This gentelman is 100% right on ALL points. In the early 90's I was a Fabricator in Aussy building V8 Super Cars. We spent hundreds If not thousands of hours on the dyno and track testing all sorts of header combinations. We had a 5 litre V8 limited to 7500rpm limit because of CAMS Rules. So to get every ounce of H.P out of such a small engine took a huge amount of development, even on the intake side also. We ended up having the engines so finely tuned an 1/8 inch change on just the intake length could cost us 45hp. All our headers ended up being 4-2-1 with an expansion chamber of the collector to an X pipe, all lengths where based on the heat (Sound pulse) signature of each tube all the way to the X pipe. If you listen to the engines (Before we had to really muffle the cars) of the higher funded teams (Holden, D.J.R etc) they ALL sound like high reving two stroke V8's, and we where making North of 600h.p reliably all day long. Awesome to see this dear man share his wisdom 👍🏻
@Jay Riley From what I've seen in my simulations it makes cylinder to cylinder balance a lot better. Exhaust blowdown pulses going the wrong way through other cylinders creates more problems with the inlet stroke (because it messes up scavenging) than having intake runners that interfere. It also lets you run a choke a lot smaller without restriction, protecting the low-mid range without top end cost. End result: Crossplane cranks with crossover headers make sexy Ferrari noises, make more mid-range torque, and can run with timing closer to ideal (because you don't get as much tendency for one cylinder to detonate from contamination). But it's difficult to get the length short enough for top end power, so on high RPM V8 racecars it's not always worth it overall
@@bmmarshall79 DJR cars. I was on the floor in the early 90's for the development of the E.A V8 Falcon and we where working along side Roush and his NASCAR boys! I know Perkins was doing the same if not more!!
My dad was a race engine builder and machinist (Pacific Northwest USA). His career spanned the 1960's through the 1990's. He did a lot of development and dyno testing (mostly V8's of every make). I know he would have been fascinated by your work. We had a customer that went to the trouble of building a 180 degree header system for a C3 Corvette running in the American SCCA Trans Am series in the 80's. The primaries came up and over the valve covers and back to vertical collectors that went down over the passenger side of the bell housing. They made the engine sound like a formula 1 engine. The pounding roar was completely gone! The power improvement was on par with what your work shows (2-4 percent). That's a lot when talking about the tiny margins we fight for in racing.
Yes' I only recently became aware of 180degree headers 'and how the sound of the engine is changed ' and I finally understood why a 289 in a mustang or fairlane etc ' doesn't sound like a 289 in a GT40.
@@blackfrost273industries4 We worked with the customer to design a quite elaborate insulated stainless steel enclosure with a lot of ventilation in the firewall and floor.
180* headers usually only work with a flat plane crankshaft. Latest tests prove that the bigger the diameter of the primary tube, the more torque the engine makes at ALL RPM & that an 18" collector is the ideal length. They also proved that running both sides into a Y pipe with a single outlet makes more power than duals or any other configuration.
Guy's - I am so busy doing new videos that getting to answer your questions is getting to be really difficult. Hopefully you will get most of the answers you are looking for in future videos. DV
Hi David, The information you provide in this video is phenomenal. THANKS for creating & posting it. I'm interested to know if headers on engines with flat-plane crankshafts need to be configured//built differently than headers for typical cross-plane crankshaft engines. Please do a follow-up video addressing headers for "flat plane" crankshaft engines. I believe the information would be very interesting & beneficial. Your consideration & attention are sincerely appreciated. Best regards, Ben
Yes great job clarifying what everyone tries to keep a dark art. You covered the typical v8 well. I know you can't do every request but it would be great to have a follow up video that fills in the gaps for other types of engines and for building a 4-2-1. For example primary length/diameter (4), how to choose the secondary (2) length & diameter, and how does this effect the collector/tailpipe (1) length/diameter for other engines as well as the typical v8, as well as any info on step tubes, merge and venturi design, and if they produce any benefits. Would also be cool l to here more on the 9' primary header and any other stories about design successes. Thanks for all the hard work you and others put in!
This is very interesting; the thing I’m finding is, the more I learn, the more I realize that I don’t know. Thanks for all you do David to give us your help and years of experience; greatly appreciated,.
I have followed your career all my life before the web, I sucked up everything you have shared via print and have several of your books next to me right now, Everybody thought Earnhardt could see air move, You were the Pioneer of Air Movement and traction Your my Hero and a Legend. Thanks, David for not keeping your wisdom to yourself and sharing your infinite knowledge, I have always appreciated it.
The key to getting it correct it must include the most important fact dual plane puts the pulses not even time period for each side. To get the max flow 4 into 2 then collector and the rest all counts. A flat plane v8 is different. The information he is sharing is spot on.
Really connected with your teachings. Understanding building my winning 2 stroke motors with proper pipe/pulse calculated really brought me to you. I feel like I now have a PhD after reviewing all of your videos. God bless you sir!
On my 2 stroke racing kart the distance from the piston to the end of the header tube is measured and you can put longer ones on for a different torque curve. Or shorter ones. But it deffinetly makes a difference. This is on my Yamaha KT 100 race kart. I change the h.p from 12 to 16 h.p with different length pipe. On race boats the make tunable headers that move with a lever. Cool stuff. Thank you for sharing your wisdom.
Sure, after watching the videos on this 99% headers are wrong, I have to agree with you whole heartedly, I have over 12 years of research and development technology from chemical plants. Understanding what you're talking about is so easy to digest. It is all based on what we used to call a batch process with multiple reactors. And measuring what comes in what's going on through the reaction phase and what is being exitive needs to be measured understood. And then if you're tying all these reactors together, you have to have the ingress. And egress shut up in an opportune manner to where they work in unison to the best yield that you can get, keep the videos up, and I love your work, and I think everybody needs to listen to you and understand your formulas and your pathway to this. But you're right on the money, buddy. Navigator
I got a friend with an h22 engine honda. No turbo. Did a naphin drawing of a tuning tube to install at the collector flange. He installed it while bolted to a mustang chasis dyno. Picked up 10 hp at 6500rpm 265 to 275hp . Changed the sound from a raspy honda with a lot of reversion slap. After it sounded like a healthy big block chevy. The tuning tube used the inertia of the exhaust pulses to stop the reversion. I think it also increased scavenging in the primary.
@@hayden6056 figure out what size your collector flange is. Buy an adapter that takes it from like 3" to 2 1/4. Weld a foot of 2 1/4 tube to it. Slip it in. At the collector. Be sure to have an air fuel ratio gauge to verify which way you need to go with idle transition feed restrictions at 1700 rpm. After that. Primary main jets at 3800 rpm. Then you can see what happens just below wide open throttle under load for the secondary mains. Why not wide open throttle as a third step keep the manifold vacuum above the power valve opening manifold vacuum. With the jetting all done. Then go wot with the manifold vacuum below the power valve rating. So you can see what you need to do with the power valve restrictions.
@@lilorbielilorbie2496 I've got four v8's here and 3 different kinds. My poncho sounds sort of similar to my Holden as they basically nicked the design off them but still different. Then nothing sounds like my LS motors due to the change in firing order and the aluminium heads. Also my two LS motors sound nothing alike because of exhaust and head and cam differences. I'm unsure where a 4 cylinder sounding like a big block Chevy went to talking to me about my jetting so maybe is cooked. Especially seeing as the only one that runs a carb isn't running anyway and probably won't be a carb next time around.
Back in the '50s, Motor Trend or Hot Rod Mag had an article on exhaust pipe tuning with a crayon. Draw a line port to end of pipe, eight individual pipes on V8, with a crayon. This is with the engine warm and the exhaust pipes cooled to ambient temperature. Fire up the engine and run it at the rpm you want peak HP, where the crayon melts is where the pipe needs cutting. I suppose several people would be needed to get this info on one run, or it could take several runs. It might be worth a try to find each cylinder's favorite length. Have not tried it, being a two stroke motorcycle racer, I went with other information. If you or others try it, let me know if it works. Good luck on your endeavors!
180 degree headers are the true equal headers. they put the pulses all 180 from the last cyl. sounds more like f1 also, a hum rather than a classic v8 rumble. dude made a set on youtube great video & informative
Mr. Vizard, absolutely the best explanation of the flow into & out of and why of the internal combustion engine. Using pressure is genius as I would bet someone made such sensors to withstand the rigors of this environment. But using them in this fashion makes perfect sense. Thank you! This is a keeper for my TH-cam save file. In your name sir. DK, ASE Master Tech since 78.
These kind of topics are great, brilliant as always. Just re assures me I'm going in the right direction as i converted a set of 4 into 1 headers made for my (4 cylinder, sub 7k rpm) engine into 421s. Interested to hear about anti reversion chambers as i dabbled with that too.
This is the real practical knowledge that comes from experience, not bench-racing or marketing speil .. Thank you sir for sharing your experience. I'd really like more details on the 2v vs 4v heads. Looking forward to more of your content.
Mr Vizard you are an engine builders hero and a saint for casting the precious pearls of your wisdom and experience before us oblivious swine. I shall endeavor to stand in the light of your experience so that I may always go faster than the other guy. My hat's off to you, sir.
Best of the best. Your accomplishments are mind numbing. I majored in auto mechanics in school... Thank God. Just love engines and working on mechanical things of all kinds. I can do my own repairs small or large. Garages are so expensive, I have saved thousands of dollars in my life. Always learning and blessed to find men like you. David you are amazing and inspire me to get even more horse power out of my lawn mower. Engine science is amazing. Thank you for sharing your hard work. Cheers to and your health
Thanks DV! I have always thought equal length’s were BS! You have really put it into perspective for me with which one’s should be different lengths! I have always figured a big torque increase, but didn’t realize the amount of HP increase at high rpm!
It's not exactly BS. It depends on the type of header we are talking about. The top fuel type of header, where each tube is self scavenging, equal length is important. All cylinders are running at the same rpm and have the same resonance. The reason equal lengths are BS in the rest of the world of headers, is that scavenging due to other cylinders is more important than the self scavenging of each cylinder.
@@kart70 Correct there are many more things to take into consideration including merge collector design (which ARE equal length anyway) and overall length etc.
It has everything to do with flow! I’v been in the HVAC industry for 33 years and it’s the same in my industry. Everyone is focused on the efficiency of the equipment I focus on the duct work and air flow.
Awesome, awesome, awesome!! I just watched Engine Masters S1 E12, where they tested dyno-tested for the right sized header primary on a street-mannered Blueprint Engines 400ci SBC, with a cylinder head that flows ~220CFM at .650" lift on the exhaust valve. Guess what they found? The 1-3/4" headers made the best power, looking at the entire graph. Yes, the 1-7/8" headers won the HP war by 3hp or so but they produced a funky torque graph down low, compared to the 1-3/4" headers. This matches up perfectly to the chart presented in your video here, David. Well done!!
The GT40 with the crossover exhaust eliminated the 2 cylinders being close to each other. I believe it alternated the firing order from bank to bank and it spaced all the pulses equally. It also ran the firing order in a circle in each collector to create a vortex.
Fantastic David thank you very much for the information and sharing your knowledge.... the younger crowd would do well to take note of what these older guys have to say there as so few of them left that are willing to share....
David listening to you is like listening to my Physics instructor Dr Carlton. Car headers are a far cry from the Denco, Carpenter, and Vance & Hines headers I used on my motorcycles years ago. theyre just now using a pyramid in the collector where the 4 pipes merge, we were using them on our sidewinders going back to 1978. So you sir are 100% correct on this which is why I built the headers on my Superstock Chevy II and Monte Carlo SS. I couldnt "buy" what I wanted so I built them. And yes theyre a stepped header design and my collector uses a tricone for the sonic pulse.
These men know their stuff, keep it coming , gentle men, and thank you both, Andy, and David, proformence is their profession, a labor of love, of going, Fast, and First, the first time,,,,,and the last,
I realized I forgot something in my comment and I apologize. My name is Jeff Thompson. I'm 63 and live in Montana. Thank you for sharing your valuable wisdom. Jeff
Great stuff, Sir. @26:42, you list the length changes for cylinders as: Cylinders 1, 4,&6 + 4 inches Cylinders 2, 3&5 - 4 inches Cylinders 1 and 8 - nominal length What about cylinder 7? Thanks.
Spent 2 years modeling NASCAR motors for Hendricks Motorsports, and designing and fabricating headers. We went from 740 to 815 hp in that time, cica 2000. Cup cars were running 9700 rpm, we learned the primary lengths were way long by instrumenting engines on a dynamic and timing the negative pressure waves.
I speculate a factor that is very important to equal exhaust flow is not just the length but also the degrees of turn the exhaust header passage , if we add up all the degrees of turn that occurs in the bends ! you can have anywhere from 180 to over 360 degrees of turning the exhaust will transition through. obviously a near straight shot out of the heads exhaust port is ideal ! every bend creates restriction and turbulence,So in my theory when calculating the length we must take into account degrees of bend , a stream of exhaust gas goes through as it gets more delay through the bends than in a straight path , thus there needs to be a formula that converts bend to equivalent straight path , so we need sum of straight lengths plus our turn coefficient which is greater than 1 multiplied by the center tube lengths in the bends ! reverse thinking to take advantage of this theory , would be to use length and bends to time pulses into the collectors. I wonder if any tuners have performed experiments to study this concept to find the formula for the coefficient table for various header diameters .
Thanks David I've had this same argument with a many of other engineers for years and years, I've also proven that there is no difference between or sometimes a gain in the use of a crush bend radius over a mandrel bend. My own testing i used water flow instead of air as the mass flow. Didn't have alot of funds in those days and a water flow tester was easier and cheaper to make. Great video.
VERY interesting. From the beginning of time I was always told that a 90 degree engine is the optimum and equal length headers are a MUST. Thank you for clarifying.
Regarding the power gain with unequal length: I think it's stunning to see the effect in the 3000 to 3500 range. Over 5%, almost 7% at best! Imagine a "boost" of that magnitude on your daily driver 😲 My main worry is though: this arrangement will surely ruin the sweet rumble of the V8, as it evens out the pulses...
Thanks, David. The great thing about the Ford 2 liter is, it's a non-inerferance engine. Timing belt breaks, easy fix. Reliable, decent power, good on gas.
I (now) choose pipe off the head based on projected horsepower thanks to some insight by an ex-NASCAR exhaust builder. Assuming long radius bends and a good port-header transition, 1.5" off the head can handle up to around 500hp on a V8. But it needs to step up fairly smartly to not be a restriction. Works well in try-y designs as the primary-secondary join is often where you'd need to put a second step (third diameter) on a 4-1 so the headers just look like the only big difference is a stepped primary. There's a ton more to it than that though...
Step tube headers improve power over the entire RPM range. Tri-Y headers improve torque BUT NASCAR teams ran them(& still does on trucks & Xfinity cars for ground clearance. Nothing new. Mickey Thompson manufactured & sold Try-Ys 60 years ago & Thorley headers STILL sells Try-Ys for trucks, Thorleys don't make as much peak power as 4 into 1s, but they do improve low end torque for trucks pulling trailers. Engine Masters dynoed several different primary tube sizes & proved that bigger primaries make more power throughut the entire RPM range. My buddy runs what's considered huge primaries on his all motor 440ci small block Olds in his 86 Cutlas Super Comp car & it runs low 5s in the 1/8 & low 8s in the 1/4, 60ft times are 1.10. Car ran 7.97/168 with his 555BBC with 4,000ft air
@@bradgriffith4231 It's not just the pipe size where the low-mid gains come from. A lot of it is the collector size. You can't fix the low end of a 250hp v8 (late C3 Corvette for example) with a 1.75 primary by running a 2" choke without restricting it a little in the top end. The expansion of a single pipe into a 4-1 collector then back down to a small choke is too much turbulence. But if you build a new set of headers that start at 1.375 then steps to a 1.5, you can run the small choke without the same turbulence. The effect is even more pronounced with a 4-2-1, you could go 1.375 primary into 1.5 for the secondary, then use a 1.75 final choke. You can get away with that because the expansion of a single pipe into a 2-1 collector is a lot smaller than a 4-1 and there's less turbulence. Are those videos here on TH-cam? I'd like to watch, see if they tested collector changes as well
I've been building exhaust systems starting with lawn mower engine go-karts I'm racing things that others weren't just because since I was about five or six years old. I definitely see where you're coming from and you're the first person I can say thinks pretty much like I do. Working with 2-stroke knowledge keeps me from being put in the box building 4 stroke exhaust systems. People have always hated my designs but have never been able to surpass them. Most of my street cars were straight pipes , very quiet because of properly phased pulses until you got bad. The funniest I ever built was an autocross full-size 69 station wagon with a 390 engine side pipe. That was so two of us and our lady friends could do comfortably the all night rally with special stages. The horrified look of the Fiat Audi VW Datsun drivers was great. I can only imagine your life was every bit the thrill living in the European continent. Chevy small-block tri-y headers 1973 in high school was when I seriously started developing my shut up and listen ability. Have a crippled math teacher with a slant 6 Dodge start I think it was that would eat anything the kids thought they had that was hot. Young ones you need to learn from the Masters 😉
Finally someone confirmed to me what I have been thinking for a long time Equal length is only for even firing engines LS and cross plane american engines requires unequal length as you are saying here depending on the fact that cylinders of 1 bank do not fire 90 degrees exactly after each other Some more which require longer and some less which require shorter With the exception of the 8 to 1 exhaust LS with rotary fire pattern where the collector recieves pulses sequentially in order In this case equal length is the way to go
Very impressive information David. Currently looking at applying this to an Australian Ford Windsor 5.0 and a Cleveland 351 small port. Same firing order on both. Same objective on both. One is a work vehicle, hauling a load all of the time, the other a 4x4 and tow vehicle. Maximising torque at low rpm for drivability, and fuel efficiency is the objective. It would appear that Ford Australia took your advice to the extreme on the Windsor. The factory 4 into 1 system pipes, range from about 22" down to 4" LOL. The information that you have provided, along with your low noise exhaust video, plus the information in your How to Build Horsepower book, means I have some design directions to follow. Thank you very much. Regards Greg from Perth, Western Australia
Hello Greg from Perth! I have a 351C in a 3500 lbs Ford Mustang 71 and I drove it around with a built engine using Australian 2V heads but did not get very good fuel econ. I tried smaller carb but it ran hot that way which is no good. It had 10:1 compression dished Ross pistons and mild cam. At least it was fun. How is your economy so far on your vehicles? Have you thought of Injection? That and overdrive tranny would probably increase my fuel econ from 8 mpg to perhaps double that but I don't feel like switching fuel delivery. I'm more open to a T5 trans with overdrive. Got C6 now.
BTW, I love Big Block Chevies. I currently own a 1997 GMC Suburban with a 454 that I just built and upgraded last year. About 400 HP now. Not too crazy and plenty of power to get out of its own way.
My reading has led me to believe that the same people selling headers have a second job of selling camshafts. Header design often locks you into a compromise because it forces you to design for a narrow range. Most people pick too short a tube and too large a diameter thinking they will magically drive above 5K rpm everywhere they go. Engines are dynamic and change constantly.
I remember when Walker and Banks did those tests! Wow David, thats a great story! thanks for sharing, never knew that was you behind those awsome walker mufflers! Flowmasters are junk, not mufflers!!!! Always learning from your vids! I learn something everytime i watch any of your videos and I've been around for awhile as well...LOL
Wow!.....your video just popped up. ...I’ve subscribed instantly! I bought & read multiple times your 1st & 2nd (tuning the mini ‘A’ series books) ...40+ years ago! Your knowledge & expertise was awesome back then, so without hesitation I will watch & support everything you do... Regards Tony 👍
I was hoping you could address a few question's I had from this video: 1.) Would you apply the primary lengths changes to "balance-out" a 4-2-1 design / would it provide similar benefit? 2.) it seems a mistake was made in your presentation given cylinder #1 was listed 2X and cylinder #7 is missing? 3.) If mistype occurred, THEN wouldn't cylinders: 4, 6 & 7 = +4" longer , 2, 3, & 5 = -4" shorter, and 1 & 8 = nominal? Any feedback you're willing to provide is GREATLY appreciated; keep up the good work!!!
I'm trying to my best to follow this for my upcoming build, and I'm getting a little lost. Basically what we're trying to do here is ensure that each cylinder's pressure wave arrives at the collector in evenly spaced intervals. At 26:35 we see our big moment that gives us the exact measurements (although there is an error here, cylinder 1 is listed twice and cylinder 7 is not listed), and it was in trying to apply those logically that I realized that I was lost. With a classic firing order of 1,8,4,3,6,5,7,2, evenly spaced at 90* intervals of crankshaft rotation, this doesn't add up in my mind. Please see my diagram below of cylinders and the number of degrees of crankshaft rotation that have occurred since the previous fire of a cylinder on that same bank: 8 - 180 7 - 90 6 - 180 5 - 180 4 - 90 3 - 270 2 - 270 1 - 180 It seems to me that we should be making the cylinders that fire after only 90* should be the ones that are 4" longer, and that the ones that fire after 270* of rotation should be the ones that are 4" shorter, however this does not match up with the "big moment". I considered the fact that this test may have been done with a firing order swap of 4&7 and perhaps even 3&2, but still it's not making sense to me. Hopefully Mr. Vizard sees this, or someone else who can clear this up and help me understand.
@@johnsnow1355 can you please elaborate more? I have an in-line six cylinder engine, and I am designing an unequal headers, the firing order 1 - 5 - 3 - 6- 2- 4 it’s a Nissan tb 48, it’s going to be a true duel exhaust , so therefore two manifolds will be available, first manifold will have its primary tubes on 1 5 3 and the second one will be 6 2 4, I am doing this to get the scavenging. I want to avoid the pulses from hitting each other. What’s the ideal situation to get the most of it
@@johnsnow1355 “exhaust pulses have a known velocity so it’s a simple math calculation” Apparently it’s a simple math calculation that you don’t know or are unwilling to share. 🤦♂️
Absolutely love your videos, I'm addicted to learning more. My project old peace of family history is going kinda in the opposite direction in performance, but laws of physics still apply. I''m a diesel guy, drive truck (boss gets the money, I put in the hours, I'm on poor mans budget) My little Cummins pulls 16,500lb EMPTY camper and horse trailer --- 19,000+ lb loaded camper and horses most weekends up extremely steep off road, logging roads, hard climbs. High country of WA. Love the sound on my over worked straight pipe Cummins, whistling 32lb of boost, at a beautiful throaty sound 1,800 RPM, blowing out some quite hot EGTs. To say there ain't much carbon build up in my lil 5.9 Cummins. BUT!!!! I also dearly miss driving my 71' F250, 360. Open up the 4bbls on my Edelbrock 600. I know the first answer will be, go put a 428 or 460 in it and be done. I'm weird about keeping as close to factory original as possible. Just want to tune the most efficiency, torque in what I got, 360 FE. Factory original (I put on the manifold and carb, single glass pack, Edelbrock cam timing chain with +/- 4 degrees / stock timing keys). Everything else stock. I'll never pull 20,000lb, but it DESPERATELY needs bottom end torque help, to help pull 10,000lb horse trailer up a grueling 6-9% grade, slow / low RPM - 25mph pull up, off road mountain. Any short quick advice is greatly appreciated. My thoughts, please correct me where I'm wrong... Advanced cam timing "power curve down, raise compression 🤷", small tube long headers, long collector. I'm still learning... I'm on a poor mans budget. Can't have something that doesn't do a job, yet I miss the sound of a 4bbl V8 working. Any pointers are greatly appreciated.
Way back in the day, one of the US "hot Rod" type magazines tried 5 different design fenderwell "headers" for a 57 Chev with a street-strip 327- there was around 50hp difference between the best and worst. IIRC, these were 4>1 designs, so there was even more potential for a 'better'. Making it even more significant, that was only a 400-450 hp engine and if you project to some of the strongest modern NA engines, that could easily be pushing 100hp!
That why I love the old 255 Indy Ford snake exhaust engine of @1965... Buick used a 15deg upward angle on their exhaust manifold, this worked like a reversion dam top side of the exhaust outflow - that 15 degs helped make a vacuum helping the exhaust flow turn on the down flow of those factory cast manifold. Back in the 80' s there was Aftermarket custom exhaust headers known as AR --- Anti Reversion
When I started watching and getting the jist of where you were going with this video I immediately thought about pressure transducer sensor as a means to graph the pulse. It was neat to see you went the same direction. This is my first exposure to your TH-cam channel. Looking forward to watching more.
Cylinders 1357 are shown to show you the degrees of spacing between impulses of those cylinders! After cylinder number 7 cylinder number 1 is shown so you see the difference between crank quotation between cylinder number 7 and number 1! This is all related to one bank of a V-8 engine
Some race engine header designs took that issue into account. They had tubes from the opposite bank collected to give 180 deg separation of pulses and equal length tubes to provide maximum scavenging of exhaust gasses.
Thank you for sharing this David. It's a pleasant review for me, with some extras, as I own and have studied judiciously, many of your books. If you would, Is primary length critical for an inertia pulsed exhaust header? I'm referring to the tri-y 4 cylinder example you showed. If you'd be so kind, as to take it a step further, have you tried a tri-y header on a V8 and do they scavenge at all?
Ive just started watching your videos and have learned a lot. If i could make a recommendation, it would be to reduce the amount of animations and cuts in your videos. Specifically the random zooming of the camera. It is quite distracting to ME. Maybe not others but to me it is. Either way lots of good info and always appreciate spreading proper knowledge.
I dont know why you hand out so much of your hard earned knowledge for free on TH-cam and next to nothing in your books. However unrecognized, it's greatly appreciated
Excellent information, DV! I'm looking to do a set of custom headers for my Ford small block and have a question. On the chart at 26:35, it shows cylinder 1 as both +4" and nominal length. Which one is actually supposed to be cylinder 7? I learn a lot from your videos, as well as Unity's. Thank you.
I'm pretty sure you increase the length of the ones that fire back to back on same bank. It would be different from sbf to sbc because of firing order.
@@danielphillips1094 I've been thinking it out - pretty sure you lengthen the primary of the cylinder that fires 90° after a same-bank cylinder and shorten the one that's 270° behind another cylinder. I have no idea at the moment how much to add or cut off, but I'm pretty sure it'll depend on where you want your torque peak.
@@johngregory4801 that sounds right. I think nominal length would be determined by where you want peak power and the other cylinders would be adjusted to even out pulses at 180 degree intervals. I think what you're after is essentially a 180 degree header setup without crossing primaries from bank to bank.
@@johngregory4801 also in order to figure out how much to alter length you'd need to know how fast the pulses are moving. I believe I've seen the 500mph figure tossed around. I'm sure the info is out there, just have to know what to search for.
Thank you, DV; When you do your next video on this could you touch on pairing cylinders for scavenging in conjunction with optimizing unequal length primaries in addition to anti reversion, collector types, and secondary lengths? If that sounds more like two more videos, perhaps you could include the secondary length discussion with the scavenging pairing strategy with the appropriate un equal primary lengths. v/r wh
For performance street use, my favorite headers have been tri-Y design. I wish that was a more common offering. Street motors don't live at the top of their RPM all the time. And tri-Y is a not-terrible answer to the cross plane crank. My favorite thing about this channel is how well the info agrees with things I know from: being a general physics nerd, building speakers, sailing boats, and aviation. Also, it's always a "science based" approach here. The dyno is the arbiter of truth.
Dynos are not the arbiter of truth because they don't apply real world loads. I know numerous people with good running Super Comp cars that had them dyno tuned & they slowed down considerably. Went back to their old tuneup & the car picked right back up. Nobody other than some "exotic" spaghetti cars run cross plane cranks in street cars. Ford tried it for about 1 year in the Coyote & went back to a 90* crank. GM is using it in their race only Corvettes. The harmonic vibrations cancel any gain one might get in performance & the Ford cranks were destroying the engines. The longer the stroke, the worse the vibrations. It's the same as 4 cylinders without counter balance shafts, only twice as bad! As far as Tri-Y headers, they're ok for trucks pulling a heavy load but they don't make nearly the peak HP as a correctly designed 4 into 1 with merge collectors & step tubes with merge collectors are even better
I don't know this guys channel, just wandered here by accident and watched the whole 30 min. SO I will summarise his answer. YOU ARE GOOD WITH EQUAL LENGTH IF YOUR ENGINE FIRES AT EQUAL TIMING. Thats it, his title is a bit misleading. Most of the time some cylinder are not firing equall timing so you have to adapt those accordignly. Thanks for coming to my ted talk.
I’m way out of my league. Glad it’s a video. I can watch to overload, stop, and go again later. Learning LOADS. I always just accepted that four equal length into one was best. Hm. It leaves power on the table.
One wonders after decades of header manufacturing, did any header manufacturers ever know about this concept? Equal length just seemed too obvious. I guess it wasn't. Great video. I love when you guys and Tony get together.
I hate to defend header manufacturers but... this is probably farther down the priority list in the design for good reason. I would imagine that fitment and plug access are highest priority and honestly, would be happy if they could just get the plug access better on most. Trying to design a header that is the 'correct' unequal length means designing a header to one very particular build; IE with 'x' cam and 'y' manifold, optimized at 'z' engine speed. Maybe they could get something 'closer' with more R&D but it's extremely doubtful many would spend the time and money on anything but the highest performance/$$$ (highest profit margin) engine platforms.
Hi David, great video as always! I’m a physicist (so please don’t shut me off right away ;-), but I’m looking at your chart around time 21:18 . I notice that 1) you have a peak in performance enhancement at about 3500 rpm, and then 2) another right at 6000 rpm at the top rpm of the data. BUT, it is important to note that the performance enhancement is STILL INCREASING at 6000 rpm. This strongly suggests that if you went to higher rpm above 6000 rpm, that you will see continued performance enhancement with higher rpm. The reason is because you are approaching the second resonance of the header (2 * 3500 =7,000). But this suggests that this design is still leaving performance on the table. If the engine can’t go higher than 6000 rpm, then we have to bring that second resonance down to 6000 rpm by lengthening the header further. Does that make sense?
I remember when I was a kid in the 60s my dad had a friend who drove sprint cars. They were a little bit different back then they had clutches and could take off without a push they were also limited to 300 cubic inches. His car seemed to dominate the track. His car was different in many ways but what I remember was it had enormous headers and an exhaust that went to the back. The other cars had stingers as a kid I remember fire coming from the during night races but of course not from my dad's friend's car. As I got older I wondered how much those wild headers with all those tubes were just for intimidation purposes. I suspect that some of them didn't even carry exhaust. I remember he sure was a lot faster and those pipes sure were convoluted.
No you are effectively swapping the firing order left to right on the engine making the two hottest cylinders sit at the front instead of the back the pulses are the same
I know you said the numbers (primary lengths) in the video aren't optimum yet and need more research, but for what you've done so far it's a very worthwhile gain. My only question is... are they "close enough" to get similar gains on a small block Mopar? I know no header company is going to spend the time on an engine that's been out of production for 30 years, but I could always have a custom exhaust shop do a one-off set for me based on those numbers.
yes, its going to make a gain that's substantial, make note that his cup car engine (the drivers side header) was displayed in this video...is Mopar! The other thing to remember he clearly stated early in this video was that the secondary(collector) length was a much more important factor than most people gave consideration to.
You're right everybody has it wrong but you have it wrong too. Equal length does not mean the pipes are all the same pipe length but equal "Flow" length. Every bend changes the flow length, and the amount of bend changes the length. There are charts and formulas to determine pipe length. Every bend lengthens the pipe flow so you add or subtract pipe length varying on the number and degrees of bends so all pipes geometrically flow the same. My 1968 Motor Manual went into great details about this. This is stuff everybody overlooks and nothing new. This geometry is also used in several other industries Water, Air and Gas and others
Wait a minute...at 26:30 into the video you show lengths of each primary tube, but where is cylinder #7?...and you list cylinder #1 twice!...Please correct.
So the on screen formula half makes sense but #1 is listed as both nominal and +4". The logical setup would be inside vs outside pipes. This would be calculable by the fps of exhaust gas flow so the timing of the gas to the collector should be predictable. Only issue is how much spent gas remains, vs how much vacuum the collector can pull. Ie, how much gas must the exhaust gas push on in front of itself. Pulse timing is all mute if the flow gets there at the speed of sound vs the speed of gas flow and the speed of gas flow is only possible if the system is relatively in a vacuum which should be possible in a race situation so ill go with it. Basically the issuse is that on a 90 degree cross plane crank v8 the cylinder timings are 180-90-180-270... Well its circular and that might not be from 1 in the firing order but i listed it this way to make a couple things easier to see. so there are always 2 pairs at 180 (out of 720) but those 2 pairs aren't 180 from each other. There are also 3 270 pairs. if you skip over to the far side of the 90 degree gap with your pairs you find 2 extra these 2 extra are the ones youd nornally use on a tri y but i suppose inside vs outside could make more power on that because the cylinders can chase the tail of another, sometimes things that arent even are still good because it can help to get the energies interacting instead of isolated but 90 degrees is generally considered too close. When you lengthen the trailing cylinder of the 90 pair you put it less than 180 from the next cylinder, that may be better than nothing bot for a true 180 all around like most 4 cylinder engines you also move to cylinder 180 from the cylinder that you intend to move to fix the 90 situation. Since all common cross plane v8 cranks are designed roughly the same this amounts to an inside pair and an outside pair. So on any particular firing order the mathematically ideal situation would be that anything done to 1 is done to 7. Anything done to 3 is done to 5. Anything done to 2 is done to 8 and anything done to 4 is done to 6. While i do see some of that in davids results its not all around so i wonder if they tried the mathematical way and maybe say something they didn't like or if they started with the transducers and pushed things around till it was good enough? Anyhow as far as i can tell with any firing order you would have 4 long pipes that would make an even running 4 cylinder and 4 short pipes that would make a different evenly running 4 cylinder. The groupings would be 1,4,6,7 and 2,3,5,8 same as you would combine them into a 180 header but only to simulate a 180 header (at a specific rpm, though it should help at all rpm) using separate banks which should help avoid excess lengths to make a true 180 header. I actually thought because it works out to inside vs outside you could use the extra pipe as part of an asthetic design and make some squigles or loops with center exits like some block huggers. If you placed everything right you could have a heart like appearance. This could be marketed with the old heartbeat logo they used to use for chevy as the heartbeat of usa. I almost don't want to post this so i have an oprotunity to make prototypes but i just don't have the time for that. Please credit my comment here if you market that and people buy it.
Hello DV, thank you for the accuracy explanation, like always. Can you make one for Turbo headers and maybe also on a plenum size too? At the end fig. is it cylinder No. 7 and 8 nominal and is the Nominal size for the calculation as explained before? Thank you and God bless you.
All the Subaru engines from early 2000s (and probably back to the 80s) up to 2014 had unequal length headers, really bad but sounded great. 2 of the pulses were running into each other with another right behind and then the 4th dragging behind. They sounded cammed with open headers. Then the turbo models sounded even hotter, until they put "equal length manifolds" on them and now they just sound like a mild engine. Still has a unique sound being a boxer 4 cylinder but just sounds like its neutered now.
DV, I think that a Tesla Valve (the scientist, not the car company) conduit design in a header could assist in solving some of the problems with back-flow due to pulse asymmetry issues that result in collector flow conflicts. Combining a high flow primary of the correct length to smooth exhaust pulse frequency alignment with a backflow prevention mechanism that doesn't significantly restrict flow would free up another HP boost like that achieved with primary pipe length tuning. That would be a fun experiment to try.
Liked the video, and agree with your points purely from a torque / HP band perspective. But (and you allude to it yourself in the vid) all the distinctive throb and personality would be removed from the exhaust note. That is sometimes the main reason people buy V8s.
@27:34, in your list, you have cylinder 1 listed twice, at +4 and at nominal, and no #7 listed. What were the primary diameters? And lengths? I would think that the gasses changing direction would need to be considered also in optimizing headers. And 4-cylinder engines would be easier than eights due to crankshaft rod journal position. And manufactures will optimize headers based on who is going to buy them, optimized engine, heads, etc. Sorry for your loss.
The 30% of sound reminded me of the one of the reasons why MP3 files are relatively small. When the audio is compressed it drops the quieter sound waves that most ppl can’t hear under the loud ones. I kinda got hopeful when you mentioned 4cyl. That you might touch on 6cyl. I’m interested in slant 6’s those things need all the help they can get. 😂
but people like me MP3/compression sounds and all-8 firing i can hear 👂clearly it's the really high pitching wines i can't anymore or at least not as much as im around industrial equipment that changed my ears and or brain tune's it out so lower human-male voice is easier for me to understand ect
there's a bit more wizardry going into mp3 than dropping 'quieter' sounds, it's dropping data that doesn't "matter" as much based on tests on people and stuff like that.
I would like to know about lengths for Chevy LS engines as the firing order differs from the standard Chevrolet V8. LS firing order is 1-8-7-2-6-5-4-3 so 1/3 fire close together as do 2/6, so they'll see the pressure peak or overlap, whereas there will be a low between 7 & 5, and 4 & 8. SBC & BBC order is 1-8-4-3-6-5-7-2 so it's 8/4 and 5/7 firing close together so they have the overlapping pressures. and a corresponding low pressure with the 6/2 & 1/3 pairs. the chart you show at 27:00 has cylinder #1 twice, but no #7. then the #5 primary tube on the 572 BBC is significantly longer in the picture. the low rpm HP & torque numbers would make a big difference for trucks especially, but I'm also looking at the high rpm gains for a drag car.
The principle seemed to be shortening the "delayed" cylinders that fire 270/450 after 0, so with the LS pattern of 1x7xx5x3 the equivalent should be 5+3 shorter, 7 longer, 1 unchanged. x8x26x4x would be 8+2 shorter, 4 longer, 6 unchanged
@@Stereomoo I wonder if it matters with an open header at all (one side will not help scavenge the other)? on an LS, it seems to me 7/1 should be long with 3/5 shorter (say 34"/30") to drive the peak between the 7/5 closer together and the 1/3 peaks apart. On the even side 6/4 should be long and 2/8 should be short (same 34"/30") for the same reasons. or maybe the 4" difference is wrong for that case? basically using a primary tube length to shorten or lengthen the time it takes for the impulse to get to the collector. combining pipes with a Y or X-pipe is where the two sides being different may make more logical sense, if the pulses land on top of one another then you still get pressure spikes after the two sides combine. Optimally timing the pulses of one side to hit exactly between the pulses from the other side would make it work best for scavenging all cylinders. however, this difference should be tested with a set of y-pipes that start at equal length and add 1/2" increments to one side until a peak is found.
This gentelman is 100% right on ALL points.
In the early 90's I was a Fabricator in Aussy building V8 Super Cars.
We spent hundreds If not thousands of hours on the dyno and track testing all sorts of header combinations.
We had a 5 litre V8 limited to 7500rpm limit because of CAMS Rules.
So to get every ounce of H.P out of such a small engine took a huge amount of development, even on the intake side also.
We ended up having the engines so finely tuned an 1/8 inch change on just the intake length could cost us 45hp.
All our headers ended up being 4-2-1 with an expansion chamber of the collector to an X pipe, all lengths where based on the heat (Sound pulse) signature of each tube all the way to the X pipe.
If you listen to the engines (Before we had to really muffle the cars) of the higher funded teams (Holden, D.J.R etc) they ALL sound like high reving two stroke V8's, and we where making North of 600h.p reliably all day long.
Awesome to see this dear man share his wisdom 👍🏻
@Jay Riley From what I've seen in my simulations it makes cylinder to cylinder balance a lot better. Exhaust blowdown pulses going the wrong way through other cylinders creates more problems with the inlet stroke (because it messes up scavenging) than having intake runners that interfere. It also lets you run a choke a lot smaller without restriction, protecting the low-mid range without top end cost.
End result: Crossplane cranks with crossover headers make sexy Ferrari noises, make more mid-range torque, and can run with timing closer to ideal (because you don't get as much tendency for one cylinder to detonate from contamination). But it's difficult to get the length short enough for top end power, so on high RPM V8 racecars it's not always worth it overall
5 L V8 "small engine"
@@NikoKyunKyun They are in my world Lol.
Kiwi, we’re those Perkins engineering cars?
@@bmmarshall79 DJR cars. I was on the floor in the early 90's for the development of the E.A V8 Falcon and we where working along side Roush and his NASCAR boys!
I know Perkins was doing the same if not more!!
I learned a lot from David’s books. Very knowledgeable guy with tons of experience.
My dad was a race engine builder and machinist (Pacific Northwest USA). His career spanned the 1960's through the 1990's. He did a lot of development and dyno testing (mostly V8's of every make). I know he would have been fascinated by your work. We had a customer that went to the trouble of building a 180 degree header system for a C3 Corvette running in the American SCCA Trans Am series in the 80's. The primaries came up and over the valve covers and back to vertical collectors that went down over the passenger side of the bell housing. They made the engine sound like a formula 1 engine. The pounding roar was completely gone! The power improvement was on par with what your work shows (2-4 percent). That's a lot when talking about the tiny margins we fight for in racing.
Yes' I only recently became aware of 180degree headers 'and how the sound of the engine is changed ' and I finally understood why a 289 in a mustang or fairlane etc ' doesn't sound like a 289 in a GT40.
@mark Hatch, that car was built by Jim McCrocklin, mike Halloran and myself, we did a ton of testing with Dennis Fisher, lots of fun
Holy cow for heat batman. I wonder if that was a concern at all.
@@blackfrost273industries4 We worked with the customer to design a quite elaborate insulated stainless steel enclosure with a lot of ventilation in the firewall and floor.
180* headers usually only work with a flat plane crankshaft. Latest tests prove that the bigger the diameter of the primary tube, the more torque the engine makes at ALL RPM & that an 18" collector is the ideal length. They also proved that running both sides into a Y pipe with a single outlet makes more power than duals or any other configuration.
Guy's - I am so busy doing new videos that getting to answer your questions is getting to be really difficult. Hopefully you will get most of the answers you are looking for in future videos.
DV
Hi David, The information you provide in this video is phenomenal. THANKS for creating & posting it.
I'm interested to know if headers on engines with flat-plane crankshafts need to be configured//built differently than headers for typical cross-plane crankshaft engines.
Please do a follow-up video addressing headers for "flat plane" crankshaft engines. I believe the information would be very interesting & beneficial.
Your consideration & attention are sincerely appreciated.
Best regards,
Ben
Yes great job clarifying what everyone tries to keep a dark art. You covered the typical v8 well. I know you can't do every request but it would be great to have a follow up video that fills in the gaps for other types of engines and for building a 4-2-1. For example primary length/diameter (4), how to choose the secondary (2) length & diameter, and how does this effect the collector/tailpipe (1) length/diameter for other engines as well as the typical v8, as well as any info on step tubes, merge and venturi design, and if they produce any benefits. Would also be cool l to here more on the 9' primary header and any other stories about design successes. Thanks for all the hard work you and others put in!
This is very interesting; the thing I’m finding is, the more I learn, the more I realize that I don’t know.
Thanks for all you do David to give us your help and years of experience; greatly appreciated,.
That is the Kruger Dunning Effect.
That is the Kruger Dunning Effect.
I have followed your career all my life before the web, I sucked up everything you have shared via print and have several of your books next to me right now, Everybody thought Earnhardt could see air move, You were the Pioneer of Air Movement and traction Your my Hero and a Legend. Thanks, David for not keeping your wisdom to yourself and sharing your infinite knowledge, I have always appreciated it.
The key to getting it correct it must include the most important fact dual plane puts the pulses not even time period for each side. To get the max flow 4 into 2 then collector and the rest all counts. A flat plane v8 is different. The information he is sharing is spot on.
Really connected with your teachings. Understanding building my winning 2 stroke motors with proper pipe/pulse calculated really brought me to you. I feel like I now have a PhD after reviewing all of your videos.
God bless you sir!
On my 2 stroke racing kart the distance from the piston to the end of the header tube is measured and you can put longer ones on for a different torque curve. Or shorter ones. But it deffinetly makes a difference. This is on my Yamaha KT 100 race kart. I change the h.p from 12 to 16 h.p with different length pipe. On race boats the make tunable headers that move with a lever. Cool stuff. Thank you for sharing your wisdom.
Sure, after watching the videos on this 99% headers are wrong, I have to agree with you whole heartedly, I have over 12 years of research and development technology from chemical plants. Understanding what you're talking about is so easy to digest. It is all based on what we used to call a batch process with multiple reactors. And measuring what comes in what's going on through the reaction phase and what is being exitive needs to be measured understood. And then if you're tying all these reactors together, you have to have the ingress. And egress shut up in an opportune manner to where they work in unison to the best yield that you can get, keep the videos up, and I love your work, and I think everybody needs to listen to you and understand your formulas and your pathway to this. But you're right on the money, buddy. Navigator
I got a friend with an h22 engine honda. No turbo. Did a naphin drawing of a tuning tube to install at the collector flange. He installed it while bolted to a mustang chasis dyno. Picked up 10 hp at 6500rpm 265 to 275hp . Changed the sound from a raspy honda with a lot of reversion slap. After it sounded like a healthy big block chevy. The tuning tube used the inertia of the exhaust pulses to stop the reversion. I think it also increased scavenging in the primary.
A twin cam vtec inline 4 sounded like a healthy V8 with triple the cubes and a pushrod cam?
That's an amazing piece of tube.
@@hayden6056 figure out what size your collector flange is. Buy an adapter that takes it from like 3" to 2 1/4. Weld a foot of 2 1/4 tube to it. Slip it in. At the collector. Be sure to have an air fuel ratio gauge to verify which way you need to go with idle transition feed restrictions at 1700 rpm. After that. Primary main jets at 3800 rpm. Then you can see what happens just below wide open throttle under load for the secondary mains. Why not wide open throttle as a third step keep the manifold vacuum above the power valve opening manifold vacuum. With the jetting all done. Then go wot with the manifold vacuum below the power valve rating. So you can see what you need to do with the power valve restrictions.
@@hayden6056 Exactly .I'm calling B.S. on a 4 cyl. anything sounding like a BBC. Ain't gonna happen unless you are on the Really good drugs.
@@lilorbielilorbie2496 I've got four v8's here and 3 different kinds. My poncho sounds sort of similar to my Holden as they basically nicked the design off them but still different.
Then nothing sounds like my LS motors due to the change in firing order and the aluminium heads.
Also my two LS motors sound nothing alike because of exhaust and head and cam differences.
I'm unsure where a 4 cylinder sounding like a big block Chevy went to talking to me about my jetting so maybe is cooked. Especially seeing as the only one that runs a carb isn't running anyway and probably won't be a carb next time around.
How do you fuck so much the sound signature of a "flat plane" 4 cyl to make it sound like a crossplane
Back in the '50s, Motor Trend or Hot Rod Mag had an article on exhaust pipe tuning with a crayon. Draw a line port to end of pipe, eight individual pipes on V8, with a crayon. This is with the engine warm and the exhaust pipes cooled to ambient temperature. Fire up the engine and run it at the rpm you want peak HP, where the crayon melts is where the pipe needs cutting. I suppose several people would be needed to get this info on one run, or it could take several runs. It might be worth a try to find each cylinder's favorite length. Have not tried it, being a two stroke motorcycle racer, I went with other information.
If you or others try it, let me know if it works.
Good luck on your endeavors!
I thought that was for where to put the x or h pipe for a v engine
That has been debunked several times. Same with x pipe placement, same with header extensions.
thats for 2 strokes.
180 degree headers are the true equal headers. they put the pulses all 180 from the last cyl. sounds more like f1 also, a hum rather than a classic v8 rumble. dude made a set on youtube great video & informative
The wisdom of senior mechanic and engineer. Very interesting stuff. I just subscribed.
Mr. Vizard, absolutely the best explanation of the flow into & out of and why of the internal combustion engine. Using pressure is genius as I would bet someone made such sensors to withstand the rigors of this environment. But using them in this fashion makes perfect sense. Thank you! This is a keeper for my TH-cam save file. In your name sir.
DK, ASE Master Tech since 78.
Thank you for your years of research and passion David. 35 years old and glad I found your videos!
These kind of topics are great, brilliant as always. Just re assures me I'm going in the right direction as i converted a set of 4 into 1 headers made for my (4 cylinder, sub 7k rpm) engine into 421s.
Interested to hear about anti reversion chambers as i dabbled with that too.
This is the real practical knowledge that comes from experience, not bench-racing or marketing speil .. Thank you sir for sharing your experience. I'd really like more details on the 2v vs 4v heads. Looking forward to more of your content.
Mr Vizard you are an engine builders hero and a saint for casting the precious pearls of your wisdom and experience before us oblivious swine. I shall endeavor to stand in the light of your experience so that I may always go faster than the other guy. My hat's off to you, sir.
Thanks David for your time to explain all that we should know about.
Best of the best. Your accomplishments are mind numbing. I majored in auto mechanics in school... Thank God. Just love engines and working on mechanical things of all kinds. I can do my own repairs small or large. Garages are so expensive, I have saved thousands of dollars in my life. Always learning and blessed to find men like you. David you are amazing and inspire me to get even more horse power out of my lawn mower. Engine science is amazing.
Thank you for sharing your hard work.
Cheers to and your health
Thanks DV! I have always thought equal length’s were BS! You have really put it into perspective for me with which one’s should be different lengths! I have always figured a big torque increase, but didn’t realize the amount of HP increase at high rpm!
It's not exactly BS. It depends on the type of header we are talking about. The top fuel type of header, where each tube is self scavenging, equal length is important. All cylinders are running at the same rpm and have the same resonance. The reason equal lengths are BS in the rest of the world of headers, is that scavenging due to other cylinders is more important than the self scavenging of each cylinder.
@@kart70 Correct there are many more things to take into consideration including merge collector design (which ARE equal length anyway) and overall length etc.
Can't wait for part II. Great job as usual.
Much thanks.
I literally just punched the air after hearing the intro. This is what I'm talking about. Thank you for sharing!
It has everything to do with flow! I’v been in the HVAC industry for 33 years and it’s the same in my industry. Everyone is focused on the efficiency of the equipment I focus on the duct work and air flow.
Awesome, awesome, awesome!! I just watched Engine Masters S1 E12, where they tested dyno-tested for the right sized header primary on a street-mannered Blueprint Engines 400ci SBC, with a cylinder head that flows ~220CFM at .650" lift on the exhaust valve. Guess what they found? The 1-3/4" headers made the best power, looking at the entire graph. Yes, the 1-7/8" headers won the HP war by 3hp or so but they produced a funky torque graph down low, compared to the 1-3/4" headers. This matches up perfectly to the chart presented in your video here, David. Well done!!
As usual another excellent education my DV.
oh this channel is such a gem
glad i found this
I have tri-ys on my 64 C10. They fit, were cheap used, they're silver coated and they look cool.
Great comments on Total Seal. Keith Jones, and Kevin Studeker have been critical assets in all of my racing engine builds. They are hands on experts.
This man was great friends with my old tech school teachers. Glad I got to meet him
I have never seen more adept adept header discussion. Best on TH-cam
The GT40 with the crossover exhaust eliminated the 2 cylinders being close to each other. I believe it alternated the firing order from bank to bank and it spaced all the pulses equally. It also ran the firing order in a circle in each collector to create a vortex.
Fantastic David thank you very much for the information and sharing your knowledge.... the younger crowd would do well to take note of what these older guys have to say there as so few of them left that are willing to share....
Thanks DV! Just finished this chapter in the big block book last night so I did the homework! Starting your small block book tonight
David listening to you is like listening to my Physics instructor Dr Carlton. Car headers are a far cry from the Denco, Carpenter, and Vance & Hines headers I used on my motorcycles years ago. theyre just now using a pyramid in the collector where the 4 pipes merge, we were using them on our sidewinders going back to 1978. So you sir are 100% correct on this which is why I built the headers on my Superstock Chevy II and Monte Carlo SS. I couldnt "buy" what I wanted so I built them. And yes theyre a stepped header design and my collector uses a tricone for the sonic pulse.
I found out about your videos when my ex was in a class taught by Marvin & I must say you make very detailed informative videos
These men know their stuff, keep it coming , gentle men, and thank you both, Andy, and David, proformence is their profession, a labor of love, of going, Fast, and First, the first time,,,,,and the last,
Your summary (~28:30) shows cylinder 1 as being both +4" and nominal, and is missing number 7. Which 1 should be a 7?
I realized I forgot something in my comment and I apologize. My name is Jeff Thompson. I'm 63 and live in Montana. Thank you for sharing your valuable wisdom. Jeff
Man, I just left out of Helena a few day ago.
Great stuff, Sir.
@26:42, you list the length changes for cylinders as:
Cylinders 1, 4,&6 + 4 inches
Cylinders 2, 3&5 - 4 inches
Cylinders 1 and 8 - nominal length
What about cylinder 7?
Thanks.
I believe this should be 7 and 8. Why? Because 1 can't be both nominal AND +4 at the same time plus 7 is missing otherwise.
@@Scarlet_1971_cuda so what you're saying is cyl. 7 & 8 lengths are the base line for the other + and - measurements
@@mfr5725 That is the way I understand what is being presented. Hopefully DV or Andy will chime in as well.
He renamed it to cylinder 1 just to see if people are awake.
the picture of the BBC showed the #5 primary tube significantly longer which doesn't match this
Spent 2 years modeling NASCAR motors for Hendricks Motorsports, and designing and fabricating headers. We went from 740 to 815 hp in that time, cica 2000. Cup cars were running 9700 rpm, we learned the primary lengths were way long by instrumenting engines on a dynamic and timing the negative pressure waves.
I would trust your info more than his. Actually out there winning races.
I speculate a factor that is very important to equal exhaust flow is not just the length but also the degrees of turn the exhaust header passage , if we add up all the degrees of turn that occurs in the bends ! you can have anywhere from 180 to over 360 degrees of turning the exhaust will transition through. obviously a near straight shot out of the heads exhaust port is ideal ! every bend creates restriction and turbulence,So in my theory when calculating the length we must take into account degrees of bend , a stream of exhaust gas goes through as it gets more delay through the bends than in a straight path , thus there needs to be a formula that converts bend to equivalent straight path , so we need sum of straight lengths plus our turn coefficient which is greater than 1 multiplied by the center tube lengths in the bends ! reverse thinking to take advantage of this theory , would be to use length and bends to time pulses into the collectors. I wonder if any tuners have performed experiments to study this concept to find the formula for the coefficient table for various header diameters .
Your How to Build Horsepower book is very good, and I am a Mopar guy, but concepts hold. Fantastic analysis and presentation.
Thanks David I've had this same argument with a many of other engineers for years and years, I've also proven that there is no difference between or sometimes a gain in the use of a crush bend radius over a mandrel bend. My own testing i used water flow instead of air as the mass flow. Didn't have alot of funds in those days and a water flow tester was easier and cheaper to make. Great video.
VERY interesting. From the beginning of time I was always told that a 90 degree engine is the optimum and equal length headers are a MUST. Thank you for clarifying.
Regarding the power gain with unequal length: I think it's stunning to see the effect in the 3000 to 3500 range. Over 5%, almost 7% at best! Imagine a "boost" of that magnitude on your daily driver 😲
My main worry is though: this arrangement will surely ruin the sweet rumble of the V8, as it evens out the pulses...
makes a crossplane sound like a flatplane. so to me is a win win situation
@@Hashiriya985 infidel!
Thank goodness for 2X playback speed.
Great idea! 😂
It is a must to make this video from a soul searching yoga video to a informative one
Lol. Ditto
2x was a bit much for my pea brain to process but 1.5x was a good compromise 😂
Talk slow, race fast !
Fascinating video.
Thank you for sharing your knowledge David.
Thanks, David.
The great thing about the Ford 2 liter is, it's a non-inerferance engine. Timing belt breaks, easy fix. Reliable, decent power, good on gas.
I (now) choose pipe off the head based on projected horsepower thanks to some insight by an ex-NASCAR exhaust builder. Assuming long radius bends and a good port-header transition, 1.5" off the head can handle up to around 500hp on a V8. But it needs to step up fairly smartly to not be a restriction. Works well in try-y designs as the primary-secondary join is often where you'd need to put a second step (third diameter) on a 4-1 so the headers just look like the only big difference is a stepped primary. There's a ton more to it than that though...
Step tube headers improve power over the entire RPM range. Tri-Y headers improve torque BUT NASCAR teams ran them(& still does on trucks & Xfinity cars for ground clearance. Nothing new. Mickey Thompson manufactured & sold Try-Ys 60 years ago & Thorley headers STILL sells Try-Ys for trucks, Thorleys don't make as much peak power as 4 into 1s, but they do improve low end torque for trucks pulling trailers. Engine Masters dynoed several different primary tube sizes & proved that bigger primaries make more power throughut the entire RPM range. My buddy runs what's considered huge primaries on his all motor 440ci small block Olds in his 86 Cutlas Super Comp car & it runs low 5s in the 1/8 & low 8s in the 1/4, 60ft times are 1.10. Car ran 7.97/168 with his 555BBC with 4,000ft air
@@bradgriffith4231 It's not just the pipe size where the low-mid gains come from. A lot of it is the collector size. You can't fix the low end of a 250hp v8 (late C3 Corvette for example) with a 1.75 primary by running a 2" choke without restricting it a little in the top end. The expansion of a single pipe into a 4-1 collector then back down to a small choke is too much turbulence. But if you build a new set of headers that start at 1.375 then steps to a 1.5, you can run the small choke without the same turbulence.
The effect is even more pronounced with a 4-2-1, you could go 1.375 primary into 1.5 for the secondary, then use a 1.75 final choke. You can get away with that because the expansion of a single pipe into a 2-1 collector is a lot smaller than a 4-1 and there's less turbulence.
Are those videos here on TH-cam? I'd like to watch, see if they tested collector changes as well
I've been building exhaust systems starting with lawn mower engine go-karts I'm racing things that others weren't just because since I was about five or six years old. I definitely see where you're coming from and you're the first person I can say thinks pretty much like I do. Working with 2-stroke knowledge keeps me from being put in the box building 4 stroke exhaust systems. People have always hated my designs but have never been able to surpass them. Most of my street cars were straight pipes , very quiet because of properly phased pulses until you got bad. The funniest I ever built was an autocross full-size 69 station wagon with a 390 engine side pipe. That was so two of us and our lady friends could do comfortably the all night rally with special stages. The horrified look of the Fiat Audi VW Datsun drivers was great. I can only imagine your life was every bit the thrill living in the European continent. Chevy small-block tri-y headers 1973 in high school was when I seriously started developing my shut up and listen ability. Have a crippled math teacher with a slant 6 Dodge start I think it was that would eat anything the kids thought they had that was hot. Young ones you need to learn from the Masters 😉
Fantastic video.
Worth its weight in gold.
Well Done.
Finally someone confirmed to me what I have been thinking for a long time
Equal length is only for even firing engines
LS and cross plane american engines requires unequal length as you are saying here depending on the fact that cylinders of 1 bank do not fire 90 degrees exactly after each other
Some more which require longer and some less which require shorter
With the exception of the 8 to 1 exhaust LS with rotary fire pattern where the collector recieves pulses sequentially in order
In this case equal length is the way to go
Very impressive information David. Currently looking at applying this to an Australian Ford Windsor 5.0 and a Cleveland 351 small port. Same firing order on both. Same objective on both. One is a work vehicle, hauling a load all of the time, the other a 4x4 and tow vehicle. Maximising torque at low rpm for drivability, and fuel efficiency is the objective. It would appear that Ford Australia took your advice to the extreme on the Windsor. The factory 4 into 1 system pipes, range from about 22" down to 4" LOL. The information that you have provided, along with your low noise exhaust video, plus the information in your How to Build Horsepower book, means I have some design directions to follow. Thank you very much. Regards Greg from Perth, Western Australia
Hello Greg from Perth! I have a 351C in a 3500 lbs Ford Mustang 71 and I drove it around with a built engine using Australian 2V heads but did not get very good fuel econ. I tried smaller carb but it ran hot that way which is no good. It had 10:1 compression dished Ross pistons and mild cam. At least it was fun. How is your economy so far on your vehicles? Have you thought of Injection? That and overdrive tranny would probably increase my fuel econ from 8 mpg to perhaps double that but I don't feel like switching fuel delivery. I'm more open to a T5 trans with overdrive. Got C6 now.
BTW, I love Big Block Chevies. I currently own a 1997 GMC Suburban with a 454 that I just built and upgraded last year. About 400 HP now. Not too crazy and plenty of power to get out of its own way.
There is a mistake at 28:00 minutes.
Cylinders 1 should be +4 inches and also nominal? One of those should be cylinder 7, right? But which one?
My reading has led me to believe that the same people selling headers have a second job of selling camshafts. Header design often locks you into a compromise because it forces you to design for a narrow range. Most people pick too short a tube and too large a diameter thinking they will magically drive above 5K rpm everywhere they go. Engines are dynamic and change constantly.
that's old school thinking. with his headers you can run anything
I remember when Walker and Banks did those tests! Wow David, thats a great story! thanks for sharing, never knew that was you behind those awsome walker mufflers! Flowmasters are junk, not mufflers!!!! Always learning from your vids! I learn something everytime i watch any of your videos and I've been around for awhile as well...LOL
Wow!.....your video just popped up.
...I’ve subscribed instantly!
I bought & read multiple times your 1st & 2nd (tuning the mini ‘A’ series books) ...40+ years ago!
Your knowledge & expertise was awesome back then, so without hesitation I will watch & support everything you do...
Regards Tony 👍
I was hoping you could address a few question's I had from this video:
1.) Would you apply the primary lengths changes to "balance-out" a 4-2-1 design / would it provide similar benefit?
2.) it seems a mistake was made in your presentation given cylinder #1 was listed 2X and cylinder #7 is missing?
3.) If mistype occurred, THEN wouldn't cylinders: 4, 6 & 7 = +4" longer , 2, 3, & 5 = -4" shorter, and 1 & 8 = nominal?
Any feedback you're willing to provide is GREATLY appreciated; keep up the good work!!!
I have the same questions like you. Have you found someone that has answered them for you?
@@frz_64 this YT-video has the math on calculating the lengths and better describes why
th-cam.com/video/Qp-I-csOcxo/w-d-xo.html
@@frz_64 yes; a YT-video called "unequal length headers - will this work" has the answer
@@ericschumacher5189 ok nice. i will search for it
I'm trying to my best to follow this for my upcoming build, and I'm getting a little lost. Basically what we're trying to do here is ensure that each cylinder's pressure wave arrives at the collector in evenly spaced intervals. At 26:35 we see our big moment that gives us the exact measurements (although there is an error here, cylinder 1 is listed twice and cylinder 7 is not listed), and it was in trying to apply those logically that I realized that I was lost. With a classic firing order of 1,8,4,3,6,5,7,2, evenly spaced at 90* intervals of crankshaft rotation, this doesn't add up in my mind. Please see my diagram below of cylinders and the number of degrees of crankshaft rotation that have occurred since the previous fire of a cylinder on that same bank:
8 - 180 7 - 90
6 - 180 5 - 180
4 - 90 3 - 270
2 - 270 1 - 180
It seems to me that we should be making the cylinders that fire after only 90* should be the ones that are 4" longer, and that the ones that fire after 270* of rotation should be the ones that are 4" shorter, however this does not match up with the "big moment". I considered the fact that this test may have been done with a firing order swap of 4&7 and perhaps even 3&2, but still it's not making sense to me.
Hopefully Mr. Vizard sees this, or someone else who can clear this up and help me understand.
Figuring out pulse widths is simple math really.
@@johnsnow1355 please share
@@lukeperry085 exhaust pulses have a known velocity so it's a simple math calculation of time and speed to know where the pulse is at a given time
@@johnsnow1355 can you please elaborate more? I have an in-line six cylinder engine, and I am designing an unequal headers, the firing order 1 - 5 - 3 - 6- 2- 4 it’s a Nissan tb 48, it’s going to be a true duel exhaust , so therefore two manifolds will be available, first manifold will have its primary tubes on 1 5 3 and the second one will be 6 2 4, I am doing this to get the scavenging.
I want to avoid the pulses from hitting each other.
What’s the ideal situation to get the most of it
@@johnsnow1355
“exhaust pulses have a known velocity so it’s a simple math calculation”
Apparently it’s a simple math calculation that you don’t know or are unwilling to share. 🤦♂️
Absolutely love your videos, I'm addicted to learning more. My project old peace of family history is going kinda in the opposite direction in performance, but laws of physics still apply.
I''m a diesel guy, drive truck (boss gets the money, I put in the hours, I'm on poor mans budget)
My little Cummins pulls 16,500lb EMPTY camper and horse trailer --- 19,000+ lb loaded camper and horses most weekends up extremely steep off road, logging roads, hard climbs. High country of WA.
Love the sound on my over worked straight pipe Cummins, whistling 32lb of boost, at a beautiful throaty sound 1,800 RPM, blowing out some quite hot EGTs. To say there ain't much carbon build up in my lil 5.9 Cummins.
BUT!!!! I also dearly miss driving my 71' F250, 360. Open up the 4bbls on my Edelbrock 600. I know the first answer will be, go put a 428 or 460 in it and be done. I'm weird about keeping as close to factory original as possible. Just want to tune the most efficiency, torque in what I got, 360 FE.
Factory original (I put on the manifold and carb, single glass pack, Edelbrock cam timing chain with +/- 4 degrees / stock timing keys). Everything else stock.
I'll never pull 20,000lb, but it DESPERATELY needs bottom end torque help, to help pull 10,000lb horse trailer up a grueling 6-9% grade, slow / low RPM - 25mph pull up, off road mountain.
Any short quick advice is greatly appreciated. My thoughts, please correct me where I'm wrong... Advanced cam timing "power curve down, raise compression 🤷", small tube long headers, long collector. I'm still learning... I'm on a poor mans budget.
Can't have something that doesn't do a job, yet I miss the sound of a 4bbl V8 working. Any pointers are greatly appreciated.
Way back in the day, one of the US "hot Rod" type magazines tried 5 different design fenderwell "headers" for a 57 Chev with a street-strip 327- there was around 50hp difference between the best and worst. IIRC, these were 4>1 designs, so there was even more potential for a 'better'.
Making it even more significant, that was only a 400-450 hp engine and if you project to some of the strongest modern NA engines, that could easily be pushing 100hp!
That why I love the old 255 Indy Ford snake exhaust engine of @1965...
Buick used a 15deg upward angle on their exhaust manifold, this worked like a reversion dam top side of the exhaust outflow - that 15 degs helped make a vacuum helping the exhaust flow turn on the down flow of those factory cast manifold.
Back in the 80' s there was Aftermarket custom exhaust headers known as AR ---
Anti Reversion
I went and searched for some more material and saw the child. It makes sense. My condolences, I can only imagine. May things be as well as possible.
When I started watching and getting the jist of where you were going with this video I immediately thought about pressure transducer sensor as a means to graph the pulse. It was neat to see you went the same direction. This is my first exposure to your TH-cam channel. Looking forward to watching more.
So glad this video Exists, thank you again David.
The formula mentions cylinder #1 twice. Is one of these supposed to be #7? if so, which one? Thanks
Cylinders 1357 are shown to show you the degrees of spacing between impulses of those cylinders! After cylinder number 7 cylinder number 1 is shown so you see the difference between crank quotation between cylinder number 7 and number 1!
This is all related to one bank of a V-8 engine
@@bigboycycles Cylinder no. 7 is not shown in the chart @ 26:33 so still a little confused.
Only confused with the chart there. The rest makes perfect sense that's why I was wondering if it was just a typo. Thanks
Some race engine header designs took that issue into account. They had tubes from the opposite bank collected to give 180 deg separation of pulses and equal length tubes to provide maximum scavenging of exhaust gasses.
Thank you for sharing this David. It's a pleasant review for me, with some extras, as I own and have studied judiciously, many of your books. If you would, Is primary length critical for an inertia pulsed exhaust header? I'm referring to the tri-y 4 cylinder example you showed. If you'd be so kind, as to take it a step further, have you tried a tri-y header on a V8 and do they scavenge at all?
Ive just started watching your videos and have learned a lot. If i could make a recommendation, it would be to reduce the amount of animations and cuts in your videos. Specifically the random zooming of the camera. It is quite distracting to ME. Maybe not others but to me it is. Either way lots of good info and always appreciate spreading proper knowledge.
I love the idea of setting track records through sheer "drivability"!!!!
Theres a lot to be said for having the torque right when you need it.
: )
I dont know why you hand out so much of your hard earned knowledge for free on TH-cam and next to nothing in your books. However unrecognized, it's greatly appreciated
Excellent information, DV! I'm looking to do a set of custom headers for my Ford small block and have a question. On the chart at 26:35, it shows cylinder 1 as both +4" and nominal length. Which one is actually supposed to be cylinder 7?
I learn a lot from your videos, as well as Unity's. Thank you.
This. Please reply.
I'm pretty sure you increase the length of the ones that fire back to back on same bank. It would be different from sbf to sbc because of firing order.
@@danielphillips1094 I've been thinking it out - pretty sure you lengthen the primary of the cylinder that fires 90° after a same-bank cylinder and shorten the one that's 270° behind another cylinder. I have no idea at the moment how much to add or cut off, but I'm pretty sure it'll depend on where you want your torque peak.
@@johngregory4801 that sounds right. I think nominal length would be determined by where you want peak power and the other cylinders would be adjusted to even out pulses at 180 degree intervals. I think what you're after is essentially a 180 degree header setup without crossing primaries from bank to bank.
@@johngregory4801 also in order to figure out how much to alter length you'd need to know how fast the pulses are moving. I believe I've seen the 500mph figure tossed around. I'm sure the info is out there, just have to know what to search for.
Thank you for sharing your great knowledge.
Thank you, DV; When you do your next video on this could you touch on pairing cylinders for scavenging in conjunction with optimizing unequal length primaries in addition to anti reversion, collector types, and secondary lengths? If that sounds more like two more videos, perhaps you could include the secondary length discussion with the scavenging pairing strategy with the appropriate un equal primary lengths. v/r wh
For performance street use, my favorite headers have been tri-Y design. I wish that was a more common offering. Street motors don't live at the top of their RPM all the time. And tri-Y is a not-terrible answer to the cross plane crank.
My favorite thing about this channel is how well the info agrees with things I know from: being a general physics nerd, building speakers, sailing boats, and aviation.
Also, it's always a "science based" approach here. The dyno is the arbiter of truth.
Dynos are not the arbiter of truth because they don't apply real world loads. I know numerous people with good running Super Comp cars that had them dyno tuned & they slowed down considerably. Went back to their old tuneup & the car picked right back up. Nobody other than some "exotic" spaghetti cars run cross plane cranks in street cars. Ford tried it for about 1 year in the Coyote & went back to a 90* crank. GM is using it in their race only Corvettes. The harmonic vibrations cancel any gain one might get in performance & the Ford cranks were destroying the engines. The longer the stroke, the worse the vibrations. It's the same as 4 cylinders without counter balance shafts, only twice as bad! As far as Tri-Y headers, they're ok for trucks pulling a heavy load but they don't make nearly the peak HP as a correctly designed 4 into 1 with merge collectors & step tubes with merge collectors are even better
I don't know this guys channel, just wandered here by accident and watched the whole 30 min.
SO I will summarise his answer. YOU ARE GOOD WITH EQUAL LENGTH IF YOUR ENGINE FIRES AT EQUAL TIMING.
Thats it, his title is a bit misleading.
Most of the time some cylinder are not firing equall timing so you have to adapt those accordignly.
Thanks for coming to my ted talk.
I’m way out of my league. Glad it’s a video. I can watch to overload, stop, and go again later. Learning LOADS. I always just accepted that four equal length into one was best. Hm. It leaves power on the table.
One wonders after decades of header manufacturing, did any header manufacturers ever know about this concept?
Equal length just seemed too obvious. I guess it wasn't.
Great video. I love when you guys and Tony get together.
I hate to defend header manufacturers but... this is probably farther down the priority list in the design for good reason. I would imagine that fitment and plug access are highest priority and honestly, would be happy if they could just get the plug access better on most. Trying to design a header that is the 'correct' unequal length means designing a header to one very particular build; IE with 'x' cam and 'y' manifold, optimized at 'z' engine speed. Maybe they could get something 'closer' with more R&D but it's extremely doubtful many would spend the time and money on anything but the highest performance/$$$ (highest profit margin) engine platforms.
Brings back memories of Phillip Smiths old book . Priceless to me.
Hi David, great video as always! I’m a physicist (so please don’t shut me off right away ;-), but I’m looking at your chart around time 21:18 . I notice that 1) you have a peak in performance enhancement at about 3500 rpm, and then 2) another right at 6000 rpm at the top rpm of the data. BUT, it is important to note that the performance enhancement is STILL INCREASING at 6000 rpm. This strongly suggests that if you went to higher rpm above 6000 rpm, that you will see continued performance enhancement with higher rpm. The reason is because you are approaching the second resonance of the header (2 * 3500 =7,000). But this suggests that this design is still leaving performance on the table. If the engine can’t go higher than 6000 rpm, then we have to bring that second resonance down to 6000 rpm by lengthening the header further. Does that make sense?
I am going to have to rewatch this a couple of times.... great work
I remember when I was a kid in the 60s my dad had a friend who drove sprint cars. They were a little bit different back then they had clutches and could take off without a push they were also limited to 300 cubic inches. His car seemed to dominate the track. His car was different in many ways but what I remember was it had enormous headers and an exhaust that went to the back. The other cars had stingers as a kid I remember fire coming from the during night races but of course not from my dad's friend's car. As I got older I wondered how much those wild headers with all those tubes were just for intimidation purposes. I suspect that some of them didn't even carry exhaust. I remember he sure was a lot faster and those pipes sure were convoluted.
do 4-7 swap cams help to correct this problem? thanks for taking the time to share your wealth of knowledge...
No you are effectively swapping the firing order left to right on the engine making the two hottest cylinders sit at the front instead of the back the pulses are the same
I know you said the numbers (primary lengths) in the video aren't optimum yet and need more research, but for what you've done so far it's a very worthwhile gain. My only question is... are they "close enough" to get similar gains on a small block Mopar? I know no header company is going to spend the time on an engine that's been out of production for 30 years, but I could always have a custom exhaust shop do a one-off set for me based on those numbers.
Just buy Dynomation6 and start plugging in your motor info.
yes, its going to make a gain that's substantial, make note that his cup car engine (the drivers side header) was displayed in this video...is Mopar! The other thing to remember he clearly stated early in this video was that the secondary(collector) length was a much more important factor than most people gave consideration to.
You're right everybody has it wrong but you have it wrong too. Equal length does not mean the pipes are all the same pipe length but equal "Flow" length. Every bend changes the flow length, and the amount of bend changes the length. There are charts and formulas to determine pipe length. Every bend lengthens the pipe flow so you add or subtract pipe length varying on the number and degrees of bends so all pipes geometrically flow the same. My 1968 Motor Manual went into great details about this. This is stuff everybody overlooks and nothing new. This geometry is also used in several other industries Water, Air and Gas and others
Wait a minute...at 26:30 into the video you show lengths of each primary tube, but where is cylinder #7?...and you list cylinder #1 twice!...Please correct.
I noticed many years ago on ford Ranger 2.3L trucks they has a Cast exh manifold that was a 4 -2-1 design.
So the on screen formula half makes sense but #1 is listed as both nominal and +4". The logical setup would be inside vs outside pipes. This would be calculable by the fps of exhaust gas flow so the timing of the gas to the collector should be predictable. Only issue is how much spent gas remains, vs how much vacuum the collector can pull. Ie, how much gas must the exhaust gas push on in front of itself. Pulse timing is all mute if the flow gets there at the speed of sound vs the speed of gas flow and the speed of gas flow is only possible if the system is relatively in a vacuum which should be possible in a race situation so ill go with it. Basically the issuse is that on a 90 degree cross plane crank v8 the cylinder timings are 180-90-180-270... Well its circular and that might not be from 1 in the firing order but i listed it this way to make a couple things easier to see. so there are always 2 pairs at 180 (out of 720) but those 2 pairs aren't 180 from each other. There are also 3 270 pairs. if you skip over to the far side of the 90 degree gap with your pairs you find 2 extra these 2 extra are the ones youd nornally use on a tri y but i suppose inside vs outside could make more power on that because the cylinders can chase the tail of another, sometimes things that arent even are still good because it can help to get the energies interacting instead of isolated but 90 degrees is generally considered too close. When you lengthen the trailing cylinder of the 90 pair you put it less than 180 from the next cylinder, that may be better than nothing bot for a true 180 all around like most 4 cylinder engines you also move to cylinder 180 from the cylinder that you intend to move to fix the 90 situation. Since all common cross plane v8 cranks are designed roughly the same this amounts to an inside pair and an outside pair. So on any particular firing order the mathematically ideal situation would be that anything done to 1 is done to 7. Anything done to 3 is done to 5. Anything done to 2 is done to 8 and anything done to 4 is done to 6. While i do see some of that in davids results its not all around so i wonder if they tried the mathematical way and maybe say something they didn't like or if they started with the transducers and pushed things around till it was good enough? Anyhow as far as i can tell with any firing order you would have 4 long pipes that would make an even running 4 cylinder and 4 short pipes that would make a different evenly running 4 cylinder. The groupings would be 1,4,6,7 and 2,3,5,8 same as you would combine them into a 180 header but only to simulate a 180 header (at a specific rpm, though it should help at all rpm) using separate banks which should help avoid excess lengths to make a true 180 header. I actually thought because it works out to inside vs outside you could use the extra pipe as part of an asthetic design and make some squigles or loops with center exits like some block huggers. If you placed everything right you could have a heart like appearance. This could be marketed with the old heartbeat logo they used to use for chevy as the heartbeat of usa. I almost don't want to post this so i have an oprotunity to make prototypes but i just don't have the time for that. Please credit my comment here if you market that and people buy it.
Hello DV, thank you for the accuracy explanation, like always. Can you make one for Turbo headers and maybe also on a plenum size too? At the end fig. is it cylinder No. 7 and 8 nominal and is the Nominal size for the calculation as explained before? Thank you and God bless you.
All the Subaru engines from early 2000s (and probably back to the 80s) up to 2014 had unequal length headers, really bad but sounded great. 2 of the pulses were running into each other with another right behind and then the 4th dragging behind. They sounded cammed with open headers. Then the turbo models sounded even hotter, until they put "equal length manifolds" on them and now they just sound like a mild engine. Still has a unique sound being a boxer 4 cylinder but just sounds like its neutered now.
DV, I think that a Tesla Valve (the scientist, not the car company) conduit design in a header could assist in solving some of the problems with back-flow due to pulse asymmetry issues that result in collector flow conflicts. Combining a high flow primary of the correct length to smooth exhaust pulse frequency alignment with a backflow prevention mechanism that doesn't significantly restrict flow would free up another HP boost like that achieved with primary pipe length tuning. That would be a fun experiment to try.
I had the exact same thought about the Tesla valve principle.
Liked the video, and agree with your points purely from a torque / HP band perspective. But (and you allude to it yourself in the vid) all the distinctive throb and personality would be removed from the exhaust note. That is sometimes the main reason people buy V8s.
Thank you for the valuable info very informative. Note: Please see if you can enhance audio quality of your productions.
@27:34, in your list, you have cylinder 1 listed twice, at +4 and at nominal, and no #7 listed.
What were the primary diameters? And lengths?
I would think that the gasses changing direction would need to be considered also in optimizing headers.
And 4-cylinder engines would be easier than eights due to crankshaft rod journal position.
And manufactures will optimize headers based on who is going to buy them, optimized engine, heads, etc.
Sorry for your loss.
The 30% of sound reminded me of the one of the reasons why MP3 files are relatively small. When the audio is compressed it drops the quieter sound waves that most ppl can’t hear under the loud ones.
I kinda got hopeful when you mentioned 4cyl. That you might touch on 6cyl. I’m interested in slant 6’s those things need all the help they can get. 😂
The leaning tower of power needs nothing but gas and time. They will outlive humanity.
When it’s all over the only things left will be roaches slant 6s and 90s jeep cherokees.
Exactly, I'm going to make my own headers for my slantie!
but people like me MP3/compression sounds and all-8 firing i can hear 👂clearly it's the really high pitching wines i can't anymore or at least not as much as im around industrial equipment that changed my ears and or brain tune's it out so lower human-male voice is easier for me to understand ect
there's a bit more wizardry going into mp3 than dropping 'quieter' sounds, it's dropping data that doesn't "matter" as much based on tests on people and stuff like that.
I tried 2" Tri Y and Equal length extractors on the mild big block chev in my Holden Monaro.
The Tri Ys always felt as though they made more power.
I would like to know about lengths for Chevy LS engines as the firing order differs from the standard Chevrolet V8.
LS firing order is 1-8-7-2-6-5-4-3 so 1/3 fire close together as do 2/6, so they'll see the pressure peak or overlap, whereas there will be a low between 7 & 5, and 4 & 8.
SBC & BBC order is 1-8-4-3-6-5-7-2 so it's 8/4 and 5/7 firing close together so they have the overlapping pressures. and a corresponding low pressure with the 6/2 & 1/3 pairs. the chart you show at 27:00 has cylinder #1 twice, but no #7.
then the #5 primary tube on the 572 BBC is significantly longer in the picture.
the low rpm HP & torque numbers would make a big difference for trucks especially, but I'm also looking at the high rpm gains for a drag car.
The principle seemed to be shortening the "delayed" cylinders that fire 270/450 after 0, so with the LS pattern of 1x7xx5x3 the equivalent should be 5+3 shorter, 7 longer, 1 unchanged. x8x26x4x would be 8+2 shorter, 4 longer, 6 unchanged
@@Stereomoo I wonder if it matters with an open header at all (one side will not help scavenge the other)? on an LS, it seems to me 7/1 should be long with 3/5 shorter (say 34"/30") to drive the peak between the 7/5 closer together and the 1/3 peaks apart. On the even side 6/4 should be long and 2/8 should be short (same 34"/30") for the same reasons. or maybe the 4" difference is wrong for that case? basically using a primary tube length to shorten or lengthen the time it takes for the impulse to get to the collector.
combining pipes with a Y or X-pipe is where the two sides being different may make more logical sense, if the pulses land on top of one another then you still get pressure spikes after the two sides combine. Optimally timing the pulses of one side to hit exactly between the pulses from the other side would make it work best for scavenging all cylinders. however, this difference should be tested with a set of y-pipes that start at equal length and add 1/2" increments to one side until a peak is found.