you do on some engines. fullsize ls powered trucks loose a lot going to big. im sure its due to the tuning and can be overcome by tuning for that but they do when stock. now 350 buicks and most carbed engines ive never seen it but it could be cause we tune the carbs to overcome it.
Sean Still much better than the square log that bolted directly to the head. I know this has been asked before but what is the exhaust port spacing difference between the SBB and the SBC? Is there any chance of making an adapter plate? Thanks Paul
Going too big too close to the engine (collector diameter) and too short of collector is a fast torque killer with any engine. Tuning only helps so much.
Hi, I personally do not think it is possible to use a SBC header with adapter plate in a Skylark anyway. Here is how tight it is with the steering shaft.. This is my old car with different turbo setup but same idea... Also keep in mind that this setup gave 450 HP and 470 Ft of TQ at the tires on low boost with NO head porting an a low $ 355 cube Buick 350. Estimated HP was about 250-260 realistic flywheel HP without boost but was making about 525 at the crank with boost. Pretty respectable considering the lack of head porting, stock valves, and no fancy parts... Very mild Poston 114 cam, about 8.7:1 compression and small turbos. Less than ideal log style manifold but hey it certainly works for a relatively low HP application like this. I am sure if it was naturally aspirated this type of manifold would kill a lot of power VS a long tube header but when turbocharging I don't think it is as big of an issue unless going for a max effort type engine.
Keep in mind this is the way I purchased the car, and other that what I can confirm with my eyes, this is what I was told was done: 350 Mild crane cam and lifters Hooker super comp headers 3" exhaust to mufflers - 2-1/2" out from there H pipe So... Good, bad or ugly? Overkill? Losing, gaining, neither?
That is a good system. Probably sounds great too. I do 3 inch to the mufflers because that is where the restriction is. 2 1/2 after the mufflers is fine because very few mufflers will outflow a 2 1/2 inch pipe (none that I know of.) Personally I think the Hot Rod video was pretty definitive. You kill nothing with big pipes after a motor with headers. Your system is a bit of overkill for a 350 but it cost you nothing. Enjoy it.
On that particular build (455) a log manifold was all there was room for. Simply no way to do headers of any configuration whatsoever. What is not clear is that for a log it is pretty well done, with steps in the diameter and directional joins. But it is still a log. I do think that if we removed the glass packs and instead put a single canister type muffler at the rear it might work better, but that's pretty low on the priority list. IIRC torque was over 400 and the car has plenty of power for anything other than drag racing. We'll soon be dropping an LS4 into one of these and my goal there is to fit it with headers that have at least a 24" primary tube. I will not argue against long tube headers being the best way to go. I know they are. Even back in the day a swap from manifolds to off the shelf headers was good for about 40 hp on a generic small block without changing anything else and that tells you right there how much more efficient they are. The point I wanted to make was that even the iron manifold systems are tuned and you can screw them up with the wrong configuration, losing much horsepower in the process, and in those cases it isn't all about restriction. I suspect, this theory that you need some back pressure with those systems is incorrect. I suspect that what you need is reversion, and because people do not understand reversion they just thought it was back pressure. Jim
Agreed! (whole post) Reflective wave tuning. I think people are forgetting that aspect and only concentrating on flow. My comment to the 24" primary headers is that it presents itself with an opportunity to merge those to "two's" rather than a 4-1 collector using the same appx space, giving the same benefit as the long tube headers (2 tuned peaks of influence). That's how I've done it anyways.
Good point, I'll look at that since the merge would be under the car. Any recommendations on secondary tube length? I'm thinking at least a foot, maybe something on par with collector length recommendations. On the topic of reversion, secondary pulses, or reflective wave tuning, that can be a very powerful force as we see in 2 stroke usage (they really have to de-tune the scavenger slightly to avoid blowing up the engines? That's almost insane!) and with a V8 you would have 2 times the exhaust pulses in the same time interval. But we've not been seeing any 2 cylinder 2 stroke engines with a common scavenger pipe that I can recall, that would indicate that either they haven't found a way to make that work, or nobody has considered it (unlikely). It seems more probable that splitting the reversion pulse between two ports cuts the scavenging effect to half for each of them, and the volume of the chamber might have to be increased. Back to 4 stroke. That's not to say that reversion isn't important in a V8 engine exhaust. Just that as the matter now stands it is of much lesser importance than good header design. But look at the exhaust design on this LT4: http://ls1tech.com/forums/conversions-hybrids/1827238-lt4-8l90-shop.html Much too short to be an effective header yet the engine produces 650 hp. Does reversion play a part here? I suspect that it does. Would that engine come with a recommendation for header pipe size and length? I'd say that would be an excellent question. Jim
I don't want to link that forum due to needing to be registered to view, but the paraphrase was intended to boil down a lengthy story illustrating clearly that scavenging is extremely influential to a 4 stroke engine, to the point that it can be considered as part of the major metrics that comprise the build. The problem the mainstream guys face is availability, as mentioned. I wouldn't arbitrarily pic a length and diameter when Pipemax or Wallace racing (among many) calcs are out there. It would be careless to not math out the engine's characteristics from the capability of the head, factor in how the cam sways the curve, and factor in where you want your header's influence to be in accordance to the goals of the vehicle and it's use. For example, peak torque and peak hp could be used for branch lengths, or maybe spread them a bit... Again, rather than A-B-A testing, it works well to just plan it out with intentions of overshooting your goals a bit rather than wishing you had 40lbs more torque or something. [haven't looked at the ls forum link, just commenting that tri-y headers can be tuned for high and low rpm, as they are for nascar, etc.. Another opportunity is for pairing the merge @ 180* (or whatever) firing intervals to eliminate the out of sequence pulse when using a 4-1 type]
The LT4 exhaust is a Tri Y header and is beneficial for low to mid range power. See post #9 Notice how the passenger side collects differently than the driver side. The primary "Y"s in all 4 cases receive exhaust pulse that are 270* and 450* degrees apart. Paul
All good points, but pipemax is too expensive if it isn't your primary business and I didn't see anything at Wallace that would help you calculate the header pipe for a 455 in terms of reversion pulse timing. I think a good starting point would be the length and diameter of a big block header pipe on a factory Buick. I know they were 2-1/2" diameter but have been unable to find the length. Finding a reversion calculator for a scavenger pipe would be another good starting point, but I have no idea if one is even available, seems like it would need to be a graphic program where you move lines. Then we could possibly come up with a reversion chamber that would work with the big block, swapped in where the glass packs are now. As for the Tri-Y for an LS4 I think my default would be to ask Ed Henneman, but I'm not sure if he is still around. He'd be in his 70's at least, and it's been over 30 years since I last talked to him. Ed was maybe the first to promote velocity based header design in the early 60's, as most everything up to then had been restriction based. And as I said, I think the OEMs have been doing reversion based work for decades now. They are way ahead of the header industry in that field. But this I think we can assume, in a primary tube longer is better, even if the amount better is in question. That probably also applies in a secondary tube in tri-y designs. Now taking Vizard's suggestion of 24" as a minimum primary length and recommendations of 12+" as minimum for collectors, a secondary tube should fall somewhere in the middle I would think. Ed really liked a primary tube length around 40". That is a little hard to do, but would indicate that the secondary could easily be 16". Here we really are not dealing with reversion but velocity based design, where a longer tube helps. It might seem like I'm just pulling numbers out of thin air here but that's really not the case. Both Ed and Vizard have done extensive real world testing, determined what actually works, not what magazine editors want to hear, and these are their recommendations for the most part with a little extrapolation by me on secondary tube length. In my opinion we should build on the shoulders of giants whenever possible and leave the flash and glitter to commercials. Jim
If you look at a long tube, the pressure pulse from the blow down portion of the exhaust cycle is pulling a low pressure behind it as it travels down the pipe which continues to evacuate the cylinder as the piston returns to TDC. This should drop the pumping losses to nothing and also present a very low pressure when the intake valve opens. I would think there would be no downside to a longer pipe. The only other consideration would be pipe diameter based on the amount of exhaust volume due to engine size, power adders ..ect. It would be interesting to have a pressure transducer at different points and actually scope what is going on in different types of exhaust systems.
The people you refer to that did testing are also using acoustical wave tuning formulas, as have been around forever and then some. You'll have to play around with the wallace site formulas, there's enough there to do what you need. It's your learning curve, but I'd keep the factory starting point in mind as a frame of reference only. For better forum reading on the topic I'd suggest Speedtalk (advanced), to hear what the top minds have to say about the subject.
The website is still there and has "2016 marks our 54th year" near the top. I bought a set of thick flanges for a 455 from there back in the early 90's. http://headersbyed.com/
Couple of things I'd like to point out: With my previous comments about a merge/crossover (I've talked about this elsewhere before in the past) was not meant as a replacement for headers implying equal gains, but as an alternative to maximize the effects of log/manifold type exhausts where headers are not a (feasible, possible, or otherwise desirable) option. If you can put it as close to the engine as possible, it will work better of course, for reasons mentioned earlier. The main idea here is to try and gain as much 'sucking' as you can muster so that it will help overcome the negative effects of not using long tube headers. Even if log tube type manifolds are created, angling the inlets more toward the flow direction is better than a "T" type inlet with 90* exhaust flow turns and ugly, power robbing turbulence, and stepping up to gradually increased sizes as more cylinders are introduced into the 'log' is a good idea, but then there's the problem with the last one which ends up with a small tube dumping into a much larger one, and ultimately ends up better off as a 'tri-y' type configuration in shorty form. Then you're like 'why?' Just use the damn manifolds. lol (as a side note, you'll see a lot of more modern OEM manifolds designed this way) Gains over the original iron manifolds would be negligible and pointless (or perhaps even worse!), especially when you'd have to use a merge/crossover just like with iron manifolds to get the most out of it. I had a design myself on this back when I was trying to come up with a 'shorty' design, but scrapped the idea in favor for other things (though the R&D I spent on it left a memorable impression and showed strengths and weaknesses I had not previously considered), which was all a valuable learning experience. So in the case with this log type manifold (as it is in the pictures) is the need to draw it away from the engine as best as possible. If there's no room for a crossover "X" anywhere, at least you could use nothing but tubing (sans muffler here), putting the muffler(s) as far back as you can, as well as coating those head pipes from the log manifold itself back to the threshold where the gasses are cooling significantly so that your 'stove pipes' can draw as best they could given their limitations. Don't underestimate thermal draw. Would this significantly increase power? I think it would at least help, and wouldn't be too terribly involved vs the headache of a custom set of tri-y's in a place with severely limited clearance. Not that you're asking for advice, I just wanted to show what could be done to improve things for maybe others who may come across similar scenarios in their applications. Back pressure and optimal velocity can often cross over and become confused. I do not see how reversion is a desirable outcome to improve exhaust evacuation, as my understanding of exhaust reversion is the reversal of gas flow back toward the engine. There are ways to reduce reversion which improve performance. My understanding (haven't read any of Vizard's stuff before, and saw references to him here and skimmed over his material and I was like 'yep') is that optimal velocity is what you need, and more so in one direction. Pulse wave reversion is what makes short tube headers/manifolds perform worse than full length headers, and is why I assert some form of assistance in the exhaust system itself where the manifolds are lacking. Hey maybe I'm wrong? I'm all about learning.
All comments regarding the tri-y' header design on that application seems very doable and would be the best way. Thanks Tony and Paul. :TU:
