Does the turbo downpipe _length_ matter?

[ptuomov]

Here's a quick question for anyone who's tested something like this on the dyno. Does the turbo downpipe length matter for power from the pulse tuning perspective?

I'm specifically interested in a twin-turbo cross-plane V8, but by my logic if the downpipe length matters for any other setting then it'll matter in that setting as well. So any data from any engine on the turbo downpipe length making any difference (or not) would be greatly appreaciated.

Why would the turbo downpipe length matter? Because it's somewhat analogous to the collector length in a normally aspirated engine. When the exhaust pulse makes it to the end of the downpipe, it'll reflect back an inverted pulse to the turbine outlet. The turbine runs on instantaneous pressure differentials and if the rarefaction wave makes it to the turbine at the right time, it can make a difference. In particular, it may be beneficial at low rpms (before the turbine has spooled) to time the rarefaction wave at the turbine outlet to coincide with the high-pressure wave at the turbine inlet, possibly spooling the turbine earlier. At high rpms, it may be beneficial to time the rarefaction wave to arrive at the time when there's low pressure at the turbine inlet, smoothing out the pressure differential over the turbine and possibly increasing the turbine flow capacity. I'm speculating.

Why would the turbo downpipe length not matter? it may be that the turbine extracts the energy in the pressure waves caused by the exhaust blowdown events so well that there's not much left coming out of the turbine outlet. The above considerations may or may not be in principle correct, but the turbine kills the waves and the magnitudes are so small that the above turbo downpipe length considerations amount to mental masturbation only.

[ptuomov]

Although this is no proper substitute for an actual test, I run some simulations with Vannik's software. The simulation results say that the downpipe length makes very little difference to the simulated pressure traces at the exhaust ports. The waves are almost completely broken up by the turbine. The wastegate pipe length has some impact on the pressure waves in the wastegate pipe, but the diameter and the overall energy in that pipe is relatively small, so it doesn't make much difference to the exhaust port conditions.

This is of course all dependent on the turbine model, so if someone has actually tested different downpipe lengths on a dyno I'd be most interested in hearing about the results. And since the car's exhaust acts as a necromancer (the exhaust is so loud that if you drive by a grave yard the dead wake up) at least some sound waves make it thru the pipes...

For context, I am trying to specifically lower the exhaust port pressure of cylinder #1 during it's overlap, as cylinder #3's exhaust blowdown pulse arrives there at just the worst possible time in the 6000-7000rpm range. I was hoping that the just right length turbo downpipe and/or wastegate pipe could send some low-pressure relief there, but looks like it's just "a pipe dream".

[RevTheory]

I was thinking the most important thing after the turbo is just keeping it as unrestricted as possible for the reasons you saw in your software simulation but I held back on posting because I can't back it up with experience.

[ptuomov]

I was thinking the most important thing after the turbo is just keeping it as unrestricted as possible for the reasons you saw in your software simulation but I held back on posting because I can't back it up with experience.

The simulation relies heavily on the turbine model, so it's not guaranteed to be correct. Still, at this point conventional wisdom, back of the envelope logic, and the simulation all point towards the pulse tuning being irrelevant in turbo-back exhaust, at least if only relatively small fraction of the exhaust gas is bypassed thru the wastegate circuit. It's just noise level vs. flow restriction tradeoff for me then, unless someone has some experimental evidence to counter that.

[Circlotron]

What might be interesting to try is one of those trumpets shaped things on the exhaust opening of the turbine housing, or at least where the downpipe meets the atmosphere. As the exhaust gases travel through a widening cross section they are forced to slow down, and this will extract kinetic energy from it's moving mass. I would think that would send a continual negative pressure back toward the turbine exhaust outlet. Might be worth looking at.

[Circlotron]

^^ Related information >> http://www.frenchriverland.com/white_hy ... gainer.htm

[user-23911]

The way I see it is that everything affects everything else............within reason.

Getting off topic slightly but when the exhaust comes out of the turbine, it's spinning very fast. If you keep the pipe diameter big for longer (same size as exit) then put vanes inside the pipe to straighten the flow, then gradually reduce the diameter.........you might see some unexpected gains.
Likewise on the inlet side of the compressor. the air is spinning in the same direction as the compressor wheel. Before it enters the compressor.
If you fit vanes inside the pipe before the compressor, you can straighten the flow or even spin it in the opposite direction.
That way you might even see more unexpected gains.......all very easy and very cheap.
More boost and less lag.

[Kevin Johnson]

I thought that effect was recently mentioned with respect to a cat (honeycomb flow straightener).

[ptuomov]

I was specifically interested in pulses and pulse tuning, which if relevant would influence the need for tuned lengths.

I think the overall flow capacity of the exhaust currently in the car is good. Here's the turbo downpipe and here are the tail pipes:

P1030699.JPG

SilentNight.jpg

The cycle average back pressure at that turbo downpipe measured with a slow sensor is less than 1psi. So I'm thinking I'm ok as far as just getting flow out of the turbine.

However, the current system is a little loud, and we're working on quieting it down with cross-overs and whatnot. I just wanted to know if where the cross-overs etc. termination boxes will affect power from the pulse tuning perspective, or do I just have to care about steady flow capacity and sound suppression while happily ignoring all pulse tuning issues. Sounds like (pun intended) I can ignore pulse tuning and focusing on just high steady flow and low noise.

[ptuomov]

The way I see it is that everything affects everything else............within reason.

Getting off topic slightly but when the exhaust comes out of the turbine, it's spinning very fast. If you keep the pipe diameter big for longer (same size as exit) then put vanes inside the pipe to straighten the flow, then gradually reduce the diameter.........you might see some unexpected gains.
Likewise on the inlet side of the compressor. the air is spinning in the same direction as the compressor wheel. Before it enters the compressor.
If you fit vanes inside the pipe before the compressor, you can straighten the flow or even spin it in the opposite direction.
That way you might even see more unexpected gains.......all very easy and very cheap.
More boost and less lag.

Well everything affects everything in principle, but does the downpipe length have any significant pulse tuning effect on anything upstream of the turbos? That's the question.

When you say reduce the diameter, you mean increase the diameter? In any case, given the turbine housing, the down pipe flow path goes to 3" diameter early.

We looked into the vanes on the inlet side. The impeller angle and the compressor inlet diameters are designed to work well without vanes. The speed at which air flow hits the impeller and the angle at which it hits is a design consideration in these compressors. The advice we got is that as long as the compressor inlet can be fed equally from all sides the vanes are not necessary or beneficial. The new inlet design is a bulb that allows the compressor to suck from all sides, whether it's perfect or not is another question. On the turbine outlet side, since we're dumping into a round pipe I don't immediately see why any flow straightening would be that beneficial. There's turn shortly after so there some vanes might help but I'd guess the gains would be minimal as these are gentle radius mandrel bent pipe. Plus the cycle average pressure in that pipe is low, so it's flowing well.

[wyrmrider]

I do not know if this has been mentioned but I always liked to come off the turbo as straight as possible for as long as possible (which sometimes is not very far)
sometimes expanding helps or going to a larger pipe when you bend

[ptuomov]

I do not know if this has been mentioned but I always liked to come off the turbo as straight as possible for as long as possible (which sometimes is not very far) sometimes expanding helps or going to a larger pipe when you bend

Straight pipe expanding gently to the largest possible diameter should work the best. I feel I am in a "good enough" situation there because the cycle turbine outlet pressure is low. I was just interested in whether the length matters to pulse tuning or does the turbo act like a sort of event horizon that no pulses penetrate. Seems like the latter.

[Erland Cox]

I believe that the down pipe should taper from little over turbine size to the required exhaust size with a taper no greater than 7 degrees included angle.
From testing on 4-2-1 headers on 4 cylinder engines I found no difference in changing length in the last pipe.
Seems like the pulses were already used up there and they should be so even more after a turbo.
But pressure recovery after the turbine will help, dumping into a big pipe creates back pressure.

Erland

[RevTheory]

That thing looks like it's going to be a rocket

[ptuomov]

That thing looks like it's going to be a rocket

Hoping so. It was already fast with 711 hp at the wheels, and hoping for more in this version.

IMG_3615.JPG

IMG_3616.JPG