Electronic DV Control

[Ralph]

Being a road circuit enthusiast I spend a lot of time at partial throttle. It took me many tries to find a DV valve for my Evo8 that worked for me (happily settled on GFB in the end), but I still have this never ending quest for better partial throttle behavior. All the DV I've tried use a pressure signal from the intake manifold as a control signal to determine when/if the DV should be opened. This works fine for fully open or fully closed throttle, but at partial throttle there is enough pressure drop across the (partially open) throttle plate to crack the DV open and loose boost pressure when I don't want to.

The way I understand it (which could be entirely wrong ...) the DV should only open when the throttle is rapidly closed to prevent air being forced backwards into the turbo cold-side, causing bad things like compressor stall/surge, strain on the turbo, unnecessary spool down, etc. So it sounds to me like the rate of throttle closing is what should really control the DV, not the difference in pressure between the IC piping and the manifold. If I close the throttle slowly, there's no need for the DV to open; same if I'm at steady partial throttle balancing the car through a turn.

Pretty much all cars have electronic throttle position sensors these days, so is there a manufacturer that offers a DV directly controlled by the TPS signal? It could be the definite solution to partial throttle flutter ...

[GoFastBits]

Hi Ralph,
Great question - partial throttle DV behaviour is an interesting subject, and one that does particularly affect EVO racers.

What I would like to determine is exactly what your desired outcome is - do you want to get rid of part throttle flutter, or hold boost in the pipes for better response?

The reason I ask is because these two outcomes are polar opposite in regards to what the valve needs to do.

If flutter occurs during any particular set of conditions, it means there is not enough airflow through the turbo compressor for the amount of boost pressure, pushing it into surge. The only way to eliminate the flutter under these conditions is to reduce pressure, increase airflow, or both. A DV does both - opening to provide a relief path increases airflow and drops pressure to avoid surge.

Contrary to this, improving throttle response is all about keeping as much pressure in the intercooler as possible. However, this can sometimes result in low RPM or partial throttle surge, since the valve is trying to stay closed to hold pressure in the intercooler.

If you lift off slowly or partially, you'll likely find that the valve actually needs to be opened MORE in order to prevent surge than if you lift off quickly and completely. This is because during a partial lift-off, there is still some exhaust gas spinning the turbine, but the reduced airflow through the throttle pushes the turbo into surge more readily.

A common problem with the EVO 8 & 9 is the Karmen vortex airflow meter is sensitive to surge, particularly at high boost. Pulsing in the intake can cause the car to "buck" during a partial lift-off due to erroneous air flow readings.

So if you are currently experiencing surge during partial or slow-lift offs, the only way to get rid of it is open the valve more by softening the spring pre-load. If your GFB valve isn't already fitted with a soft spring, that's something you might try. The factory valve works like a "blanket" solution for surge, in that it typically opens fully regardless of whether or not it needs to. But this is counter-productive in regards to throttle response.

In short, there are no manufacturers that offer electronic DV control. There are a couple of technical issues that would need to be addressed to get such a system to work, but there's no major reasons why it wouldn't work. The cost involved of such a system however would be a very large factor, and the gains vs cost questionable.

I hope this helps!

Best regards
GFB Pete

[Ralph]

Hi Pete, thanks for the detailed response! You're absolutely right about the Evo wanting to "buck" during throttle lift due to the MAF setup, especially if it's cold outside and the DV isn't fully recirculated. Both will make things worse, but that's why I like the adjustable GFB valve so much - in the winter I can set it to full recirc to get smooth driving, and when it gets warmer I can VTA a bit to get my sound effects back .

The main reason I got thinking about TPS driven DV control is partial throttle flutter (the lift-off bucking is just something I've learned to work around by lifting completely and then getting back on it). I may be using the wrong terminology though: what's happening is (I think) that at steady partial throttle and positive boost, there is enough pressure differential across the throttle plate to drop the pressure in the DV signal line, and so it opens. That causes the boost to drop, which causes the pressure difference to drop, the DV closes, and boost builds up again, and then the cycle repeats. This happens very quickly (4-5 times a second maybe). So I'm thinking it's the DV that is fluttering, and not the turbo surging - although admittedly this is based solely on what I can hear, since I don't have separate pressure sensors for the manifold and the IC piping (upstream of the throttle).

While thinking about this, I came across this article on electronic DV control, and that made me curious about the whole thing:
http://autospeed.com/cms/A_2188/article.html