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

[ptuomov]

AR interface at each head/manifold junction.

How about this idea.

Suppose the firing order is 13726548, that is, on the passenger side bank 13-2--4-. Suppose one is trying to solve the 90-degree exhaust blowdown interference problem at high rpms. The problem is that #3 blows into #1 during its overlap. Adam Christian's Coyote idea was to get #1 and #3 pulses as much on top of each other as possible, such that we only get pumping losses but not cylinder evacuation problem in cylinder #1. So it's desirable to delay the blowdown of #1 and advance blowdown of #3 to get them on top of each other as much as possible.

Make the black path from #3 to exit a large diameter "expressway". This way, hopefully #3's blowdown gets done early. Make the red path from #1 a small diameter "cowpath". This will hopefully delay the blowdown of #1 and keep the dynamic pressure facing the exit high so it'll fight hard the pulse from #3. Advance and shorten the blowdown of #3 and retard and lengthen the blowdown of #1, and hopefully you get the pulses and flow more on top of each other and save the overlap of #1.

hooker_exhaust_manifolds_hero.jpg

[Calypso]

What if, instead of the ”cowpath”, you’d put venturi type of contraction to the end of #1. Less lossy from flow perspective, same speed as ”cowpath”, poor flow to reverse direction and the change of area would reflect part of the pressure wave energy from #3 back to the collecting tube with same sign. It would do that also back towards valve, when #1 opens, but perhaps that is less detrimental.

[ptuomov]

What if, instead of the ”cowpath”, you’d put venturi type of contraction to the end of #1. Less lossy from flow perspective, same speed as ”cowpath”, poor flow to reverse direction and the change of area would reflect part of the pressure wave energy from #3 back to the collecting tube with same sign. It would do that also back towards valve, when #1 opens, but perhaps that is less detrimental.

With the caveat that I don't know what I'm talking about, it's my intuitive impression that to create the same sort of mass flow effect, I could either use a medium-diameter long-length restriction or a small-diameter short-length restriction. Which one is more efficient from pumping loss perspective? Which version of the "cowpath" would be more effective in placing the pressure pulse from #3 at the #1's exhaust valve as early as possible?

[user-23911]

The easy answer is to use a short cam.

[ptuomov]

The easy answer is to use a short cam.

Shorter cam is the easy answer for reducing the 180-degree exhaust blowdown interference at low rpms. Once you go below 180 degrees on exhaust duration, you can't get 180-degree interference no matter what your manifold looks like.

For the high rpm, 90-degree interference, just a shorter cam is not necessarily a very good answer. Longer the exhaust duration, the easier it is to get the two blowdown pulses effectively on top of each other. That Porsche Cayenne Turbo uses those extra short exhaust cams on half the cylinders, and the 90-degree blowdown interference appears to get worse with them.

Less overlap may work to mitigate the worst consequences of the high rpm 90-degree blowdown interference in the victim cylinder. So spreading the lobes apart is one crutch in the combat medic's kit.

I'm hoping that whatever the best camshaft crutch is that there's an exhaust manifold crutch that's additive...

[hoffman900]

Smaller exhaust valve
50* seat on exhaust valve
Sink exhaust valve
Less seat-to-seat duration on exhaust lobe
^potential combination of all three.
A/R Chamber at some point in the exhaust system.
^potential combination of all four.

[ptuomov]

Smaller exhaust valve
50* seat on exhaust valve
Sink exhaust valve
Less seat-to-seat duration on exhaust lobe
^potential combination of all three.
A/R Chamber at some point in the exhaust system.
^potential combination of all four.

Those sound like fine methods for combating generic exhaust reversion. They may also work with the 180-degree exhaust blowdown interference.

I'm concerned about the 90-degree exhaust blowdown interference at high rpms. Given the space available, I believe the only workable strategy is to get the 90-degree separated cylinders to blow as much on top of each other as possible. I don't understand how generic reversion reduction techniques are going to help with that. Intuitively, and this has not facts or even just theory backing it up, it seems to me that one should try to do the polar opposites to cylinder #3 and cylinder #1. Applying the above methods to #3 seem counterproductive, no?

[statsystems]

Smaller exhaust valve
50* seat on exhaust valve
Sink exhaust valve
Less seat-to-seat duration on exhaust lobe
^potential combination of all three.
A/R Chamber at some point in the exhaust system.
^potential combination of all four.

I definately like the smaller valve idea with a 50* seat.

[ptuomov]

Smaller exhaust valve
50* seat on exhaust valve
Sink exhaust valve
Less seat-to-seat duration on exhaust lobe
^potential combination of all three.
A/R Chamber at some point in the exhaust system.
^potential combination of all four.

I definately like the smaller valve idea with a 50* seat.

Smaller than what? And for both #3 and #1, or just one of them?

[cjperformance]

A couple of ideas if room is very tight,,,

Very short stub pipes from each cylinder that protrude into a tapering log manifold

OR manifold exit right under #3 & #3 having a larger but the shortest flow path to the exit.

[ptuomov]

OR manifold exit right under #3 & #3 having a larger but the shortest flow path to the exit.

That's what I was thinking with the "expressway #3" idea.

Does the "cowpath #1" part make sense?

[statsystems]

Smaller exhaust valve
50* seat on exhaust valve
Sink exhaust valve
Less seat-to-seat duration on exhaust lobe
^potential combination of all three.
A/R Chamber at some point in the exhaust system.
^potential combination of all four.

I definately like the smaller valve idea with a 50* seat.

Smaller than what? And for both #3 and #1, or just one of them?

You can shrink the valve to the size that still let's you get your bottom cut on it. As an example, you can use a 1.500 valve in a SBC with a 50* seat.

[cjperformance]

OR manifold exit right under #3 & #3 having a larger but the shortest flow path to the exit.

That's what I was thinking with the "expressway #3" idea.

Does the "cowpath #1" part make sense?

I think we're on a similar track as in by making #3 as direct as possible to the exit and as big in diameter as possible it gives area for the volume of gas but kills velocity as much as possible. Seeing that #1 gas pulse will have good velocity before dumping into the exit near #3 ,, #3's low velocity pluse then 'theoretically' has less chance of affecting #1 , keeping the exhaust passage for #1&2 tight/high velocity and paired as in the pic you posted also gives a basic ar dam against reversion into #1 at the #1 / #2 junction. I would also keep #4 as high velocity as possible , having a two cylinder firing space between #2 & #4 gives 'some' time for pressure pulses to settle so having a high velocity shot of gas after this pause should be a good thing.

[hoffman900]

Smaller exhaust valve
50* seat on exhaust valve
Sink exhaust valve
Less seat-to-seat duration on exhaust lobe
^potential combination of all four.
A/R Chamber at some point in the exhaust system.
^potential combination of all five.

Apparently, I can't count.

The reversion is happening at overlap, regardless of low or high rpm. Killing low lift flow will allow your duration to be where it needs to be, but will kill some flow at overlap.

We had to do something like this on an engine of mine. With OHC and no way to run higher rocker ratios, you need a certain amount of seat to seat duration to maintain cam dynamics for the lift you want. Shrinking, sinking, and a 50* cut on the exhaust valve all played a roll in reducing overscavenging, while allowing the cam to be able to lift where it needs to be.

Calvin Elston gets into this with headers that scavenge too well. You shrink the exhaust and run a bigger intake. That's not your problem here though.

[modok]

If the system is restrictive the 90 degree interference might be a problem at higher rpms
At lower rpms 180 degree interference is likely a greater problem.

What do you mean by restrictive?

And any ideas how to cure the 180 and 90 degree interference problems in a short exhaust manifold?

Well, actually yes. I can think of a number of solutions but you have to be careful about solving problems that might not actually be problems. i think I've said before sometimes you think there is a "dip" in the power curve and blame that on the exhaust, fix it, and find out it wasn't a dip you just got rid of the peak before it So do some tests before you work too hard on it.
One thing I have been realizing is because of firing order and the layout of the manifolds and so forth, even if an engine LOOKS symmetrical, it does not necessarily RUN symmetrically anyway.....so the header does not need to be equal length and symmetrical, or all the same size.
Four cylinder thoughts:
The two middle cylinders sharing a primary....like an old five port head, is not the worst idea. A "three into one" on a four cylinder, if that makes it fit where a four into one won't...then it's a great solution. You can also have two downpipes on a cast manifold like a toyota 22r did for many years. If the point of shorty headers is to save space.... then that saves space eh?? You can also do a twin y, use a muffler with two entries. If the primaries enter separately into a volume of several liters the 180 interference is GONE. You have to put something "in the way" of the exist to make the interference travel up the other primaries, such as a length of pipe. If they enter separate into a volume then it greatly delays and dampens the pulse.
I haven't thought much about that on a v8 but probably can work along the same lines.