Overkill, but looks very nice and effective. Most systems out there will be either 3" or 2 1/2", and stepping down from 3" from header collectors to 2 1/2" tails is a good design. Helps with moisture evacuation and ease of fitment around the axle, and in no way harms performance on a combination such as this. H-pipe looks to be in a good spot, and would do more for acoustics than anything, but isn't hurting anything. Headers are doing all your scavenging work, and exhaust system is more than adequate.
If you apply the acoustic wave formulas to many commercially made "competition headers" you might find that they focus their benefit around the rpm range seen near the shift recovery rpm, keeping in mid how limited of a scope that 'customization' to an application can possibly be to a mass marketed product. I don't recall that specifically enough to say across the board if it still rings true but going through the catalogs and applying the educational analysis a couple of decades ago, along with conversations with the above mentioned header makers back then....at least that's how I recall it. Keep in mind that mass produced products have to fall within a demographic. The testing isn't so much to see 'if' the header does it's job so much as verifying the rest of your combination executes as well as YOU planned it to... ***this is the catch 22 of the whole discussion*** There isn't a universal way to apply a profitable product being big box priced, and to further obfuscate the topic it all depends on your specific combination and how it all balances together. With that being said, just the math alone to engineer your total combo is well past the hobbyists full understanding and patience, lest they go on a years long expensive learning curve but without the track time of a superstock racer. So either guess, work with $omeone, buy whatever is on the market and live with the result, or apply 50+ year old engineering (still current) to get your starting point and steer it from there.
IME 3" collectors lose torque for small cube applications, but bear in mind that this is with fairly unrestricted systems.
Shift recovery? Where have I seen this phrase and its importance mentioned before? (shhh, not too loud) Good stuff Tony. Thanks again!
I totally overlooked that aspect. Buick 350 headers have a 3" collector, so there is no real 'reducer' effect going on (when used with a 3" header pipe to mufflers). In which case, maybe the H-pipe is helping more here? I don't suppose it would be a day and night difference though, considering the rest of the combination.
I've welded in the reducers supplied with headers into the 3" head pipes and had been happy with the feel of things. [edit...anti reversionary is a huge question mark as the full wave reflections are generally taken advantage of in a more competitive world]
Hey, that's an awesome idea! 3" to 2 1/2" reducer inside a 3" pipe would help to reduce reversion and create a harder draw. :TU: I knew about this trick using manifolds and head pipes prior to a crossover, so makes sense.
If you look at formula suggestions and compare visually with Burns or other top racing headers, it makes you want to subscribe to the theory that only one pulse at a time passes the collector merge. Not that it matters for a Buick or any other street car with restrictive exhaust.
Good info. I've been concentrating lately on reversion theory as opposed to mass*velocity theory as it applied to long tube headers. The latter makes perfect sense with headers where you have a slug of exhaust moving down a tube long enough to pull a partial vacuum behind it when the port closes. Beyond that basic theory there are a whole lot of details but the average guy doesn't need to know very much of that beyond the effect on tube size and collector dimensions. But reversion theory is something else again. Here I'm not talking about the effect you get with shorty headers where the exit of the exhaust pulse before the port opens again creates turbulence in the collector and backflow up the tube, but rather the reflected pressure wave that results from hard surfaces like the baffle in a canister type muffler. In this case there is no actual flow back up the tube, or it might be more proper to say that such flow as does exist is very incremental, sort of like a wave on top of water where the water has very little motion but the wave travels nonetheless, and has power. Yet a pressure pulse still reflects off that hard surface and travels back towards the port, followed by a low pressure pulse or vacuum. We know for a fact that this occurs because a 2-stroke scavenger pipe could not work without this specific phenomenon. If you want to get into the specific physics of how it works, I suppose it has a lot to do with the compressibility of exhaust gasses. But the bottom line is that it does. My point is that this reversion theory has long been used by the makers of 2-stroke exhausts and is a proven science. I firmly believe that OEM's have known about this and used it for decades, and the advent of computer engineering analysis has allowed them to refine it to the point where that LT4 is capable of 650 streetable hp with the shorty headers we see in the photo. Without the specific header pipe and muffler used in the application it was built for I doubt it would. Reversion theory won't work with long tube headers. In those you have something like 6 ft of primary tubing between any two ports so every reversion pulse is divided by four. But in a log or shorty header that distance is shortened to the point where a low pressure reversion pulse which arrives at all 4 valves simultaneously twice per crankshaft revolution, can suck down any port that opens when it arrives. On our conversion cars I suspect we severely hamper our outputs by attenuating the reversion pulses and by causing them to arrive out of phase. This approach still does not appear to be as effective as long tube headers but obviously they are getting better at it, and if scavenger designers have to nerf their designs to keep from blowing up their engines there's reason for hope. I'd love to know what that LT4 makes with long tube headers. So in reversion design, the volume of the header pipe and the design of the muffler become all important. Much might be learned from expansion chamber exit and stinger theory. Jim
Hmm, quite thought provoking! I'm going to have to chew on this for a minute. lol It's refreshing to see I'm in congenial company. It seems I'm on the right track then, I just need to increase my understanding of these waves and their ability to create a drawing effect. Maybe I'm looking at it from the 'wrong' angle? Or maybe it's just a different angle. Not sure yet. What we know for certain, no matter the semantics of communicating thoughts, is that the exhaust needs to move fast and draw as hard as it can, whether it be from 'scavenging' or using reversion waves (synonymous and/or part of the same phenomenon). Am I getting warmer? We saw (if that smashed up header video was watched) that short restrictions introduced into the path of exhaust flow didn't hinder the power output whatsoever, as many would intuitively suspect.
Reducing reversion is what I coined it, but maybe it would be understood as something else? As far as 'in the competitive world' I was never truly interested in making money off this stuff or trying to create something that could draw capitalism's attention to it, as I believe that money will always get in the way of a true solution, which takes a back seat when it should be behind the wheel. (((no no, we can't do that even though it makes it better, simply because it would cost too much, and this other way here is more profitable, etc. etc. so on and so forth))) Anyway, perhaps this short-lived restriction (which encourages velocity) has more to do with enhancing the reversion wave phenomenon and its negative draw effect on the gasses behind it, rather than physically limiting how much gasses actually pass back up the system toward the engine--or maybe both? Regardless, we know it works well with a manifold setup placed before the merge pipe/crossover. I suppose it would just boil down to refining it.
??? There might be some that disagree with that. The way I interpret the formulas is that the longer the pipe, the lower the rpm range the wave reflection works in.
This would be true in a 'general' sense, but is relative and dependent on engine displacement, gas release timing/duration, the actual time it takes for the gasses to reach the end of the tube vs engine RPM/timing (where will the first pulse wave be when the next one is introduced, indicating that it is dependent on RPM and cam timing/duration, among many other factors that would affect the 'scavenging' effect), and would indicate the synergistic 'sweet spot' on tube diameter and length, as well as collector dimensions to accentuate this effect. Volume and velocity (and the balance between these two) become crucial factors in determining how well something will work, and makes one realize there is sooo much more to all this than simply bolting/welding parts together to move a quantity from point A to point B (as you already know). Blast it all, I'm out of time again and wish I could move this thought process forward a bit more, but alas it must wait until next time. It'll give me more time to digest and maybe realize the areas that I may have overlooked or misinterpreted/misunderstood.
I guess I would ponder those thoughts in the forums that the Nascar and Prostock header designers and engine developers post in. Their headers are fairly long for 9-11,000 rpm levels. I wonder how sound travels in a high temperature and pressurized environment? I'd be curious if the merges and branches act as reflective points or if there's influence from the other cylinders. I imagine sound has a pretty good influence over distance if concussion from noise bombs are effective. Rhetorical thoughts, obviously. All good discussion so far.
"Reversion theory won't work with long tube headers" Fair enough, that statement is not completely accurate. Of course it will work, but the effect will be diminished to the point where it is far overshadowed by the mass*velocity effect of the exhaust pulses coming down a long tube. The primary tubes very effectively isolate each port from one another so only 25% of each reversion pulse will reach each port, while each pulse is 25% of one engine cycle. In other words, each port provides one pulse but only gets back 1/4 of one reversion pulse in sync with its requirements. It would be true that some additional scavenging could be achieved that way and maybe that is the direction the next phase of exhaust development should go in, but while the two effects could be additive they are very different from each other in the way they work so it might not be all that simple. That's a direction I had not even considered. By contrast in the log manifold nearly 100% of the reversion pulse would be available to draw exhaust out of the open valve, but there is no or rather, very little benefit available from the mass*velocity effect. Can the system be designed to boost this feature? Maybe. I guess that is where these reversion cones might come in. I expect their exact dimensions and configuration could tailor them to operate in either the mass*velocity mode or the reversion mode, or perhaps they could benefit both. Quite a lot of unknown territory there. As for the sound, I consider that to be a manifestation of what is actually going on in the system more than a method of manipulating it, but maybe there's more to it than that also. At the very least another means of visualizing pulse propagation. Anyway fun stuff to think about, huh? Jim
Good advice. What are your conclusions? I suspect you already know, since sound waves travel faster and farther (before dissipation) within greater densities and not at all in a true vacuum, because the waves are the effects of energy being transferred through particles (the more there is, the greater the effect because there is more to be acted upon). Temperature tends to increase or decrease density with the expansion or contraction of molecules, and so cooler would indicate greater density with compressible mass. EDIT: I just looked up some of this basic stuff to see how my memory was, and it turns out I was incorrect about how temperature affects the sound waves in a gas, which is not dependent on density, but rather the tension of the medium. Seems I need to brush up on my definitions among other things, but we still know that there are circumstances affecting exhaust gas travel outside of the readily obvious. There is, however, much more going on that simply these two aspects, with overlaying and intertwining effects either accentuating or counteracting others. I suppose we could discuss how photons (and in fact most, if not all matter) can behave both as a wave and a particle, depending on how it is observed, and can change location in timespace depending on whether or not it is observed (!) Wrap your brain around that. Scientists are still trying to figure that out to this day. There's a few 'theories' (remember the scientific vs layman semantics of the term 'theory') so far, but the pursuit of knowledge is always exciting for any inquisitive mind, and so the quest continues. The formation of stars starting off as compressed hydrogen (understanding how this happens is fundamental to the formation of everything in the universe, including ourselves...) and the nuclear fusion that triggers ignition under gravitational forces, then the gradual stabilization leading into main sequence shows how density, mass, and temperature can affect particles/waves and release of energy, but I fear this would be too in depth for the purposes of this thread--or would it? Just how deep down this rabbit hole do you wish to traverse, and is it applicable to this discussion? Mayhap...but this is only my limited understanding of things, and I can always be wrong. Indeed sir.
I'm stating that both OEM and highly developed racing using different methods don't disprove each other. There's a reason each does what it does. The circumstances are completely different. It's well known that opening an exhaust valve into a pressurized factory exhaust system running a computer would have very different camshaft events than and general needs than an unrestricted one. I believe that people selling headers and engineers writing articles are applying already well known principles to an 'in between' area (OEM and top level racing), and describing their suggestions in an easy to understand, somewhat foolproof method. I don't think it was too difficult to engineer a 650 hp LS engine, many have been done with other platforms. I believe many other builders can get them more driveable than the typical hot rod engine. Someone has to get paid for that time. Basically in any midrange consumer market with competitive pricing a couple things happen...1. is the customer and builder basically have one shot to get it right and 2. if a builder already has an outstanding package that vastly exceeds the competition, how do you price it knowing that anybody REALLY good can buy it and duplicate it and therefore negate your exclusivity to it's development? So...off on a tangent we are, trying to figure out if the main point is collaborative learning, deciding on which aspects have more influence and when how to use them, or IF any discussed concepts are valid. The internet is a treasure trove of 'free info', with the reader trying to discern the BS and popular opinion from the real poop. Books are inexpensive, SAE papers are still cheap. Buying and trying parts are a bit more $$ but the user is rarely qualified to discern more than 'it works for me'. A decades long racing effort is more expensive yet... Paying a shop for their expertise doesn't sound too bad except that one can easily 'buy a bigger stick', lol.
This is all fun to me, but I've always been a proponent of working with what you have. This would include trying to understand what's going on and how/why things work, not so much to develop something that's already been done for other applications for a different application, since there is no point in trying to 'reinvent the wheel', but to apply this to the platform(s) on this forum (Buick 350's here, mostly), using what we already have as a baseline. This would include maximizing the efficacy of using exhaust manifolds and existing header designs through the rest of the exhaust system itself, and wouldn't necessarily be exclusive to the Buick 350. It would be far more cost effective to optimize the rest of the exhaust system when considering these things. Perhaps the discussion would be more appropriate and 'back on track' with this.
Always the over explainer here... As far as buying a bigger stick, consider the cost of paying a guy to kind of quick engineer a custom or altered current header to work with your already well planned out package and you'll quickly see that many more cubes, expert head or machine work would be a better starting point financially...which is why we don't see so much of this now. Anyone buy some Lemon's or ARH's lately? Not for Buick most likely. (then what you you put behind that?)
