Man, you've got some hateful people in your audience. Don't let the stupid closed minded remarks diminish the value you add here on youtube. Thanks for all you do to teach people about these amazing passions!
@@VEC7ORlt I really don't see it? there is no "over-explaining" in the context of learning. if you already know stuff, good for you, but thorough explainations are exactly the goal in an explainer video.
@@TacticalFluke09 maybe, maybe not, but it sure reminds me of a uni lecturer we had at one moment - tells same thing in 3 different ways, and not the most important stuff, and there is no remote control to fast forward, unlike YT.
I've been in the trades as a welder and fabricator for about 16 years now. I have worked in all sorts of sectors like oil and gas, construction, manufacturing, etc. I am working to leave the trades and go back to school as an engineer in mechatronics. A common complaint i run into regardless of where I'm working or what industry I'm working in, is about engineers not designing things to be manufactured, repaired or worked on. I think the problem is the lack of practical experience a lot of engineers have. It's great that 3d printing is becoming more and more mainstream because it will expose some of these problems that the people actually building the designs have to deal with on a regular basis.
Well, as you well know, machinists have a lot of practical experience and familiarity with achieving things like proper fit and are the first ones to be handling precision made parts (because they made them precisely). I'm not sure what familiarity mechanical engineers have with machining and how far they are educated in the trade, but I can say this: machining is tough and demanding and often it is the machinist that is the better engineer overall.
@@OceanusHeliosi cant speak to how well a machinist or welder would perform as an engineer because i have never been an engineer. I suspect that mindset is the same as an engineer not having any practical experience, except reversed. People seem to forget that engineers, and the tradesmen that execute the engineers design, are two sides of the same coin.
Practical experience installing stuff like the engineer in the video when he installed his under sink filter in an awkward location, with no regard for the proper bend radius / length between his isolation and prefilter, or working length boxed in between 90s after the prefilter to his main filter? @27:13 His isolation valve is shut; is there a leak?
@@OceanusHelios Disagree on a machinist being the better engineer. They have valuable knowledge that should be borrowed and consulted when it comes to making a part easy to manufacture and maintain, but theres more to engineering than that. Considerations like strength, failure modes, alternate possibilities, optimization and even getting to a point where you understand whats actually required from a part/ assembly have a lot more nuance behind it than it might seem at first. Unfortunately engineers cant do all of that perfectly, but thats why you're supposed to consult with others, like machinists, maintenance crews, end users, the lists goes on.
As a mechanical engineer, I can say some. Paradigms such as design for assembly, design for service or design for manufacturing are parts of mechanical engineering education (Design for X is the common name). However today’s industry limits the design space with cost, industrial design and safety. More and more companies want unmaintainable/irreparable systems so that the customer will come back. Even at such cases some very smart solutions may be possible but project managers do not want to spend time there. They allocate their limited resources on where the customer will interact with rather than the guys at the factory or service. This approach is killing the artistic part of mechanical design and I hope the new sustainability wave will force companies to manufacture long lasting, serviceable products. i.e. Problem is real but the root cause is not the inability of the engineers but the competition out there
Pvc pipe, Fender washers, Paper clips, Ball bearings, Skateboard bearings, Dollar store cutting boards, Finishing nails, Insulation supports, Kickboards, All of these work great with 3d printed parts to achieve precision, toughness, and ease of manufacture.
'Dollar store cutting boards,' The small plastic ones are great for just a general plastic board. I just used one to mount a camera over a door sill, so I wouldn't have to drill holes everywhere. Camera weight sits on the door sill, mounting screwed into the cutting board, and then the edge of the cutting board slid into the J channel and behind the door sill. Camera is secure at the correct angle and doesn't move, without having to be drilled into the wall.
I think another term that could be used is "Design for Manufacture." Essentially we must know the limitations of a manufacturing process, including 3D printing, which has many positives, but can often fall short if off-the-shelf components are not integrated into an assembly or project. As you have demonstrated, the simple addition of off-the-shelf hardware can drastically increase the capabilities of 3D Printed components and their integration with more "traditional" manufacturing methods. My main piece of advice as a fellow engineering major is to try to address an audience more consistently. You clearly have a great understanding and knowledge of your field, but if you are going to address an audience that may not have expertise in your field be careful with how complex the terminology that you use is. Sometimes surface level terminology is all that is needed to accurately describe a concept, and you seemed to swap from expert level terminology to very simple descriptions of a problem throughout the video. Great video and keep up the great work!
One of the most important things I learned early when I started getting into designing with my 3d printer was the important of stress tests. Know your overhang angles, know your tolerances between 2 printed parts and know your tolerances between a printed part and a manufactured part, know the exact spacing for heat set inserts, screws, etc, and the exact minimum distance you need for a wall to not be too thin. All of these things will make designing for the printer so much easier. I understand this is basic knowledge for anyone who's already in manufacturing, but I came from a software background and had to learn it all as a hobby. It seems so obvious now, but it's kind of a blind spot going into your designs at the start.
I disagree in one comment. It is good for anyone learning design engineering as a hobby or professionally to learn the proper language. As long as a brief educational explanation as to the meaning of engineering terminology is given at some point early in the video as was done here, I would not discourage the continuance of such language.
@@RexAnothership Yep I totally agree with you! I could just see it being a bit of whiplash for less technical viewers if not explaining is done and both simple and complex terms are being thrown around.
Using a toothed belt upside down as gear teeth is a good example of this. My favorite example of borrowing precision is a design for a turntable I saw that used a Harley valve and valve guide as a radial bearing, sitting on top of a ball bearing in a hole as the thrust bearing. I've also seen ball point pens used as bearings. The ball tips have absolutely insane tolerances and until a year ago there was only a single manufacturer of them.
@@VaughnRhinehart pretty much anything works better than 3d printed gears, unless you're making very large tooth low load systems out of tpu. Although I've never used the belt tooth trick, I just end up using a belt conventionally or a chain for force transfer.
I taught GD&T (Geometric Dimensioning and Tolerancing) in ITT (No longer around). It is odd that college engineering majors don't get into GD&T. This is so crucial to assemblies and the reason for GD&T. At minimum, engineers need to speak this language (in drawings with GD&T callouts) so machininst can properly fabricate the part.
Well machinists like me can still machine using the old standards, but there is a caveat. GD&T lets us have more wiggle room because GD&T calls out where we have that wiggle room and where we don't. It is a better indication to have GD&T callouts on the blueprints so that we know what is going to be critical and what is not going to be critical and can better understand what the design intent is.
A big part of my job at the moment is remediating issues that our industrial designers caused by not considering the manufacturing capabilities of our vendor, the order of operations, and not indicating critical dimensions. It's frustrating because a lot of this work should have been done at the start.
@@dittilio I couldn't agree more. ID and other engineering majors should have GD&T in at least one core class. In my comment, I totally forgot about ID, but that may be one of the most important fields for this subject.
We do here in the UK, as long as the course is IMECHE certified that is, others may also, but it's a requirement for IMECHE, which is why I specified that.
@@dittiliowhen I worked for the old Hewlett-Packard they assigned me, a manufacturing engineer, to sit in the engineering design area, and provide input during design reviews. It was called “concurrent engineering.” Yes. It did decrease time to market and reduced product costs.
As an amateur maker on a budget (I'm both poor and raised not to spend money on stuff that I don't absolutely need... it gets in the way more than anything honestly) I kind of learned that on my own. Any time you overthink something, check if there are ready made solutions. Nearly every time you can save yourself money, time and hassle this way. I'm slowly becoming a prop maker and my mind is SO primed to just keep thinking of solutions when I think up a project. 3D printing really made me get into it even more. Few things that absolutely ruin me: Overthinking the details and doing a stupidly simple mistake with the very basic part of the project... This happens constantly. And another one is screw ups... I hate myself every time I screw something up, especially when I can't easily fix it. Today I was cutting up a mirror to fit the irregular hexagonal frame for the first time and I screw up all of them (first one was slightly too large and the tiny piece to make it fit didn't snap well and the other one I somehow cut up to wrong dimensions... yeah) and getting more is a hassle because shipping will end up being more than the mirror.
Mistakes are inevitable, don’t be too hard on yourself. If you still need a mirror, I’d highly recommend checking local thrift stores/second hand stores or similar, usually you can find some great bargains in those kind of stores.
Tolerance is a function of design, precision is a function of manufacture. A good engineer designs for the largest possible tolerances that still allow manufacturing precision ti achieve a functiinal product, while still practical and cost-efficient. That's why an engineer with actual productiin experience is preferable to a pure theorist.
YES! It annoys me no end when I watch 3d printer channels misuse the word tolerance as some kind of function of the machine. Clearance is another word I would like them to add to their vocabulary.
@@carneeki Printers do have tolerance limitations but they mostly make repeatable mistakes which the model can be adjusted for to achieve decent accuracy. I agree that clearance is a needed addition to their vocabulary. Making a log book for your machine to achieve proper clearance for the design intent is a wise move. Also printing with exterior first with slower travel speeds and or smaller road/pass width makes a huge difference to tolerance and surface quality. Some slicers allow for different perimeter layer heights than infill. This is good for great surface quality and decent print speeds. I often use 0.4mm wide 0.07mm layer height perimeters with 0.7mm wide 0.2mm layer height inner walls and infill.
Refered here by a short and stayed for the full video. Well done nailing what is missing from the typical engineering coursework that profs expect you to discover in the lab portion (or when thrust into the real world). Assembly concepts also apply to electronics design! Diodes, digital latches and precision rectifiers and other techniques for example to get precise voltages.
Same here. I don’t know why it was so intriguing but stayed for the whole video. It’s interesting to supplement the imperfect additive process with classic more precise items. The spring steel is a good example. Borrowing the tolerance from those industries to improve a print makes a lot of sense.
This is a great follow-up to your short video on this topic. I feel like I am able to understand when a part needs to "borrow a tolerance", but determining which pre-existing part is most optimal is the hard part. I will definitely try to keep these examples in mind for when I design models in the future!
Not sure if you asked us to subscribe in the video, but I subscribed immediately after this video because this is the exact type of channel that I’m looking for. I love taking deep dives into engineering topics, especially the ones that don’t get talked about that often. Keep up the great work, I look forward to seeing what you make in the future!
I am going to enjoy mentally digesting all this in the coming weeks/months until I have that "AHA!" moment where it applies to what I'm doing. Fascinating video, thank you!
Your short video on Borrowing a Tolerance led me here. The full video did not disappoint, and it expanded the discussion into directions only hinted at in the short. Thanks! I am not a design or manufacturing professional but am 3D-print-curious, and this video is enlightening. Regarding your specific questions at the end of this video, category 5 was most interesting to me but others appeal as well.
You're not just borrowing a tolerance. That seems to be one case of a more generic concept. Borrowing a property. You might be borrowing tolerance, or borrowing smoothness, or borrowing plastic deformation, or hardness, or heat resistance, etc. Etc. Many other topics to explore here
Awesome video! 100% practical design information. You included useful examples of cheap things for improving tolerances when designing a part for 3d printing. Haters gonna hate no matter how good a video you make. As a degreed ME I can confirm we didn’t put these ideas together in the context of assemblies. This helps regular people and companies with crappy consumer 3d printers make higher quality functional parts on the cheap. Don’t mind the haters… They don’t know how to contribute knowledge to the world in an effort to make it better.
I tear down lots of thrown out stuff - like tear it way down - and that is a great way to learn fastening and fabrication techniques - as you are showing - my Twin brother had his daughtesr (goggles and headphones) - tear down anything that was throw out - vacuum cleaners - tools - etc - the brain has endless hard drive visually
Another thing that is possible is borrowing a clearance. This actually often makes precision fits work interestingly when 3d printing. I unintentionally made a gearbox that only worked when not completely screwed together. If screwed together it bound up. Turns out the play in the self-threading M2 socket head screws was allowing the gearbox to work.
Balls in bearings are not supposed to contact all the race, just two points. Otherwise there will be not rolling friction, but sliding, excess heat and wear. If you need more load bearing capacity, you need to use roller bearings.
This was a wonderful talk on the subject of the necessities and limits of design tolerance. Your words emphasize the importance of using tools and materials for their distinct advantages. Utilizing off-the-shelf parts in key areas is superior to single-component manufacturing, as many readily available components have acceptable tolerances imbued from the manufacturing method. I also appreciate the notes on how high tolerance can be achieved with redundant and feedback loops.
I admit the explanation needed to be a bit clearer, but the concepts are actually smart, i enjoyed the video and was shocked to notice its 32min after i finished it, didnt feel like 32 min, thanks for sharing your thoughts, and for the new term "borrowing a tolerance" i will borrow it from you 😅
You make this world a better place my friend! Sub'd! First of all, set up a patreon account and promote it in you videos. You reall need to get paid for being this awesome. Second, I love to learn things you can't learn in 6 years of grad/post grad in anyfield. It is usually when you're in the workforce the true learning begins. There are so many tips and tricks out there that need to be exposed in order for forward progess.
I'm starting my mechanical engineering study next year, and I am super interested in the tolerances I could get from my 3d printer! I love this video and I can't wait for some more! I'm going online and buying some ball bearings for some hinges! I love designing parts and have created my own electric motorcycle with one of my friends and all the power transition parts are 3d printed (excluding motor and chain and the base of a bicycle) our designs are far from accurate and having such inaccuracies in mind can help tremendously in our next design or project. Thank you so much and I'm looking forward to the next video!!
Great work! I have been back on a 3D printing kick and tried my first printed spring a week ago. It works great, but applying it to something that will hold significant weight, like holding a monitor is a joke. I really like your idea of stealing the strength and spring properties from the binder clips!
Great video, I saw your other videos thumbnails and am definitely interested. When I clicked on your video iPod this was a bigger youtube channel with many many views and subscribers. Great content you definitely deserve more. Definitely got my support, even inspires me to start a channel again about my Hobby (Mechatronics engineer here) Keep up the great work! 💪
Great video, thanks. Some good stuff in the comments too - most good videos (and, equally, most bad videos) often attract lots of valuable comments. As for the key subject, speaking as a programmer with decades of experience (and hacker, hobbyist engineer etc with interests and experience in electronics, metal work, woodwork, 3d printing/design etc etc) it really chimes with me. It's about the mindset rather than specific techniques. Few, if any, good or great programmers keep everything in their heads, exactly the same as lawyers and doctors etc. It's about knowing what you need to do, how - roughly - to do it, how you've tried, failed, and succeeded in the past in similar issues, and - crucially - how to find and understand more detailed information from others when/if you need it. Theory alone is useless. Application/self-learning without theory is limiting. Combine the two - even to a small extent - and, especially, adopt the type of mindset shown here and you're set up for an enjoyable hobby or enhanced career. When the shit hits the fan, no one wants theory alone if it fails at the first lack of practical experience, and vice-versa. Subscribed. Thanks for the content and the mental reminder.
I love the premise of this video, I feel like the topic discussed heavily relates to the general idea of reverse engineering! It is true that this is a concept and approach that many engineers use to learn and develop intuition and perspective into the world of engineering, but what you are doing is putting it to words in way that can incite passion and curiosity into newcomers to the wonderful world of manufacturing and design. :)
Generally helpful, but the last topic (examples) was best for me. Very keen to understand how 3d printing can be merged with generally available precision components to achieve accruate, strong and repeatable parts. The spring steel from bulldog clips is great example as well as the hdpe tubing.
I've been thinking about this a lot lately.. The algorithm brought me here for a reason.. Thank you for this video.. Still figuring out all the ins and outs of FreeCad,, not sure if paying for a good CAD is worth yet but I have too many ideas and not enough knowledge to bring them into existence.. Thank you for teaching me so many useful engineering tricks to think about while I build my first projects.. I'm almost 33 and I feel like I'm really far behind because I haven't built anything useful yet.. Learning,, Understanding,, before Building and Tinkering.. Maybe I should find some videos on the engineering process.. But I kind of want to figure it out myself with some 1st principals thinking but it's the more difficult process.. (probably)
I just found your channel and really like it. Keep up the good work. On this topic I would add bronze bushings to add precision and strength to a 3d print. A flanged bushing with a shoulder bolt for rotation of parts.
Solid work, a YT recommendation success story. Subscribed. I guess if I were to go out of my way to find things to improve, perhaps the edit could be different such that there's less jumping around and backtracking in topics or something. I also didn't come away with the clearest idea of what "borrowing a tolerance" is supposed to mean exactly, but you did say that it's kind of a nebulous concept so I think that's par for the course. Anyway, lots of food for thought, thanks.
loved this video, subscribed. what got me, personally, was 'will be open source'.. which is a great hook. and the other major thing is 'thinking outside the box' which are the best examples ever. if you show someone 10 examples of how to do things differently, they'll come up with at least 3 ways on their own in different situations. examples of functional variety open up peoples ability to think wider. replacing a belt on an engine? you need the same belt. but.. if you're out in the middle of nowhere, maybe a belt, pantyhose, duct tape, water hose, or something else might be tried (won't necessarily work, but it might).. I use things 'not for their intended purpose' all the time. and it's interesting to see how many people are stuck in the "why would that work?" thought process as opposed to, "well, I supposed grinding a nut shape into a metal ruler might just work in this case"... or ratcheting 2 pieces of hard wood together around a nut... the more people can think like this, the more their minds become like the inventors of old... bicycle mechanics made an airplane, not by thinking of how they don't have airplane parts or that planes might not work, but by thinking that, y'know, with enough gearing, light weight materials, or wings.... this might just do the trick".... and these are great people to be surrounded by. the can-do'ers. and the ones willing to try. so.. I see your approach as very inspirational, and also useful to myself, as there's never enough examples of a different way to do something, which might translate to a completely different solution years later in a completely different situation. kudos. :-) Mind, I'd rather not see you turn into the 'Look, a bottle cap and a drill = a new way to wing string!" sort of thing.. you're way better than that. like a skateboard compared to a motorcycle. looking forward to learning more from you. :-)
In your example of the hex chuck vs. 3 jaw chuck, the precision of the 3 jaw comes from being able to make afirmative contact with the tool being chucked, whereas the hex chuck is non adjustable. If one jaw in a 3 jaw chuck is displaced 1mm radially, the contact point with the tool also displaces that same 1mm, and as such the center line of that tool displaces that same 1mm. Hex chucks wobble because they favor rapid interchangability and utilize flank contact to drive a tool. To facilitate rapit interchangability, they need to be loose. Even if your hex chuck is perfectly centered, there will still be room for the tool to move within the chuck. 3 jaw chucks on the other hand are not favoring rapid changability as much as a hex chuck, and have adjustability built in to facilitate positive contact with the tool, hopefully eliminating wobble of the tool within its chucking contacts
I think people might be confused as you have several different related concepts that all fall under "borrowing" but that term doesn't completely explain any of them: (Stay til the end for a question) 1) On imprecise parts, creating a machined surface as an indexing point/datum for *relative* precision -- the dimensions of the cast/printed part don't matter, as long as they provide sufficient strength for the required precise *relative* geometry of the machining operations - e.g. concentricity. 2) Using the properties of precisely-machined spheres and cylinders, which contact at exactly three/two points, which if you assume a normal distribution of error, on average will cancel out each other's errors by 1/N standard deviations. *see question below* 3) Using compliant materials to bridge gaps in tolerances, such as gaskets, or the deformation/break-in of 3d-printed plastics against harder materials. (this one is to me the biggest stretch for using "borrowing") 4) Using the observed (if not necessarily specified) tolerances of mass-manufactured objects to supplement the tolerances of 3d printed parts. 5) Even more fundamental is to stop the easy thinking you can just 3d print everything, and start thinking in terms of assemblies to get the properties you need from multiple materials. Maybe a better term might be (to borrow from computing) "mixed precision" -- cast/printed + machined or mass-manufactured parts, or using combinations of materials of varying ductility. My question about 2) above is about the assumption of normal distributions of error--might there be some inherent biases to the error in the 3d printing process that need to be taken into account? Print orientation is the biggest one, but also layer overhang, internal stress relief, thermal factors, etc, that might bias the errors of the spherical contact points to bias the errors in some direction, rather than a normal distribution of errors? It'd be interesting to explore how use of the circular/spherical error averaging might need to be limited/adapted based on materials/manufacturing.
Loved the video! It really sparked some ideas and solutions for certain problems. My only complaint is how monotone your delivery is. Kind of ironic being an engineering topic where you do everything to help minimize or eliminate variance. In the case of your vocal delivery some, variance would be helpful 😉
Awesome stuff man! I only work with wood and concete (high tolerances) but have considered engineering tools for some time. This channel is really cool. Dont mind all the hate over the wording, i totally understand what you mean because we use things that expand and contract in heat, all of this needs to be accounted for in the beginning. And your ball bearing needing to be "broken in" concept made perfect sense. Oh and im a skater too... so thats a real thing wel alllll know about as skaters...Even with all steel bearings. Subbed for sure!
This is an excellent review, probably especially for folks just getting into the hobby aspects of mechanical engineering through 3D printing. There are some things where it's probably important to emphasize that "very precise" in some of these contexts - when actually quantified - may mean something very different. The example I'll use is your post-processing of the print with a handheld size-O twist drill; if you're chasing the difference between a metric 8mm and an ANSI Size O, a two-flute twist drill is going to be very disappointing to use, even in a lathe. But if you're designing in metric and living in the US, the size-O is a very nice alternative to trying to source an 8mm six flute reamer - especially if the size, roundness, and perpendicularity over the length of the hole are more critical than what you can get from FDM but less critical than those last fifty microns.
This reminded me of that hot debate of having absolutely precise jigs to do welding. They are trying to sell their gears to do precise cutting and welding by deliberately not talking about how to design things to be easier to put together or borrowing tolerance like you're doing here. People need to know more about this, it's just easier to get higher consistency and cost effectiveness this way than trying to perfect machining of every parts, that's just unnecessary cost.
This was great. Yes he gave a name to "macgyvering", but it is such a fundamental part of rigid manufacturing that it deserves discussion. The metalworking example showing the flattened contact point needed to create another flat contact for a joint was not something I considered as someone that has only worked with 3D printing. I incorrectly assumed metal parts just had better tolerances, but no instead a precision design will include a step to ensure precision. The sloped shape edges for the sake metal part release are another nice detail that makes sense but I hadnt considered. Mass production 3D printing has similar tricks, such minimizing bed contact area so an impact from the printer tool head can be used to release the part. This is a great topic, something the average printer user wouldn't consider, despite affordable 3D printer design being rooted in this concept (rigid 20x20 bars, rods, and other non-plastic parts being used harmoniously with plastic ones).
I never thought of ball bearings as a force mitigators. I'm a 79y/o Fusion & 3D print tinkerer. I really enjoyed the video. I liked and subbed up. I hope I can contribute something.
those machined zones on cast parts are often called indexes, and you will very frequently see index stripes on large axles, as they have no desire to precision machine every single part of the axle, but you still need SOME zone to calibrate your lathe against for further machining steps
Meaning: No matter of schooling you have to work in companies where you really gets your hands on . Or learn it yourself ! Experience Experiences = Knowledge
The golf cart I have at work has urethane bushings in the steering column. In the summer, turning the steering wheel causes a honking noise and the wheel just stays wherever you last pointed it. As soon as the cooler fall temperatures hit, it becomes quieter and steers easily. No amount of graphite, Moly or teflon changes this. Need to print an ABS replacement.
Stumbled across this channel randomly and am very very thankful for your work. This kind of knowledge seems greatly underrepresented on youtube, especially directed at "non-professionals". The intro premise about assemblies ist just on point. are there any sources, books or other channels, you could recommend on this matter? Most channels show i.e. a cycloidal actautor leg, often in great details, but the main focus seldomly aims toward an "assembly discussion". On the other hand you have the "making of" genre, which focuses on fusion360 or machining workflow. If lucky you can find a thing or two of those you have discussed here, spread over a two hour tutorial video
Just a tip, those ridges are called seams. It's where the printer performs a retraction to move to the next layer. The slicing program should have a feature to change the seam behaviour, either to align it so they are all aligned, or randomise it so none of them are aligned. There are some other options too, but I figured those might be the most pertinent to you. Great work on the video btw. I just recently finished my mechanical engineering degree and was surprised to hear your course didn't do much in the way of design for assembly. Design for manufacture and design for assembly are a requirement for a course to be IMECHE certified here in the UK. May I ask where you studied (purely out of curiosity)? :)
In the future if you're making these videos, you could probably run them off a script to run em faster. Watched the whole thing at 1.75x and it was comfortable. It'd drop the overall length of the video, and would hold better attention, and that would also help with the algo and the overall viewership you get. Cool video.
I’m curious how many viewers would agree. folks, feel free to weigh in. it’s definitely a lot of work to script and condense. i’ll get there over time.
Whenever I hear "I was never taught this" it seriously breaks my brain because I go to school and I feel like I am listening to psychobabble struggling with classes but I have been making and designing things for most of my life. I will meet Mechanical engineering students what they like to make (I am EE type) and almost all of them are fascinated when I tell them what I make and what know (A lot of it being semi-complex mechanical stuff, nothing crazy). I simply don't understand how people can REALLY want to be an engineer and never teach themselves anything. I find myself in countless labs finishing projects in 5 minutes when it takes others 6+ hours. But when it comes to classes I could really care less once I understand the fundamentals because I know that is what I need to make something which gets me in trouble since you get tested on the hyper niche and deep crap which for some reason people can care about. I find it quite hard to care about things that I know have almost 0 translation to what I want to do.
...and thats part if the reason why i chose a study+apprenticeship model. I hate useless theory but excel when asked to apply knowledge in projects. Our education system is mostly focused on amassing specialized niche knowledge at least when it comes to universities. The dual system is a really great thing for people that think differently. Do you have something like that where you are studying? Btw, im looking for some help with a few projects that include programming. Ok its rly early in the morning so i fear my answer sprouted larger than initially intended xD Have a good day! Greetings! O/
I absolutely hear what you're saying... but I entirely disagree. I'm not an engineer. But I have one niche that I'm personally involved in. Amd one niche where I've been let down by professionals. I'm an alcohol industry expert. I've been a professional brewer. I've been a professional vintner. I studied German to expand the textbooks and methodologies I have available to me. I've done more or less what you did. I taught myself based on my interests. I gained a HUGE baseline knowledge, and have gone extremely deep down a number of professional rabbit holes. My interests in my field HAPPENED to be wide. My dedication HAPPENED to be far above the norm. I became an expert in my field through the methods youre talking about. But MOST people in my field who follow theor interests only ever learn about the most popular 25% of styles and methods. And are unable to work creatively within that bottom 75% wothout making mistakes that were worked out over 100 years ago. In a world where brewers are only brewing american standard lagers or IPAs... thats not a problem. But when an employer wants to branch out, they need to know if somebody has experience working with certain ingredients, certain techniques... things outside that top 25% of styles/methodologies. And KNOWING their employee can do that... it puts more weight behind their staff. If we ever had a brewing degree that was meaningful to getting a job that pays equivalent to even a tradesman... it genuinely would need to have those niches. And for my second point? Gunsmithing. Most gunsmiths work exclusively on ARs, striker fired pistols, or parts swapping. The number of times I've called somewhere and asked somebody advertising gunsmithing skills to refinish and recrown a barrel... only to find out that theyre a SELF DECLARED gunsmith, and dont work on absolutely basic (but absolutely critical) things is astounding. An engineer might only need a skill once in their life, based on market demands... but if I'm hiring an engineer, I want to know that they KNOW how to figure it out in a timely manner. If they cant, why should I hire a qualified engineer over a kid with ADHD whose hyperfixation was running a mill? Skills matter. But if I cant see all of your skills quickly and accurately for a job that pays enough to afford a home... I need to see that your degree or certification lets you step a ways outside the norm for your work... so that I dont fond out in 9 months that I need to pick between giving you 9 months for a 1 month niche job... or hiring an engineer that hyperfixated dofferently than you to complete the job in 1 month for 10 times your pay.
You can't be a good engineer without teaching yourself things. Sadly, I'm not sure you can teach people to be curious engineers who regularly self teach themselves additional skills. I no longer encourage people to go into engineering unless they are already building things. "You can't learn to be an engineer. You recognize that you are one and decide to accept it and run with it"
What language do you program in? I am also looking for people to collaborate with my projects as well. I could provide assistance if it is within my power.
@@davidmalaweythis is the same reason why integer gear ratios are avoided. An irrational ratio will wear much slower and more evenly and they're typically referred to as being "hunting".
In post processing parts tolerances can be borrowed from hand tools and other odd objects. For instance at a big box store like Home depot you can pick up reasonably flat to high tolerances marble tiles. With sandpaper taped to it, surfaces of parts can be quickly flattened, and it can also be used to create a reference surface to push up against a square and then another flat perpendicular surface can be applied. In this way, if the intitial design gives maybe a +.001 overage over the final design, more precision parts can be achieved. Yes, a bit of extra work, but depending on the application it can be much quicker than repeatedly printing parts to achieve the most ideal dimension.
Interesting! I already used staples to face the metal claws of some ancient video connector I wanted to make replacement mating parts for, which works quite well for the test I made but I still have to finish the design.
Your mission, should you choose to accept it: please do the math on the eccentricity of a round object in a three jaw chuck when one jaw deviates from concentric. I would like to have a formula for how much to shim to get a defined eccentricity or (eliminate an unwanted eccentricity). It will be dependent on diameter of clamped material. deriving the formula is particularly difficult. You may assume an ideal chuck (self centering scroll chuck or tapered surface drill chuck which you've shown) and that only one jaw has a different radial thickness from the other two. also that there is a flat on each jaw which is broad enough to make tangential contact with the round material. many thanks to you or your noblest viewer
as amateur designer, these types of discussions are very helpful in 3d printing, I think going over common issues that come to assemblies on cad to 3d print would be something I would be interested in. Thank you for your time… subscribed and liked.
The way to achieve a high degree of accuracy for 3D printed holes or features is to print perimeters first. All other walls and fill do not affect the tolerance. For holes parallel to the layer you need minimum layer height. Some slicers allow for variable layer heights so that you decrease layer height on the layers with parallel holes and the increase layer height to speed up print time for regions that do not require high detail. This can be even more effective if you have a Prusa XL with multi tools in that you can print scaffolding plugs of dissimilar plastics that do not bond to each other with zero tolerance to give a highly accurate horizontal hole detail feature that requires no post processing other than popping out the plug.
I believe you are really on to a good topic here. There is much to be learned from this video and likely this is just the start of the topic. Also, I see you are using SolidWorks and likely either professional or premium because you spoke of simulation. I use it as well, what a wonderful resource for us!!! One item (possibly a strange one) is PEX crimp rings. I wanted to strengthen the boss on a part that kept splitting. So, we fashioned the boss just so the copper crimp ring would compress the PLA and strengthen it. I suppose a PEX Pinch Ring would be ok too but we have the crimping tool so I think that worked better. I have repaired wooden handles on wood chisels this way, by crimping a copper PES ring on the wood after turning it to a correct size. Good luck with this concept. Thanks for the video.
I think this video would benefit from compring this ball bearing hinge design with something else, then the benefits of this design will become more apparent. The main benefit of this design seems to be controling the axes of all the hinges, they all stay parallel with each other as they rotate. Vs using a hole and pin hinge design, if the 3D printer doesnt locate the axis of each hinge perfectly vertical, as the parts rotate they'll fall out of plane with one another. His point seems to be that its easier to control the location of ball bearings (while relying on their high precision to be the same shape no matter how they're orrianted), than it is to control the concentricity and perpe dicularity of a hole for a hinge (relying on a 3D printer that introduces more error as features become smaller demensionally).
"borrowing tolerance" could be better called as "engineering tolerances in 3D printing" your term is catchy and short but all in all it also seems to confuse some people (excluding me)
I feel the 3 D printing has not matured enough as other Manufacturing process, and it requires a lot of Analysis and prototyping to achieve precision goals required in manufacturing industry. This is a good introduction which make the people to think what more can go to 3 D printing process.
I overheard a boo boo you made at the start of the video. You need to look into some policies regarding representing yourself as a toyota employee as validation of your skillset before making public videos. You could be construed as speaking for or on behalf of the company in a PR capacity. Also the intro was very condescending wether you meant it to be or not lol. I'm autistic so I understand how hard that can be and i'd personally appreciate the tip. Also, I am familiar with Toyota's policies surrounding this... Most large companies have the same types of policies and its why you rarely hear other YT engineers actually say where they work.
Do you think it will be a problem if you have detritus inside a part coming from wear on sliding contacts or should you design a way for those particles to fall out?
Man, you've got some hateful people in your audience. Don't let the stupid closed minded remarks diminish the value you add here on youtube.
Thanks for all you do to teach people about these amazing passions!
Content is not bad, but OP has really obtuse way of overexplaining things, maybe that why.
@@VEC7ORlt If English is not your native language, I can understand why you feel this way.
@@AngryMarkFPV nah, don't think thats the problem, some peeps literally think like that, they have a different 'order of things' in their head.
@@VEC7ORlt I really don't see it? there is no "over-explaining" in the context of learning. if you already know stuff, good for you, but thorough explainations are exactly the goal in an explainer video.
@@TacticalFluke09 maybe, maybe not, but it sure reminds me of a uni lecturer we had at one moment - tells same thing in 3 different ways, and not the most important stuff, and there is no remote control to fast forward, unlike YT.
I've been in the trades as a welder and fabricator for about 16 years now. I have worked in all sorts of sectors like oil and gas, construction, manufacturing, etc. I am working to leave the trades and go back to school as an engineer in mechatronics. A common complaint i run into regardless of where I'm working or what industry I'm working in, is about engineers not designing things to be manufactured, repaired or worked on. I think the problem is the lack of practical experience a lot of engineers have. It's great that 3d printing is becoming more and more mainstream because it will expose some of these problems that the people actually building the designs have to deal with on a regular basis.
Well, as you well know, machinists have a lot of practical experience and familiarity with achieving things like proper fit and are the first ones to be handling precision made parts (because they made them precisely). I'm not sure what familiarity mechanical engineers have with machining and how far they are educated in the trade, but I can say this: machining is tough and demanding and often it is the machinist that is the better engineer overall.
@@OceanusHeliosi cant speak to how well a machinist or welder would perform as an engineer because i have never been an engineer. I suspect that mindset is the same as an engineer not having any practical experience, except reversed. People seem to forget that engineers, and the tradesmen that execute the engineers design, are two sides of the same coin.
Practical experience installing stuff like the engineer in the video when he installed his under sink filter in an awkward location, with no regard for the proper bend radius / length between his isolation and prefilter, or working length boxed in between 90s after the prefilter to his main filter? @27:13 His isolation valve is shut; is there a leak?
@@OceanusHelios Disagree on a machinist being the better engineer. They have valuable knowledge that should be borrowed and consulted when it comes to making a part easy to manufacture and maintain, but theres more to engineering than that. Considerations like strength, failure modes, alternate possibilities, optimization and even getting to a point where you understand whats actually required from a part/ assembly have a lot more nuance behind it than it might seem at first. Unfortunately engineers cant do all of that perfectly, but thats why you're supposed to consult with others, like machinists, maintenance crews, end users, the lists goes on.
As a mechanical engineer, I can say some. Paradigms such as design for assembly, design for service or design for manufacturing are parts of mechanical engineering education (Design for X is the common name). However today’s industry limits the design space with cost, industrial design and safety. More and more companies want unmaintainable/irreparable systems so that the customer will come back. Even at such cases some very smart solutions may be possible but project managers do not want to spend time there. They allocate their limited resources on where the customer will interact with rather than the guys at the factory or service. This approach is killing the artistic part of mechanical design and I hope the new sustainability wave will force companies to manufacture long lasting, serviceable products.
i.e. Problem is real but the root cause is not the inability of the engineers but the competition out there
Pvc pipe,
Fender washers,
Paper clips,
Ball bearings,
Skateboard bearings,
Dollar store cutting boards,
Finishing nails,
Insulation supports,
Kickboards,
All of these work great with 3d printed parts to achieve precision, toughness, and ease of manufacture.
Now do you mean ball bearings, or ball bearings
And whilst we're here, fork handles or four candles?
'Dollar store cutting boards,' The small plastic ones are great for just a general plastic board. I just used one to mount a camera over a door sill, so I wouldn't have to drill holes everywhere. Camera weight sits on the door sill, mounting screwed into the cutting board, and then the edge of the cutting board slid into the J channel and behind the door sill. Camera is secure at the correct angle and doesn't move, without having to be drilled into the wall.
I think another term that could be used is "Design for Manufacture." Essentially we must know the limitations of a manufacturing process, including 3D printing, which has many positives, but can often fall short if off-the-shelf components are not integrated into an assembly or project. As you have demonstrated, the simple addition of off-the-shelf hardware can drastically increase the capabilities of 3D Printed components and their integration with more "traditional" manufacturing methods.
My main piece of advice as a fellow engineering major is to try to address an audience more consistently. You clearly have a great understanding and knowledge of your field, but if you are going to address an audience that may not have expertise in your field be careful with how complex the terminology that you use is. Sometimes surface level terminology is all that is needed to accurately describe a concept, and you seemed to swap from expert level terminology to very simple descriptions of a problem throughout the video.
Great video and keep up the great work!
One of the most important things I learned early when I started getting into designing with my 3d printer was the important of stress tests. Know your overhang angles, know your tolerances between 2 printed parts and know your tolerances between a printed part and a manufactured part, know the exact spacing for heat set inserts, screws, etc, and the exact minimum distance you need for a wall to not be too thin. All of these things will make designing for the printer so much easier.
I understand this is basic knowledge for anyone who's already in manufacturing, but I came from a software background and had to learn it all as a hobby. It seems so obvious now, but it's kind of a blind spot going into your designs at the start.
I disagree in one comment. It is good for anyone learning design engineering as a hobby or professionally to learn the proper language. As long as a brief educational explanation as to the meaning of engineering terminology is given at some point early in the video as was done here, I would not discourage the continuance of such language.
@@RexAnothership Yep I totally agree with you! I could just see it being a bit of whiplash for less technical viewers if not explaining is done and both simple and complex terms are being thrown around.
I didn't have any trouble with it at all, but then again, I am an actual machinist.
He’s presenting a useful mental model that will help when doing DFM but there’s way more to DFM
Using a toothed belt upside down as gear teeth is a good example of this.
My favorite example of borrowing precision is a design for a turntable I saw that used a Harley valve and valve guide as a radial bearing, sitting on top of a ball bearing in a hole as the thrust bearing.
I've also seen ball point pens used as bearings. The ball tips have absolutely insane tolerances and until a year ago there was only a single manufacturer of them.
A toothed belt is totally not gear teeth! It’s the wrong shape. I guess it could work, but does it really work better than 3D printed gears?
@@VaughnRhinehartgear teeth com in many shape and profile.
@@VaughnRhinehart pretty much anything works better than 3d printed gears, unless you're making very large tooth low load systems out of tpu. Although I've never used the belt tooth trick, I just end up using a belt conventionally or a chain for force transfer.
I taught GD&T (Geometric Dimensioning and Tolerancing) in ITT (No longer around). It is odd that college engineering majors don't get into GD&T. This is so crucial to assemblies and the reason for GD&T. At minimum, engineers need to speak this language (in drawings with GD&T callouts) so machininst can properly fabricate the part.
Well machinists like me can still machine using the old standards, but there is a caveat. GD&T lets us have more wiggle room because GD&T calls out where we have that wiggle room and where we don't. It is a better indication to have GD&T callouts on the blueprints so that we know what is going to be critical and what is not going to be critical and can better understand what the design intent is.
A big part of my job at the moment is remediating issues that our industrial designers caused by not considering the manufacturing capabilities of our vendor, the order of operations, and not indicating critical dimensions.
It's frustrating because a lot of this work should have been done at the start.
@@dittilio I couldn't agree more. ID and other engineering majors should have GD&T in at least one core class. In my comment, I totally forgot about ID, but that may be one of the most important fields for this subject.
We do here in the UK, as long as the course is IMECHE certified that is, others may also, but it's a requirement for IMECHE, which is why I specified that.
@@dittiliowhen I worked for the old Hewlett-Packard they assigned me, a manufacturing engineer, to sit in the engineering design area, and provide input during design reviews. It was called “concurrent engineering.” Yes. It did decrease time to market and reduced product costs.
As an amateur maker on a budget (I'm both poor and raised not to spend money on stuff that I don't absolutely need... it gets in the way more than anything honestly) I kind of learned that on my own. Any time you overthink something, check if there are ready made solutions. Nearly every time you can save yourself money, time and hassle this way. I'm slowly becoming a prop maker and my mind is SO primed to just keep thinking of solutions when I think up a project. 3D printing really made me get into it even more.
Few things that absolutely ruin me: Overthinking the details and doing a stupidly simple mistake with the very basic part of the project... This happens constantly. And another one is screw ups... I hate myself every time I screw something up, especially when I can't easily fix it. Today I was cutting up a mirror to fit the irregular hexagonal frame for the first time and I screw up all of them (first one was slightly too large and the tiny piece to make it fit didn't snap well and the other one I somehow cut up to wrong dimensions... yeah) and getting more is a hassle because shipping will end up being more than the mirror.
@ares395 Don't beat yourself up over mistakes, it's the cost of your education.
Mistakes are inevitable, don’t be too hard on yourself. If you still need a mirror, I’d highly recommend checking local thrift stores/second hand stores or similar, usually you can find some great bargains in those kind of stores.
Tolerance is a function of design, precision is a function of manufacture.
A good engineer designs for the largest possible tolerances that still allow manufacturing precision ti achieve a functiinal product, while still practical and cost-efficient.
That's why an engineer with actual productiin experience is preferable to a pure theorist.
YES! It annoys me no end when I watch 3d printer channels misuse the word tolerance as some kind of function of the machine.
Clearance is another word I would like them to add to their vocabulary.
@@carneeki Printers do have tolerance limitations but they mostly make repeatable mistakes which the model can be adjusted for to achieve decent accuracy. I agree that clearance is a needed addition to their vocabulary. Making a log book for your machine to achieve proper clearance for the design intent is a wise move. Also printing with exterior first with slower travel speeds and or smaller road/pass width makes a huge difference to tolerance and surface quality. Some slicers allow for different perimeter layer heights than infill. This is good for great surface quality and decent print speeds. I often use 0.4mm wide 0.07mm layer height perimeters with 0.7mm wide 0.2mm layer height inner walls and infill.
@@RexAnothership while true and a great idea to log good settings, this doesn't really have any bearing on the misuse of language.
Refered here by a short and stayed for the full video. Well done nailing what is missing from the typical engineering coursework that profs expect you to discover in the lab portion (or when thrust into the real world).
Assembly concepts also apply to electronics design! Diodes, digital latches and precision rectifiers and other techniques for example to get precise voltages.
Same here. I don’t know why it was so intriguing but stayed for the whole video. It’s interesting to supplement the imperfect additive process with classic more precise items. The spring steel is a good example. Borrowing the tolerance from those industries to improve a print makes a lot of sense.
This is a great follow-up to your short video on this topic. I feel like I am able to understand when a part needs to "borrow a tolerance", but determining which pre-existing part is most optimal is the hard part. I will definitely try to keep these examples in mind for when I design models in the future!
Yup, this is where creativity and paying attention to everything around you can really help.
Not sure if you asked us to subscribe in the video, but I subscribed immediately after this video because this is the exact type of channel that I’m looking for. I love taking deep dives into engineering topics, especially the ones that don’t get talked about that often. Keep up the great work, I look forward to seeing what you make in the future!
I am going to enjoy mentally digesting all this in the coming weeks/months until I have that "AHA!" moment where it applies to what I'm doing. Fascinating video, thank you!
you enrich my world today
a day you not learning is a day wasted!!!
going to watch you for more !!
I enjoyed seeing the integration of high tolerance parts into 3d prints sich as the 3dp hing and borrowed pin
Your short video on Borrowing a Tolerance led me here. The full video did not disappoint, and it expanded the discussion into directions only hinted at in the short. Thanks! I am not a design or manufacturing professional but am 3D-print-curious, and this video is enlightening. Regarding your specific questions at the end of this video, category 5 was most interesting to me but others appeal as well.
You're not just borrowing a tolerance. That seems to be one case of a more generic concept. Borrowing a property. You might be borrowing tolerance, or borrowing smoothness, or borrowing plastic deformation, or hardness, or heat resistance, etc. Etc. Many other topics to explore here
Awesome video! 100% practical design information. You included useful examples of cheap things for improving tolerances when designing a part for 3d printing. Haters gonna hate no matter how good a video you make. As a degreed ME I can confirm we didn’t put these ideas together in the context of assemblies. This helps regular people and companies with crappy consumer 3d printers make higher quality functional parts on the cheap. Don’t mind the haters… They don’t know how to contribute knowledge to the world in an effort to make it better.
The DIN-rail clip design is really great. A great idea to reuse those very common spring-clips for something else.
I tear down lots of thrown out stuff - like tear it way down - and that is a great way to learn fastening and fabrication techniques - as you are showing - my Twin brother had his daughtesr (goggles and headphones) - tear down anything that was throw out - vacuum cleaners - tools - etc - the brain has endless hard drive visually
Another thing that is possible is borrowing a clearance. This actually often makes precision fits work interestingly when 3d printing. I unintentionally made a gearbox that only worked when not completely screwed together. If screwed together it bound up. Turns out the play in the self-threading M2 socket head screws was allowing the gearbox to work.
Balls in bearings are not supposed to contact all the race, just two points. Otherwise there will be not rolling friction, but sliding, excess heat and wear. If you need more load bearing capacity, you need to use roller bearings.
This was a wonderful talk on the subject of the necessities and limits of design tolerance. Your words emphasize the importance of using tools and materials for their distinct advantages. Utilizing off-the-shelf parts in key areas is superior to single-component manufacturing, as many readily available components have acceptable tolerances imbued from the manufacturing method. I also appreciate the notes on how high tolerance can be achieved with redundant and feedback loops.
Finally found a TH-cam channel that I 100% match with
man, PLEASE make a video of your workshop organisation 👀 it looks like paradise !
i was barely paying attention until 6:25 and now I believe everything you say.
Excellent ideas.
When I have something I can contribute, I will.
Until then, this was fascinating.
Thank you.
I admit the explanation needed to be a bit clearer, but the concepts are actually smart, i enjoyed the video and was shocked to notice its 32min after i finished it, didnt feel like 32 min, thanks for sharing your thoughts, and for the new term "borrowing a tolerance" i will borrow it from you 😅
You make this world a better place my friend! Sub'd! First of all, set up a patreon account and promote it in you videos. You reall need to get paid for being this awesome. Second, I love to learn things you can't learn in 6 years of grad/post grad in anyfield. It is usually when you're in the workforce the true learning begins. There are so many tips and tricks out there that need to be exposed in order for forward progess.
Some really brilliant concepts here. Thank you very much for this.
I'm starting my mechanical engineering study next year, and I am super interested in the tolerances I could get from my 3d printer! I love this video and I can't wait for some more! I'm going online and buying some ball bearings for some hinges! I love designing parts and have created my own electric motorcycle with one of my friends and all the power transition parts are 3d printed (excluding motor and chain and the base of a bicycle) our designs are far from accurate and having such inaccuracies in mind can help tremendously in our next design or project. Thank you so much and I'm looking forward to the next video!!
Great work! I have been back on a 3D printing kick and tried my first printed spring a week ago. It works great, but applying it to something that will hold significant weight, like holding a monitor is a joke. I really like your idea of stealing the strength and spring properties from the binder clips!
All of it but I found the material/object examples most helpful. Like HDPE tubing for axles, BBs for precise and light balls etc
lol this is blowing my mind! Really great reminder that there are lots of ways to do even simple things.
Great video, I saw your other videos thumbnails and am definitely interested. When I clicked on your video iPod this was a bigger youtube channel with many many views and subscribers. Great content you definitely deserve more. Definitely got my support, even inspires me to start a channel again about my Hobby (Mechatronics engineer here)
Keep up the great work! 💪
Really enjoyed all of the different examples to illustrate the general point. Well done David!
&t=1186 - Neat, that's my design :) Good suggestion with the binder clip spring.
which one? small world we have now!
Great video, thanks. Some good stuff in the comments too - most good videos (and, equally, most bad videos) often attract lots of valuable comments.
As for the key subject, speaking as a programmer with decades of experience (and hacker, hobbyist engineer etc with interests and experience in electronics, metal work, woodwork, 3d printing/design etc etc) it really chimes with me.
It's about the mindset rather than specific techniques. Few, if any, good or great programmers keep everything in their heads, exactly the same as lawyers and doctors etc. It's about knowing what you need to do, how - roughly - to do it, how you've tried, failed, and succeeded in the past in similar issues, and - crucially - how to find and understand more detailed information from others when/if you need it.
Theory alone is useless. Application/self-learning without theory is limiting. Combine the two - even to a small extent - and, especially, adopt the type of mindset shown here and you're set up for an enjoyable hobby or enhanced career. When the shit hits the fan, no one wants theory alone if it fails at the first lack of practical experience, and vice-versa. Subscribed. Thanks for the content and the mental reminder.
I love the premise of this video, I feel like the topic discussed heavily relates to the general idea of reverse engineering!
It is true that this is a concept and approach that many engineers use to learn and develop intuition and perspective into the world of engineering, but what you are doing is putting it to words in way that can incite passion and curiosity into newcomers to the wonderful world of manufacturing and design.
:)
Generally helpful, but the last topic (examples) was best for me. Very keen to understand how 3d printing can be merged with generally available precision components to achieve accruate, strong and repeatable parts. The spring steel from bulldog clips is great example as well as the hdpe tubing.
I've been thinking about this a lot lately.. The algorithm brought me here for a reason.. Thank you for this video.. Still figuring out all the ins and outs of FreeCad,, not sure if paying for a good CAD is worth yet but I have too many ideas and not enough knowledge to bring them into existence.. Thank you for teaching me so many useful engineering tricks to think about while I build my first projects.. I'm almost 33 and I feel like I'm really far behind because I haven't built anything useful yet.. Learning,, Understanding,, before Building and Tinkering..
Maybe I should find some videos on the engineering process.. But I kind of want to figure it out myself with some 1st principals thinking but it's the more difficult process.. (probably)
I just found your channel and really like it. Keep up the good work. On this topic I would add bronze bushings to add precision and strength to a 3d print. A flanged bushing with a shoulder bolt for rotation of parts.
Solid work, a YT recommendation success story. Subscribed.
I guess if I were to go out of my way to find things to improve, perhaps the edit could be different such that there's less jumping around and backtracking in topics or something. I also didn't come away with the clearest idea of what "borrowing a tolerance" is supposed to mean exactly, but you did say that it's kind of a nebulous concept so I think that's par for the course.
Anyway, lots of food for thought, thanks.
Honestly ball bearings really are crazy. You'd think making a near-perfect sphere would be hard yet they're (literally in some cases) a dime a dozen
loved this video, subscribed. what got me, personally, was 'will be open source'.. which is a great hook. and the other major thing is 'thinking outside the box' which are the best examples ever. if you show someone 10 examples of how to do things differently, they'll come up with at least 3 ways on their own in different situations. examples of functional variety open up peoples ability to think wider. replacing a belt on an engine? you need the same belt. but.. if you're out in the middle of nowhere, maybe a belt, pantyhose, duct tape, water hose, or something else might be tried (won't necessarily work, but it might).. I use things 'not for their intended purpose' all the time. and it's interesting to see how many people are stuck in the "why would that work?" thought process as opposed to, "well, I supposed grinding a nut shape into a metal ruler might just work in this case"... or ratcheting 2 pieces of hard wood together around a nut...
the more people can think like this, the more their minds become like the inventors of old... bicycle mechanics made an airplane, not by thinking of how they don't have airplane parts or that planes might not work, but by thinking that, y'know, with enough gearing, light weight materials, or wings.... this might just do the trick"....
and these are great people to be surrounded by. the can-do'ers. and the ones willing to try.
so.. I see your approach as very inspirational, and also useful to myself, as there's never enough examples of a different way to do something, which might translate to a completely different solution years later in a completely different situation. kudos. :-)
Mind, I'd rather not see you turn into the 'Look, a bottle cap and a drill = a new way to wing string!" sort of thing.. you're way better than that. like a skateboard compared to a motorcycle.
looking forward to learning more from you. :-)
In your example of the hex chuck vs. 3 jaw chuck, the precision of the 3 jaw comes from being able to make afirmative contact with the tool being chucked, whereas the hex chuck is non adjustable. If one jaw in a 3 jaw chuck is displaced 1mm radially, the contact point with the tool also displaces that same 1mm, and as such the center line of that tool displaces that same 1mm. Hex chucks wobble because they favor rapid interchangability and utilize flank contact to drive a tool. To facilitate rapit interchangability, they need to be loose. Even if your hex chuck is perfectly centered, there will still be room for the tool to move within the chuck. 3 jaw chucks on the other hand are not favoring rapid changability as much as a hex chuck, and have adjustability built in to facilitate positive contact with the tool, hopefully eliminating wobble of the tool within its chucking contacts
I think people might be confused as you have several different related concepts that all fall under "borrowing" but that term doesn't completely explain any of them: (Stay til the end for a question)
1) On imprecise parts, creating a machined surface as an indexing point/datum for *relative* precision -- the dimensions of the cast/printed part don't matter, as long as they provide sufficient strength for the required precise *relative* geometry of the machining operations - e.g. concentricity.
2) Using the properties of precisely-machined spheres and cylinders, which contact at exactly three/two points, which if you assume a normal distribution of error, on average will cancel out each other's errors by 1/N standard deviations. *see question below*
3) Using compliant materials to bridge gaps in tolerances, such as gaskets, or the deformation/break-in of 3d-printed plastics against harder materials. (this one is to me the biggest stretch for using "borrowing")
4) Using the observed (if not necessarily specified) tolerances of mass-manufactured objects to supplement the tolerances of 3d printed parts.
5) Even more fundamental is to stop the easy thinking you can just 3d print everything, and start thinking in terms of assemblies to get the properties you need from multiple materials.
Maybe a better term might be (to borrow from computing) "mixed precision" -- cast/printed + machined or mass-manufactured parts, or using combinations of materials of varying ductility.
My question about 2) above is about the assumption of normal distributions of error--might there be some inherent biases to the error in the 3d printing process that need to be taken into account? Print orientation is the biggest one, but also layer overhang, internal stress relief, thermal factors, etc, that might bias the errors of the spherical contact points to bias the errors in some direction, rather than a normal distribution of errors? It'd be interesting to explore how use of the circular/spherical error averaging might need to be limited/adapted based on materials/manufacturing.
Loved the video! It really sparked some ideas and solutions for certain problems.
My only complaint is how monotone your delivery is. Kind of ironic being an engineering topic where you do everything to help minimize or eliminate variance. In the case of your vocal delivery some, variance would be helpful 😉
Awesome stuff man! I only work with wood and concete (high tolerances) but have considered engineering tools for some time. This channel is really cool. Dont mind all the hate over the wording, i totally understand what you mean because we use things that expand and contract in heat, all of this needs to be accounted for in the beginning. And your ball bearing needing to be "broken in" concept made perfect sense. Oh and im a skater too... so thats a real thing wel alllll know about as skaters...Even with all steel bearings. Subbed for sure!
This is an excellent review, probably especially for folks just getting into the hobby aspects of mechanical engineering through 3D printing.
There are some things where it's probably important to emphasize that "very precise" in some of these contexts - when actually quantified - may mean something very different. The example I'll use is your post-processing of the print with a handheld size-O twist drill; if you're chasing the difference between a metric 8mm and an ANSI Size O, a two-flute twist drill is going to be very disappointing to use, even in a lathe. But if you're designing in metric and living in the US, the size-O is a very nice alternative to trying to source an 8mm six flute reamer - especially if the size, roundness, and perpendicularity over the length of the hole are more critical than what you can get from FDM but less critical than those last fifty microns.
Subbed cause of the tolerance short. I don’t even 3d print and loved it
This is a great summary. Well presented
My God Man! Am I ever glad there’s a full video on this. I smoked a fat J and watched the Short on loop about 20 times in a trance!
Hah! I did the same thing, stone sober
Getting high makes you content to waste your life. The overseers didn't decriminalise it for your benefit.
@@MaximilianonMars Making assumptions about how others live their lives result and posting it on the internet is an even bigger waste of one's life...
I hope to be as clever as you someday. I’ll be borrowing this bit of wisdom until then!
by god, he done did it. my man's gone long form 🎉
This reminded me of that hot debate of having absolutely precise jigs to do welding.
They are trying to sell their gears to do precise cutting and welding by deliberately not talking about how to design things to be easier to put together or borrowing tolerance like you're doing here.
People need to know more about this, it's just easier to get higher consistency and cost effectiveness this way than trying to perfect machining of every parts, that's just unnecessary cost.
This was great. Yes he gave a name to "macgyvering", but it is such a fundamental part of rigid manufacturing that it deserves discussion. The metalworking example showing the flattened contact point needed to create another flat contact for a joint was not something I considered as someone that has only worked with 3D printing. I incorrectly assumed metal parts just had better tolerances, but no instead a precision design will include a step to ensure precision. The sloped shape edges for the sake metal part release are another nice detail that makes sense but I hadnt considered. Mass production 3D printing has similar tricks, such minimizing bed contact area so an impact from the printer tool head can be used to release the part. This is a great topic, something the average printer user wouldn't consider, despite affordable 3D printer design being rooted in this concept (rigid 20x20 bars, rods, and other non-plastic parts being used harmoniously with plastic ones).
I never thought of ball bearings as a force mitigators. I'm a 79y/o Fusion & 3D print tinkerer. I really enjoyed the video. I liked and subbed up. I hope I can contribute something.
those machined zones on cast parts are often called indexes, and you will very frequently see index stripes on large axles, as they have no desire to precision machine every single part of the axle, but you still need SOME zone to calibrate your lathe against for further machining steps
Meaning: No matter of schooling you have to work in companies where you really gets your hands on . Or learn it yourself ! Experience Experiences = Knowledge
The golf cart I have at work has urethane bushings in the steering column. In the summer, turning the steering wheel causes a honking noise and the wheel just stays wherever you last pointed it. As soon as the cooler fall temperatures hit, it becomes quieter and steers easily. No amount of graphite, Moly or teflon changes this.
Need to print an ABS replacement.
really nice video! looking forward to more.
Stumbled across this channel randomly and am very very thankful for your work.
This kind of knowledge seems greatly underrepresented on youtube, especially directed at "non-professionals".
The intro premise about assemblies ist just on point.
are there any sources, books or other channels, you could recommend on this matter?
Most channels show i.e. a cycloidal actautor leg, often in great details, but the main focus seldomly aims toward an "assembly discussion".
On the other hand you have the "making of" genre, which focuses on fusion360 or machining workflow. If lucky you can find a thing or two of those you have discussed here, spread over a two hour tutorial video
Just a tip, those ridges are called seams. It's where the printer performs a retraction to move to the next layer. The slicing program should have a feature to change the seam behaviour, either to align it so they are all aligned, or randomise it so none of them are aligned. There are some other options too, but I figured those might be the most pertinent to you.
Great work on the video btw. I just recently finished my mechanical engineering degree and was surprised to hear your course didn't do much in the way of design for assembly. Design for manufacture and design for assembly are a requirement for a course to be IMECHE certified here in the UK. May I ask where you studied (purely out of curiosity)? :)
My guess is University of Texas A&M based on the shirt he’s wearing
what a nice video, thank you so much
just realised you made the video on tapes as well. Great work man
In the future if you're making these videos, you could probably run them off a script to run em faster. Watched the whole thing at 1.75x and it was comfortable. It'd drop the overall length of the video, and would hold better attention, and that would also help with the algo and the overall viewership you get. Cool video.
I’m curious how many viewers would agree. folks, feel free to weigh in.
it’s definitely a lot of work to script and condense. i’ll get there over time.
@@davidmalawey for sure dude. I don't mean to knock but I'd hate to see your TH-cam not pop off like it would / should.
Whenever I hear "I was never taught this" it seriously breaks my brain because I go to school and I feel like I am listening to psychobabble struggling with classes but I have been making and designing things for most of my life. I will meet Mechanical engineering students what they like to make (I am EE type) and almost all of them are fascinated when I tell them what I make and what know (A lot of it being semi-complex mechanical stuff, nothing crazy). I simply don't understand how people can REALLY want to be an engineer and never teach themselves anything. I find myself in countless labs finishing projects in 5 minutes when it takes others 6+ hours. But when it comes to classes I could really care less once I understand the fundamentals because I know that is what I need to make something which gets me in trouble since you get tested on the hyper niche and deep crap which for some reason people can care about. I find it quite hard to care about things that I know have almost 0 translation to what I want to do.
...and thats part if the reason why i chose a study+apprenticeship model. I hate useless theory but excel when asked to apply knowledge in projects. Our education system is mostly focused on amassing specialized niche knowledge at least when it comes to universities. The dual system is a really great thing for people that think differently. Do you have something like that where you are studying? Btw, im looking for some help with a few projects that include programming. Ok its rly early in the morning so i fear my answer sprouted larger than initially intended xD
Have a good day! Greetings! O/
I absolutely hear what you're saying... but I entirely disagree.
I'm not an engineer. But I have one niche that I'm personally involved in. Amd one niche where I've been let down by professionals.
I'm an alcohol industry expert. I've been a professional brewer. I've been a professional vintner. I studied German to expand the textbooks and methodologies I have available to me. I've done more or less what you did. I taught myself based on my interests. I gained a HUGE baseline knowledge, and have gone extremely deep down a number of professional rabbit holes.
My interests in my field HAPPENED to be wide. My dedication HAPPENED to be far above the norm. I became an expert in my field through the methods youre talking about. But MOST people in my field who follow theor interests only ever learn about the most popular 25% of styles and methods. And are unable to work creatively within that bottom 75% wothout making mistakes that were worked out over 100 years ago. In a world where brewers are only brewing american standard lagers or IPAs... thats not a problem. But when an employer wants to branch out, they need to know if somebody has experience working with certain ingredients, certain techniques... things outside that top 25% of styles/methodologies. And KNOWING their employee can do that... it puts more weight behind their staff. If we ever had a brewing degree that was meaningful to getting a job that pays equivalent to even a tradesman... it genuinely would need to have those niches.
And for my second point? Gunsmithing. Most gunsmiths work exclusively on ARs, striker fired pistols, or parts swapping. The number of times I've called somewhere and asked somebody advertising gunsmithing skills to refinish and recrown a barrel... only to find out that theyre a SELF DECLARED gunsmith, and dont work on absolutely basic (but absolutely critical) things is astounding.
An engineer might only need a skill once in their life, based on market demands... but if I'm hiring an engineer, I want to know that they KNOW how to figure it out in a timely manner. If they cant, why should I hire a qualified engineer over a kid with ADHD whose hyperfixation was running a mill?
Skills matter. But if I cant see all of your skills quickly and accurately for a job that pays enough to afford a home... I need to see that your degree or certification lets you step a ways outside the norm for your work... so that I dont fond out in 9 months that I need to pick between giving you 9 months for a 1 month niche job... or hiring an engineer that hyperfixated dofferently than you to complete the job in 1 month for 10 times your pay.
You can't be a good engineer without teaching yourself things. Sadly, I'm not sure you can teach people to be curious engineers who regularly self teach themselves additional skills. I no longer encourage people to go into engineering unless they are already building things. "You can't learn to be an engineer. You recognize that you are one and decide to accept it and run with it"
What language do you program in? I am also looking for people to collaborate with my projects as well. I could provide assistance if it is within my power.
Awesome David!
In the ball-bearing example, the ratios of diameters would have to be irrational so as to prevent a periodic pattern from forming during break-in.
that’s insightful… i’ve not used these enough to make observations but i see what you mean.
@@davidmalaweythis is the same reason why integer gear ratios are avoided. An irrational ratio will wear much slower and more evenly and they're typically referred to as being "hunting".
well done, will definitely inform my designs in the future
In post processing parts tolerances can be borrowed from hand tools and other odd objects. For instance at a big box store like Home depot you can pick up reasonably flat to high tolerances marble tiles. With sandpaper taped to it, surfaces of parts can be quickly flattened, and it can also be used to create a reference surface to push up against a square and then another flat perpendicular surface can be applied. In this way, if the intitial design gives maybe a +.001 overage over the final design, more precision parts can be achieved.
Yes, a bit of extra work, but depending on the application it can be much quicker than repeatedly printing parts to achieve the most ideal dimension.
Interesting! I already used staples to face the metal claws of some ancient video connector I wanted to make replacement mating parts for, which works quite well for the test I made but I still have to finish the design.
incredibly insightful
Brilliant explanations, real engineering - more of this pls
Very good video, a lot of ideas .... Now I can see it is good to have a 3D printer.
Crazy useful info. Cheers!!
8:05 thanks, now I know why I should wear a centurion helmet when riding a motorcycle
Borrowing tolerance is an excellent technique for 3D printed design. Great video!
Thanks for sharing your knowledge!
Your mission, should you choose to accept it:
please do the math on the eccentricity of a round object in a three jaw chuck when one jaw deviates from concentric. I would like to have a formula for how much to shim to get a defined eccentricity or (eliminate an unwanted eccentricity). It will be dependent on diameter of clamped material. deriving the formula is particularly difficult.
You may assume an ideal chuck (self centering scroll chuck or tapered surface drill chuck which you've shown) and that only one jaw has a different radial thickness from the other two. also that there is a flat on each jaw which is broad enough to make tangential contact with the round material.
many thanks to you or your noblest viewer
Sand molds should be pre heated to achieve clean smoothness.
Still hard but if done properly they can produce very precise copies of an object.
excellent video, learnt a lot!
Wow. Amazing content
Very nice topic. Thanks for the video
dont like the "rebranding" for using off the shelf parts but some very good and practical ideas in this video
as amateur designer, these types of discussions are very helpful in 3d printing, I think going over common issues that come to assemblies on cad to 3d print would be something I would be interested in. Thank you for your time… subscribed and liked.
Excellent video
Schools and parents be like teaching 'you must achieve perfection', Engineering and real life; 'allow tolerance for anything to work'
The way to achieve a high degree of accuracy for 3D printed holes or features is to print perimeters first. All other walls and fill do not affect the tolerance. For holes parallel to the layer you need minimum layer height. Some slicers allow for variable layer heights so that you decrease layer height on the layers with parallel holes and the increase layer height to speed up print time for regions that do not require high detail. This can be even more effective if you have a Prusa XL with multi tools in that you can print scaffolding plugs of dissimilar plastics that do not bond to each other with zero tolerance to give a highly accurate horizontal hole detail feature that requires no post processing other than popping out the plug.
I believe you are really on to a good topic here. There is much to be learned from this video and likely this is just the start of the topic. Also, I see you are using SolidWorks and likely either professional or premium because you spoke of simulation. I use it as well, what a wonderful resource for us!!! One item (possibly a strange one) is PEX crimp rings. I wanted to strengthen the boss on a part that kept splitting. So, we fashioned the boss just so the copper crimp ring would compress the PLA and strengthen it. I suppose a PEX Pinch Ring would be ok too but we have the crimping tool so I think that worked better. I have repaired wooden handles on wood chisels this way, by crimping a copper PES ring on the wood after turning it to a correct size.
Good luck with this concept. Thanks for the video.
it would be cool to see a photo of how you used the pex ring you described. thanks for your inputs!
Very informative
Love your video, nicely explained
I think this video would benefit from compring this ball bearing hinge design with something else, then the benefits of this design will become more apparent.
The main benefit of this design seems to be controling the axes of all the hinges, they all stay parallel with each other as they rotate. Vs using a hole and pin hinge design, if the 3D printer doesnt locate the axis of each hinge perfectly vertical, as the parts rotate they'll fall out of plane with one another.
His point seems to be that its easier to control the location of ball bearings (while relying on their high precision to be the same shape no matter how they're orrianted), than it is to control the concentricity and perpe dicularity of a hole for a hinge (relying on a 3D printer that introduces more error as features become smaller demensionally).
Nice video from a fellow ME.
"borrowing tolerance" could be better called as "engineering tolerances in 3D printing"
your term is catchy and short but all in all it also seems to confuse some people (excluding me)
I feel the 3 D printing has not matured enough as other Manufacturing process, and it requires a lot of Analysis and prototyping to achieve precision goals required in manufacturing industry. This is a good introduction which make the people to think what more can go to 3 D printing process.
Where do you recommend buying your ball bearings of different materials?
I overheard a boo boo you made at the start of the video. You need to look into some policies regarding representing yourself as a toyota employee as validation of your skillset before making public videos. You could be construed as speaking for or on behalf of the company in a PR capacity. Also the intro was very condescending wether you meant it to be or not lol. I'm autistic so I understand how hard that can be and i'd personally appreciate the tip.
Also, I am familiar with Toyota's policies surrounding this... Most large companies have the same types of policies and its why you rarely hear other YT engineers actually say where they work.
Do you think it will be a problem if you have detritus inside a part coming from wear on sliding contacts or should you design a way for those particles to fall out?
Excellent
Are you really borrowing a tolerance when it's a global stack up of the 3d print xy tolerance plus a ball bearing tolerance?