How to fight exhaust reversion when the primaries are way too short

General engine tech -- Drag Racing to Circle Track

Moderator: Team

User avatar
ptuomov
Guru
Guru
Posts: 3587
Joined: Fri Aug 07, 2009 3:52 am
Location:

Re: How to fight exhaust reversion when the primaries are way too short

Post by ptuomov »

The turbine efficiency works out such that as long as you don’t have exhaust blowdown interference (due to connectivity and distance), the shortest manifold is the best. Diameter is a trade off between pumping losses, power to drive the turbine, and scavenging effects.

Because of the emissions regulations, the cats must be placed close to the exhaust ports. And more and more passenger cars have turbos from the factory. Because of this, emissions legal aftermarket exhaust headers and manifolds will forever be shorties. Shorties which are so short that, for at least cross plane V8s without outboard exhaust, the exhaust blowdown pulse interference continues to be a problem at least at some rpms. This is the reality, and I don’t think the aftermarket performance industry is paying enough attention to that reality.

It would be nice if someone could figure out a new set of rules for exhaust design that are specific to exhaust manifolds and shorty headers. Long tube headers automatically solve the biggest problem of shorties (sufficient distance to avoid pulse interference), and I feel that the existing long tube header design playbook isn’t going to work with shorties or with manifolds.
Paradigms often shift without the clutch -- https://www.youtube.com/watch?v=cxn-LxwsrnU
https://www.instagram.com/ptuomov/
Put Search Keywords Here
User avatar
modok
Guru
Guru
Posts: 3321
Joined: Sun Jun 06, 2010 1:50 am
Location:

Re: How to fight exhaust reversion when the primaries are way too short

Post by modok »

I agree. In this case tuned lengths can be ignored. I think it is a case for the scatter cam. :D
User avatar
modok
Guru
Guru
Posts: 3321
Joined: Sun Jun 06, 2010 1:50 am
Location:

Re: How to fight exhaust reversion when the primaries are way too short

Post by modok »

exhausted wrote: Sun Nov 26, 2017 7:30 am
modok wrote: Sun Nov 26, 2017 12:46 am
exhausted wrote: Sat Nov 25, 2017 2:14 pm At some point here the observation is that a "plenum" on the exhaust side of a engine after a bit of "blowdown Length" does the same thing a plenum on the intake side...It dampens and attenuates pressure waves...What a AR device would look like in thos LS aftermarket manifolds would be the same as shearplates in intake manifolds...

We are getting close here to my talk about IR intake engines needing a exhaust collector volume, (Plenum). I know that that plenum can then be choked before atmosphere, ...Cats?
There may be a situation that exists where the turbo can be placed close to the head and far from the head and work ok, but in the middle will work worse. I found a similar case with the intake side with a turbo or blower

The turbine is probably functioning both as a centrifugal turbine and ALSO as a simple impulse turbine.
It does seem most here are aware of how unsteady the exhaust gas flow actually is, with very high flows during the peak of blowdown, and also some periods with very little action.
Energy being velocity squared, it may be seen that averaging the flow would result in less energy. A header that was had enough volume to make the flow going into the turbo more steady would actually reduce the available energy to run the turbine, but the turbines ARE tested at steady flows, and likely more efficient at steady flows, so we make it steady and get the turbo efficient at that flow it might make up for the loss. Both ways can work. As an impulse turbine you want to minimize volume and velocity loss from the exhaust port to the turbo.
Assuming the turbine is a steady state flow device you'd want constant steady flow into the turbo.
It may end up in a state where you want as short/tight as possible to preserve energy, OR place the turbo where the choke would be best placed in a conventional header.... so the header functions to efficiently couple unsteady flow INTO steady flow for the turbine. In the middle between those two choices could be worse than either, if you reduce the peaks of flow but still don't approach steady flow nor good pulse timing for the engine. I have not heard any discussion of this.
Ok, makes sense but who the hell is going to make two completely different setups to determine any difference much less measure it, A dyno can measure static load power but drivablility even for racing is what the real issue here is I think which means you have to build it. Some want to build it in a computer. Is that actually easier>\? takes one hell of a computer and software to do that...anyway, alas, we are stuck guessing.
The panacea of boredom and birth of "hey, watch this...

Anyway this whole thing is about controlling a very rapidly heated MASS that has been let out of a confined space and has created pressure waves as well. We can attenuate the pressure waves but until all the heat energy has been dissipated through the various mechanisms the question is how much of the available energy have we harnessed to augment the engines performance.

The folks who are only concerned about pressure wave tuning wind up with very large tubes because they are not trying to conserve momentum energy or heat energy. And they also find out that because they did the former they have to contain the pressure wave in the tubes for a long time in order to protect the other cylinders in the manifold from the positive waves going the wrong way at the lower rpms of the desired band. So when I look at a common conventional header, I shake my head. A ubiquitous testimony to the simpler task of building intake manifolds.
I built my first set of Cup headers in a race shop. Spent a lot of time in the engine and dyno rooms. I have said before but I can remember all the intake manifolds hanging on walls and in engine carts, discarded, cut up etc but there was only one dirty old set of headers in the dyno room usually under the bench. A loud statement of something, i"ll leave that to you.

Pressure wave tuning is obviously real but there remains is a lot more we can do.IMO (Sunday mornings can get a bit dour, sorry.) I appreciate speed talk forum. I hope my musings and time can help other folks who have the energy and desire to push forward in exhaust systems can gain, from the things I have seen and done, a more complete understanding of exhaust systems.
Darn right! :D :D
cjperformance
Guru
Guru
Posts: 3661
Joined: Tue Aug 19, 2008 7:20 am
Location: South Australia

Re: How to fight exhaust reversion when the primaries are way too short

Post by cjperformance »

ptuomov wrote: Sat Nov 25, 2017 9:36 am In my case, the primaries are 40mm or 1.577” inside diameter and the plan is to push about 1000hp/8 thru each of them. Of course, pressure and density matter, too, but I don’t think the ports or primaries are currently large. They are the same size that the stock 320hpnornally aspirated 5L engine has from the factory.

Exhausted, the LS7 four-in-a-row collector that gave the last prescious +10hp for the design team to take the engine above 500hp has the crossectional area at the collector that is more than four times the cross sectional area of individual primary. It just like the four pipes would continue except the inner walls removed. No choke whatsoever and, given the pulsed nature of the flow, a significant slowdown in the mean gas velocity. Since then, many of Chevy’s cast manifolds have mimicked that expansion, the primaries seem to exhaust into a plenum. If I understood what you wrote above, this is exactly the opposite of what you are recommending. Did they leave hp on the table, or are the LS7 primaries so short (pulse interference problem not solved) that the usual header design wisdom doesn’t apply?


With exhaust manifolds, as opposed to good tube headers be it long tube, 4.2.1 etc, especially with primaries that are very short and of very different lengths there are often "not ideal" or "all else being correct then technically not ideal" things that can be done to crutch or band aid a very unideal situation. As per Coyote and that LS 505hp manifold, love it!
Couple that with a turbo set-up and as i said i just think and stick to , short, hot, tight on pipe size, get it to the turbo asap. It works. But im not K-segg or any other big namer and can only wish i had a tenth of their R&D cash!!  I'd sure have fun if i did!!
But,,
 I can say that you'll be perfectly fine at 1.577" primaries for your 1000/8 target. Thats about .0156"2/hp prior to the turbo. Its actually a pretty good size for turbo stuff IME, one exampke close to your target is my customers LS at 930hp with 1.5"id fabricated manifolds is at .0152"2/hp with a perfectly smooth driveable and emission approved  (in South Australia) engine combo and tune. It will/has gone over that but its really well over what the vehicle can handle on street tyres so why push it.
 I will add that due to room the final branch where flow from all 4 primaries come together right before the turbo flange is actually around .013"2/hp but only for about 3 inches before entering the turbos.
In this case the manifolds on both sides are close to the same and both combine the front 2 cylinders asap then the rear 2 cylinders and the combined front pipe come together right below the rear cylinder then that ultra tight section into the turbo. I only have one not real informative pic of the manifold in this phone but you can kind of see what i mean i hope.
 Cheers,
 

PhotoPictureResizer_171127_130754784-600x800.jpg
You do not have the required permissions to view the files attached to this post.
Craig.
User avatar
ptuomov
Guru
Guru
Posts: 3587
Joined: Fri Aug 07, 2009 3:52 am
Location:

Re: How to fight exhaust reversion when the primaries are way too short

Post by ptuomov »

If those are hand fabricated exhaust manifolds, they look very factory.

What’s the path length between the exhaust valves for the 90-degree interfering cylinders? Which cylinders knock first?
Paradigms often shift without the clutch -- https://www.youtube.com/watch?v=cxn-LxwsrnU
https://www.instagram.com/ptuomov/
Put Search Keywords Here
cjperformance
Guru
Guru
Posts: 3661
Joined: Tue Aug 19, 2008 7:20 am
Location: South Australia

Re: How to fight exhaust reversion when the primaries are way too short

Post by cjperformance »

ptuomov wrote: Mon Nov 27, 2017 8:52 am If those are hand fabricated exhaust manifolds, they look very factory.

What’s the path length between the exhaust valves for the 90-degree interfering cylinders? Which cylinders knock first?
Manifolds are hand tig welded mild steel steam pipe.

Lengths between ex valves on 90 deg cyls are (aprox, i can get more accurate measurements next time its in here) 28" on the rhs between #2&#6 and 23" on the LHS between #1&#3.
1.8.7.2.6.5.4.3 F.O.
I was not there when the tune was done but with knock ears apparantly #3 & #8 were the first to knock which is not what I would have expected unless its some sort of intake manifold or chamber cooling related problem.
Craig.
User avatar
ptuomov
Guru
Guru
Posts: 3587
Joined: Fri Aug 07, 2009 3:52 am
Location:

Re: How to fight exhaust reversion when the primaries are way too short

Post by ptuomov »

cjperformance wrote: Mon Nov 27, 2017 8:02 pm
ptuomov wrote: Mon Nov 27, 2017 8:52 am If those are hand fabricated exhaust manifolds, they look very factory.

What’s the path length between the exhaust valves for the 90-degree interfering cylinders? Which cylinders knock first?
Manifolds are hand tig welded mild steel steam pipe.

Lengths between ex valves on 90 deg cyls are (aprox, i can get more accurate measurements next time its in here) 28" on the rhs between #2&#6 and 23" on the LHS between #1&#3.
1.8.7.2.6.5.4.3 F.O.
I was not there when the tune was done but with knock ears apparantly #3 & #8 were the first to knock which is not what I would have expected unless its some sort of intake manifold or chamber cooling related problem.
What’s the cylinder numbering convention?
Paradigms often shift without the clutch -- https://www.youtube.com/watch?v=cxn-LxwsrnU
https://www.instagram.com/ptuomov/
Put Search Keywords Here
cjperformance
Guru
Guru
Posts: 3661
Joined: Tue Aug 19, 2008 7:20 am
Location: South Australia

Re: How to fight exhaust reversion when the primaries are way too short

Post by cjperformance »

Ah sorry! I should have added cyl numbering #-o

1.3.5.7 left hand bank , front to rear
2.4.6.8 right hand bank , front to rear
Craig.
User avatar
ptuomov
Guru
Guru
Posts: 3587
Joined: Fri Aug 07, 2009 3:52 am
Location:

Re: How to fight exhaust reversion when the primaries are way too short

Post by ptuomov »

cjperformance wrote: Mon Nov 27, 2017 9:18 pm Ah sorry! I should have added cyl numbering #-o

1.3.5.7 left hand bank , front to rear
2.4.6.8 right hand bank , front to rear
So 8, 6, 5, and 1 are 180 degree exhaust blowdown interference victims at lower rpms and 3 and 2 are 90 degree exhaust blowdown interference victims. So if one runs the same ignition advance across cylinders, I’d expect one of the first group to be the limiting factor at low to mid rpms and one of the second group to be the limiting factor at mid to high rpms. Sounds like it’s 8 in the first group and 3 in the second group.
Paradigms often shift without the clutch -- https://www.youtube.com/watch?v=cxn-LxwsrnU
https://www.instagram.com/ptuomov/
Put Search Keywords Here
cjperformance
Guru
Guru
Posts: 3661
Joined: Tue Aug 19, 2008 7:20 am
Location: South Australia

Re: How to fight exhaust reversion when the primaries are way too short

Post by cjperformance »

Would be great to be able to fit the manifolds and turbos on the opposite sides (totally impractical in a vehicle of course) and see if the same cylinders knock or wether the corresponding primary pipes knock becoming cyls 1 & 6 !?
I guess we're off the topic here but learning what is actually doing what is a good way to learn how to stop/reduce the problem.
Craig.
User avatar
ptuomov
Guru
Guru
Posts: 3587
Joined: Fri Aug 07, 2009 3:52 am
Location:

Re: How to fight exhaust reversion when the primaries are way too short

Post by ptuomov »

cjperformance wrote: Mon Nov 27, 2017 10:49 pm Would be great to be able to fit the manifolds and turbos on the opposite sides (totally impractical in a vehicle of course) and see if the same cylinders knock or wether the corresponding primary pipes knock becoming cyls 1 & 6 !?
I guess we're off the topic here but learning what is actually doing what is a good way to learn how to stop/reduce the problem.
I think that anything that makes the distance between 180-degree interfering cylinders longer and the distance between 90-degree interfering cylinders EITHER very long OR very short will help the blowdown interference. With 1.8.7.2.6.5.4.3 F.O., one would want to maximize (within reason) the distance of 1-7, 8-2, 6-4, and 5-3. One would want to either make the distance 3-1 and 2-6 either very long or very short. For 3-1, very short is relatively easy but for 2-6 it’s difficult and very long is usually impossible.

What’s unknown to me is how the various merge angles etc influence the issue. Koenigsegg and LS7 manifolds seem to suggest that collector with parallel inlets/flow paths and no choke that is directed towards the cat or the turbine somehow works well.
Paradigms often shift without the clutch -- https://www.youtube.com/watch?v=cxn-LxwsrnU
https://www.instagram.com/ptuomov/
Put Search Keywords Here
User avatar
ptuomov
Guru
Guru
Posts: 3587
Joined: Fri Aug 07, 2009 3:52 am
Location:

Re: How to fight exhaust reversion when the primaries are way too short

Post by ptuomov »

Here's a question. From the pure flow perspective (assuming lengths take care of pulse tuning and interference), how much do I need to increase the cross-sectional area when combining different cylinders' exhaust flow?

Obviously, the separation of blowdown events in crank degrees matters a great deal and the exhaust cam duration matters to some extent. I'm looking at my case and I'm guestimating there's about 110 or so degrees when the mass flow from a cylinder is within 50% of its peak.

So how many percent should one step up the pipe size when combining cylinders that fire with following separations:90, 120, 180, 240, 270? I'm guessing that you'd want to step up maybe 25% when combining two 90 degree pulses, but when combining 180 degree or more separated pulses you don't have to step up much (at all?) from the pure flow capacity perspective? Is this correct? When building a four cylinder engine 4-1 header with equally spaced 180-degree separated cylinders, how large should the collector area be relative to the primary area?

The practical application here is that I'm trying to improve separation in my exhaust manifold, and that might be possible by grouping the following cylinders together: 1, 2, 4 on one side and 5, 6, 8 on the other side. The firing order is 1--2--4-1 on one side and 65-8----6 where each character location corresponds to 90 degrees. How large does the pipe have to be that brings out those three pulses on each side, compared to the primary pipe area? I am guessing that the 1,2,4 doesn't have to step up much because everything is separated by at least 180-degrees. On the 5,6,8 side it seems that one needs to step up as much as space allows, and by at least 25% more?

Any thoughts? Am I way off here?
Paradigms often shift without the clutch -- https://www.youtube.com/watch?v=cxn-LxwsrnU
https://www.instagram.com/ptuomov/
Put Search Keywords Here
User avatar
ptuomov
Guru
Guru
Posts: 3587
Joined: Fri Aug 07, 2009 3:52 am
Location:

Re: How to fight exhaust reversion when the primaries are way too short

Post by ptuomov »

Here are some sizing recommendations from Cone Engineering web site: http://www.coneeng.com/pdf/Area_Calculation_Table.pdf

I'm interpreting this as follows. This is ignoring pulse tuning, just sizing for flow:

When combining two pulses with 270 degrees or more, the recommendation step up 15-30% which in practice means one pipe size. I say 15%. The step up is not necessary from flow perspective, but it's like a step in a primary used to to reflect back a wave. So depending on where the merge is, up one size or no change needed *in my opinion*.

When combining 90-degrees separated pulses, such as when when using 2-1 collectors to join two banks of a V8 which combine a lot of 90-degree separated pulses but pretty far from the exhaust port, they recommend 30-50% area increase. I say 50% increase in area if combining 90-degree separated pulses relative close to the exhaust port, or usually three pipe sizes.

By interpolation, when combining two pulses about 180 degrees apart, increase the area by 30% or about two pipe sizes.

Opinions, not facts.
Paradigms often shift without the clutch -- https://www.youtube.com/watch?v=cxn-LxwsrnU
https://www.instagram.com/ptuomov/
Put Search Keywords Here
User avatar
ptuomov
Guru
Guru
Posts: 3587
Joined: Fri Aug 07, 2009 3:52 am
Location:

Re: How to fight exhaust reversion when the primaries are way too short

Post by ptuomov »

I bought a set of Ford Motorsports cast iron exhaust manifolds for the Coyote engine. Some observations. First, someone should make a turbo kit using these manifolds, the turbos would bolt on in place of the catalytic converters. Second, I don’t think these are quite as optimized as the Coyote tubular manifolds. The cross sectional areas aren’t quite as logically decided. Also the collector could align the flow paths better to be more parallel before merging them. I think (perhaps naively) they I could revise the casting to make it a little better, but who knows if those tweaks could be cast easily.

Can’t post photos from my phone because the site complains file too large. It’s 2017, getting file too large message feels like a hot tub time machine....
Paradigms often shift without the clutch -- https://www.youtube.com/watch?v=cxn-LxwsrnU
https://www.instagram.com/ptuomov/
Put Search Keywords Here
RCJ
Expert
Expert
Posts: 802
Joined: Wed Jul 25, 2007 6:15 am
Location: Oklahoma

Re: How to fight exhaust reversion when the primaries are way too short

Post by RCJ »

With the flat ls7 style collectors, could the pulses be providing a ''dam'' between cylinders.Cutting back on interference between clyinders
Post Reply