chassis most failures are caused by fatigue at stress raisers. This will happen whether or not a scarf joint or a straight butt has been used. Incidentally, beam theory doesn't "go out the window". It will give the fluctuating stresses which when used with an SN curve will give a fatigue life estimate.
Finally I dispute your assertion that vehicle structures are designed by rule of thumb. Vehicle manufacturers will use detailed finite element models backed up by extensive testing in order to optimise the design and cut costs.
I hope this clarifies.
@@defendermodsandtravels Great response, thanks for the depth and consideration. For sure the calculations were the wrong ones, and that's why the ramp failed like a tin can; good thing it was all just for fun 😂.
2) Crippling, cool. I'll have to read up on it if I ever pick that project back up.
I'm under the impression that buckling is by very definition a rule of thumb, as is fatigue cycle quantity - these are aggregated from statistics and from experience, not like it is going to fail at exactly one million cycles.
FEA like you say provides insight into design but despite best efforts every auto model has it's unforeseen failure points, and they work to fix them as early as possible with test & warrantee data, then feed that back into the knowledge base. Very much rule of thumb & best practice.
Again, the debate & knowledge sharing is invaluable. Thank you for setting this up!
@@owlclaw So you had two open sections (I or C) as your loading ramp? If you had put in a bit of cross bracing between the two top flanges it would have been fine. The bracing could be from L section with say 20% the area of one of the flanges arranged in a V pattern if you can visualise what I am saying. You also want one or two braces from the bottom flange to the top bracing to stop that flange from twisting sideways. If you had done that you would have been able to mobilise the full strength of the ramp members.
@defendermodsandtravels Back to the carpenters, recalling the Victorians were up on this by adding strings of top to bottom braces in the centre of long spans of floor joists to stop them from twisting. You'd think that the combination of floorboards above and laths below would make this measure somewhat redundant however?
A couple things that I have run into:
Crane structures with horizontal I-beams may require a stepped weld seam if the load was high and the total length of weld was short due to a beam with a small height. In the US, an engineer would have to calculate the requirements, but I am not that kind of engineer.
The other thing I have seen is for aircraft frames made of round tube: splices are angle cut and a sleeve is added, utilizing plug welds and full seam weld. I am wondering if stresses in multiple directions, including tension, are the drivers of such design?
Edited after some related reading: it seems to be generally agreed that welded joints fail in shear more often than tension. Your calcs based on the dimensions don't seem to include the presence of the weld (unless I missed it)
I suspect the angled cut is a carryover from the days of wooden structural members such as frames and airplane wing spars. Wood is generally joined with a scarf joint to maximize the surface area available for gluing.
The angular or complicated joint looks pretty . Tradition may be the main reason for this type of joint, that is this is the way it has always been done where you worked or learned your trade. I just think it looks better.
Wow, some real engineering analysis. I like everything in the analysis but take exception to the fundamental assumption of uniform loading along the frame span where in actuality we have four point loads, two on the front hangers and two on the back. It will precipitate different stress values but the analysis will be the same.
It makes little difference in fact. With a UDL the bending moment diagram will be a parabola whereas with point loads it will consist of a series of straight lines between the loading points but with a similar max value. I was hardly going to bother to do a detailed survey to measure the individual body and engine / transmission mounting points, and then to estimate the point load at each location. A lot of work for a small improvement in accuracy (although you would do that if you were doing a real design).
You will note that I calculated the equivalent UDLs to give the correct axle loads so they can't be far wrong.
I take it you aren't an engineer because any engineer would have made the same approximation as me.
The reason is the same reason bicycle lugs are scalloped: if you change the stiffness of the frame ABRUPTLY, then you create a "stress riser" as the frame flexes in use, setting yourself for fatigue failure at the discontinuity
The bridge you referenced earlier has a far more static load.
SEE ALSO: DeHavilland Comet and square windows.
I am well aware of stress raisers and fatigue failures.
I am not sure what your point is though. If you are referring to butt welds there will be a notch stress at the weld toe if the weld hasn't been profiled. I didn't consider the case of welded attachments.
Very interesting video (makes me nostalgic over my past as a welder). If not mentioned yet it seems to be easier to get the two parts straight to each other with an alongated fit rather than a cut fit (like a pair of light beacons used in navigation). Elongatet fit only for measuring purpose(?)
If it makes it easier to align cut it how you wish, but don't think it will be stronger.
@@defendermodsandtravels As he says in the video, if it were wood the need of transfering shear strength to the other part would need an elongated joint. It's not necessary when welding metal.
I agree with you but your forwared remarks were pointing out but joints not splices .but I appreciate your point .
First I'd like to say that this video is great. Congratulations!!!
I agree with you, the direction of the cut should not influence the results. In Eurocode EN 1993-1-8, that deals with joints calculation, just mention that full penetration welds have the same resistance as the weakest material in the joint. Exactly like you said.
A 45 degree (more like a zig zag) weld can be used as a crack arrester for fatigue purposes in other situations. But it's clearly not the case in this video.
Very nice approach!
Sorry but welds aren't good crack arrestors, a fatigue crack will go right through if the weld is in the way (it takes the path which releases most strain energy). This is my specialist field. Happy to debate it if you have any doubts.
You can find factory approved frame splice methods on instructions for installing custom work bodies. It's interesting reading but as always weld quality is everything. My welding instructor butt spliced farm truck frames to lengthen day cab trucks with 7018 with zero problems but the beads he laid with stick looked like robot welds. He also normalized with a torch.
The correct way to make the joint work is my way.
Thank you
Very educational, there is still a Mercedes structural guide for chassis modification that requires or suggests the angled joint. It would be interesting to find out why they recommend it.
Yes it would. Out of interest I looked at the chasses of a number of trucks and semi-trailers, incliding Mercedes, and all of the butts were straight cut.
I wull have to take your word for it about the manual. I find it odd that there's a factory manual on chassis mods.
From the trucking viewpoint. Vocational truck frames are modified all the time when new for various jobs. The U.S manufacturers do provide guidelines for what can and shouldn't be done when welding on frames. The manuals will claim you shouldn't weld frames and then tell you how.
@@marioreali5925read more closely, the manufacturer clearly states "do not weld or drill on frame FLANGES". The flange being the top and bottom of the C channel. In the United States, the Federal DOT ONLY approves of the Z cut but this guy is an expert on torsional twisting force in the tens of thousands of pounds of torque on semi truck frames, oh wait he isn't.
Good video.
I think the scarf style joint could actually be worse than a straight butt joint. Surly the forces don’t like the welds changing at 90 degrees…
I wonder if there has been some confusion with the welding joint of the chassis rail and the welding of fish plates, or doublers on to chassis rails?
As I tried to explain in the video the member will fail (if overloaded) by bending, and all that really matters is the strength of the top and bottom flanges. If you want to make it stronger add some material to the flanges, not to the web.
Don't think of lines of force following the line of the weld - it doesn't work like that. There will be a stress field which is in equilibrium with the external loads and this will flow through whatever material is there (either the base metal or the weld metal). It doesn't follow the weld.
@@defendermodsandtravels I didn’t mean the line of force would change due to the joint shape. I ment the forces flowing through a change in direction of the weld.
Fish plates or doublers are added for many reasons other than bending of the chassis in the vertical plane. You might be attaching link mounts or a cross member to the webs….
I extended my caravan chassis rails by 1 metre. The two rails were 50mm x 100mm x 4mm RHS. I used a butt joint weld and then welded angled fishplates either side and only welded the top and bottom edges of the plates, leaving the side faces ‘angled at 45 degrees’ free from welding. I hope this is OK because we used to extend truck chassis this way years ago.
Here in the Alberta oilfield many of the bigger heavy trucks used for moving drilling rigs had the chassis lengthen. Nowadays it is always a straight or 90 degree joint and has been for about 30 years now. But years ago I was always told by some of the older men I worked with that it had to be an angled joint.
Very interestng to hear that. The older guys will have had it fixed in their minds that an angled weld is better without being able to explain why (except perhaps with a lot of arm waving).
Conditions where you live are particularly harsh because of the low temperatures.
Thanks for the comment.
@@defendermodsandtravelsoh stop with the condescending bullshit. You neither live in Canada or the united states, so you are oblivious to the federal regulations on frame splices. How many comments do I have to reply to make you understand that you are not an expert in this field. Aside from the arm waving, please explain why federal regulations only approve the Z cut. Also, please explain why numerous states in the US WILL NOT PASS A MOTOR VEHICLE FOR ANNUAL SAFETY INSPECTION with a vertical frame weld repair...oh because that repair is suseptible to cracking.
@@TheRoadhammer379 I made it clear in the video that I know nothing about local regulations on chassis welding however I'd like to have a conversation with those who drafted them.
You are very quick to assume that I nothing about engineering in North America. I have an engineering degree from Washington University; I am a professional member of ASME and ASCE; the design codes I have always worked to are API, AISC and AWS; I have worked on projects in the USA, principally in Houston.
The comment I would make is that there's a disconnect between practice in the Civil, Structural and Energy industries on the one side and those involved in chassis repairs on the other.
This was an outstanding presentation with some very in-depth analysis. However, something you might consider is fatigue failure. Auto chassis are subjected to all manner of impact forces and vibration from rotating, and possibly out of balance, elements. Architectural constructions are not subjected to much of this except low frequency wind loads. However, a high-rise apartment building near a supersonic test sight would certainly expose the building to square wave events. A Fourier analysis of these waves will disclose that something in the design will likely go into resonance and possibly fail by the members exceeding their elastic limit. It might be instructive to analyze wrecked automobiles that had square cut modifications to try to determine if the square cut was a contributing factor. In addition, an in-dept look at Formula one racing cars and inquire about the protocols they employ for builds and the numerous modifications.
Clearly fatigue will most likely be the cause of any failure Dynamics associated with the suspension will be a factor (which is why I applied a simple DAF to the static load however the dynamics associated with the body and chassis will be minor.
The guys who propose those exotic joint profiles claim they are "way stronger" so I refuted that argument based on static strength.
@@defendermodsandtravels I loved your very in-depth analysis ...we don't see much of that on TH-cam from people who actually work on cars. I hope Elon considers fatigue issues for his interplanetary missions where vibration damping i is most likely totally absent due to the absence of air. I have visions of some of the proposed structures flying apart from resonant vibrations. I sleep well at night because I have a three wheel car !!!
Welds shrink as they cool leaving the area right alongside the weld in stress. That’s why weld repairs usually fail at the edge of the weld, it’s rare for decent welds to actually in the weld bead. The idea of the more complicated joints is with the longer welds there’s more area that has to crack before failure. As a maintenance welder with over 50 years’ experience I like to fishplate my weld jobs on frames to avoid any future failures. I use plates at least 3 times longer than the height of the frame rails and attach the fishplates with only horizontal welds. The mistake that I’ve seen several times is welders will run vertical welds at the ends of the fishplate, figuring it’s stronger but in actuality they are only causing 2 more vertical stressed areas in the frame. Vertical welds on frames should be avoided if possible, especially on highly stressed truck frames.
Angle cuts mean longer welds, increasing the opportunity for flaws in the weld is my line of thinking.
Practical application speaks louder than mathematical speculations. Many times over I've corrected their "ideal" , to deem it actually safe to use . Overbuild for safeties sake , less failures . If a person took all this time to do mathematical equations, they would be employed long in the competitive work force .
Good stuff that got me looking. In America the Trailer Body Builders organization, GM Upfitters, and some tech articles from the SAE pretty much say the same thing. My issue is his example for Mor's is waay to close to the spring mount. But most of us hopefully will take it for what it was intended
He is not suggesting that the splice should be at the spring mount. He is doing a hypothetical study to find the maximum stresses in the beam and the maximum hogging moment is at the spring mount. This is obvious from the video.
Thank you, Sir. Very instructional and logic video base on rhe application of mechanics of materials to the chasis consruction.
Might get an extended half for my d2 after looking at the technical bits of this 😂 not sure if i rewlly wanna blob it together just Infront of the shocks anyway but this backs it up
The international harvester service manual instructs how to repair the heat treated c channel frame of their 1960s -1970s pickups. They instruct you to:
1) use only an arc welder, not a gas welder as it reduces the amount of heat you put into the metal
(Although there are in fact ways to control that heat).
2) when replacing a damaged piece of frame cut the frame at an angle.
3) reinforce the repair with angle stock of a minimum specified thickness and a minimum specified length.
4) cut 1" holes in the angle stock and weld it around the perimeter of those holes to the frame.
If you do a single vertical weld you will create a vertical anealed section. A strip of soft metal in the heat affected zone. But there is a difference in a joint from the factory. The welds in the frame are done and THEN the frame is heat treated, and thus the welds are as strong as the frame metal. (on my IH there ARE no (or at least very few) welds, its all riveted)
I can however demonstrate that the argument that joint configuration in a heat treated frame does nothing for strength is wrong with a simple thought experiment. This isnt to say that i will prove the differences are enough to worry about, or that they are not enough. That is another matter. But i can establish that its not irrelevant under all configurations. The argument does not hold.
The thought experiment goes like this: Consider a leaf from a 4" wide automotive leaf spring. Its a piece lf heat treated steel. Its very diffficult to break.
Make a straight cut through the center of the spring perpendicular to the length. Then weld it back together.
Now take an identical spring, cut it in half parallel to its length. Then weld it back together.
Now take these two springs , support the ends and load them up in the middle. We dont need a computer similation to tell us which will fail first. The one cut perpendicular to the length will fail in the heat affected zone where the heat treatment of the spring is destroyed. There is a section in the center where, all across there is NO properly heat treated steel. The other spring is weakened somewhat for the same reasons, but it still has heat treated metal along its entire length. The second example still has metal that acts as a spring along its entire length while the first example has a section in the center where its just no longer a spring. Depending on how it cooled and what filler was used its going to be hard and brittile or soft and malleable, either breaking or folding.
But the second example is extreme and not at all representative of a frame repair. (Although perhaps not as bad as one might think as it could represent the longitudinal section of a z shaped cut) So consider a diagonal cut from one corner of the spring to its opposite corner. That does not suddenly make it just as weak as the one that was compromised in the middle. It still has the property that there is good heat treated metal along the entire length.
Now repeat the experiment by making a diagonal cut just forward of one corner, to just shy of the opposite corner.
And again, just forward of where you started the last cut to just forward of where you ended.
I posit that as you change the angle of the cut (and consequently also change the length) the properties of the cut and rewelded spring will smoothly transition from those of the one where the spring was cut and welded corner to corner to those of the spring that was cut perpendicular.
But what of all the math? And the official structural engineering guidelines? How is it that those very smart people say with such cerrainty that the weld does not depend on that angle? The answer is, they didnt quite say that. They said that it doesnt depend on the angle, but only the strength of the underlying material. There is an implicit assumption baked into this statement that the strength of the weld is the same as that of the underlying material. This is true if your welding mild steel structural beams with mild steel filler. This is absolutely NOT true when your welding heat treated steel. The strength of the steel in the heat affected zone is going to be quite different.
But wasnt i talking about a spring and isnt that quite different than a frame? Well, not really. Everything is really a spring until you load it to the point you make a permanent change to it, at which point, whether it be a leaf spring or a structural beam, it's no longer acts like a spring.
So if your vehicle has a mild steel frame, them I agree. The weld should be as strong as the underlying metal if you use the right filler and the right technique, regardless of the angle of the cut. But if heat treated steel, its a different ball game.
Yes this is the problem when angle cuts are preached but they don't know why so they repeat it anyway .
This is 100%. Great job. Ive always understood that welding is as strong as base metal (excepting i think weld is harder?) But essentially i think youve got it, this is carpentry joinery making its way into welding. At least in this case it does no harm.
These fancy joints are no weaker but they are a lot more work. Fine for enthusiastic amateurs welding chasses but never used in industry.
Inclined and Z joints are appropriate for wooden glue-ups where the glue line length needs to be long enough and/or a mechanical "trick" like a scarf joint needs to be produced. Metal is different. The assumption is that the welded joint approximates any other part of the metal in terms of strength.
the only reason i would not square cut is when i have a cross member in the way in 2 places, where i have to go in between them, or to miss a gusset for structural support.
the other reason might be to use a piece of metal that would be too short if you have to cut it :) but then.... if in doubt reinforce it hey> yip.. over engineerd how many times do we see it in DIY jobs..i bet 100 % of the time. thank you for a entertaining video.
People assume you need slanted or stepped joints in welded steel because you you need them when gluing other materials such as wood or plastic, where the glue may not be as strong as the base material.
I’ve seen 140’ of suspended 24” seem-less .375WT pipe that was welded in the 40s by butt welds dangle in the open ditch for 3 weeks without failure.
It was supposed to be supposed by skids but it was a river bed crossing.
Very interesting. I just built a car ramp from short pieces and welded them with an angled cut. My instinct was that it would be better to spread sheer and compression forces over a longer weld, than a shorter weld. I think in a perfect weld, well executed with no welding flaws, the metal of the joint is as strong as parent metal or even stronger, if the weld is proud of the parent metal. I'm wondering if, due to possible welding flaws, it would be better to spread the forces on the welds along a greater distance. I don't know, but could the axiom about mitered cuts be found in weld failure analysis and not in "ideal" mathematics?
If you get a weld flaw in an area of high stress it will start to fail there irrespective of whether the weld is straight cut or angled.
@@defendermodsandtravels
Correct.
Very interesting analysis. I do however think that unless you grossly overload the vehicle, chassis failure will be caused by fatigue and dynamic loading. Under such a failure mode the quality of the weld is vitally important. Manufacturers usually have their weld procedures and quality checks in place hence you only see vertical welds. A poor vertical weld together with fatigue failure can result in catastrophic failure. The longer weld may first give an indication of failure. A modified chassis weld will more than likely have many inclusions worsening the fatigue strength.
Something like 90% of failures in metal structures are due to fatigue which emanate in the zone of highest fluctuating stress usually initiating at a defect or stress raiser. For most bending members this will be in one of the flanges. Once the crack has started it'll just grow at an accelerating rate until fast fracture takes place. Angling the web weld at 45 deg will make no difference.
Okay, novice here, and I'd have to say the advantage to an angle weld is a longer weld length, which can compensate for margins of error in workmanship, i.e. sloppy cuts, sloppy welding, inadequate alignment. This theory seems to have been confirmed by other comments from experienced and informed persons.
A method to avoid the issues of poor welding during work on the chassis.
I read that Mercedes absolutely forbid drilling hole on their sprinter chassis to stop failures at those points.
Thanks for doing all the calculations.
The "z" cut or the long angle cuts are more for alignment from how it was explained to me. It's easier to line the parts up straight with the longer sides.
Very interesting. I just did this to the wifes truck and I cut the splice at an unknown angle. Part of my logic was that of increased area of weld and I added plates to the lower flange figuring that tensile load would likely be the biggest factor. Another factor was that the angle made the welding easier as it gave more room to work, although the box was off the cab was still in place and close to the splice location.
You show several examples in industry and I'm afraid I have to respond to that with "what a silly little engineer man". 😂 That's the bean counters not wanting to spend any labour time creating scrap.
John
Your logic is quite wrong. Adding more weld metal in areas of low stress achieves nothing, particularly when a well made weld is at least as strong as the parent metal. If however it makes you feel good that you "did the job right" that's OK.
You mention that I show a few examples from industry of straight cut butt welds. In 40 years I have never seen a joint which wasn't like that. There must be one somewhere but I didn't see it.
From the silly little engineer.
Please don't think I'm suggesting you're wrong, you clearly showed I wasn't right. For me it was mostly about working room and manufacturers holes. It is good to see the numbers though.
I have a few heavy commercial trucks that we run at 65000 kilos, double frames, square cuts and no welds, everything is riveted or bolted.
Our accountant has come up with some great ideas over the years on how to "do it better"😂
Cheers and thanks,
John
Neither an engineer a bean counter.
@@joerogi8401 OK thanks for the comment John.
you do understand the bean counters have to operate within safety parameters dictated by the 'silly engineers' in order to meet safety requirements? they can shave costs here and there, but if a bridge (for example) is to be approved for public use then it has to be safe. there are plenty of examples of what happens when engineers & designers are ignored.