There is quite a sharp jump in the manifold pressure when you put your foot down. It looks like the system just isn't responding quick enough to the change in fuelling demand. Can you get the MAF's output into your datalogger? If you look at when you back off the throttle there is also a smaller rich spike, the change in manifold pressure seems less agressive when you back off the throttle though. I guess that is you controlling the throttle so the car doesn't lurch.
Engine build progress
- Thread starter TTM
- Start date
Where are you measuring boost from Thom? I am surprised there isn't an initial drop in boost pressure when your floor the throttle? I would have thought it would take some time for the turbo too speed up to match the ammount of air the engine is sucking and this delay would cause an initial drop in manifold pressure.
TTM
Well-known member
Tom, boost is measured from the intake manifold, from the line that goes to the KLR. Let's say "turbo lag" is the "ramp" on the white curve (after the first sharp "step"), at the end of which peak pressure is reached. Could we say that the little sharp "step" is an "excess" of pressure provided by the turbo that translates into the lean spike after the mixture has reached the lambda probe down under in the exhaust pipe? My thinking is, if the boost was increasing in a more linear way, without this sudden change of shape, then perhaps there might not be that lean spike?
Here is my current theory: I think the initial spike must represent the pre and post throttle pressure balancing when the throttle is opened. Before that I guess the pressure pre throttle is much higher than post throttle. This is a near instant and large increase in manifold pressure. Look how quick it happens compared to how long it takes to get full boost. This will cause a rapid, near instant increase in the ammount of air drawn into the cylinders. There will be some delay before the MAF sees this change, normally the delay probably isn't an issue but I think in this case the pressure rise is so sharp it causes such a quick increase in the ammount of air sucked by the engine that the effects are quite apparent. Most systems would have a way to cope with this, old carburettors used to have a feature to squirt extra fuel into the manifold when the throttle was depressed quickly. The system on the standard 944T might not cope with this, it does switch instantly to the WOT maps but at those revs it wouldn't expect such a high pressure so may still run lean. Normally at those RPM at full throttle there would be a good vacuum in the manifold, say if you had floored it from idle. In fact I am sure I experience similar in my turbo, especially in low gears at low rpm. Now I tend to control the throttle to eliminate it I think. Now I think of it I'm sure in low gears at low rpms the engine will stutter slightly if I floor it. I have no idea how the vitesse system deals with this. I wouldnt expect it to use a WOT map as the whole thing about the MAF is that it is adaptive to changing boost levels etc which would change the WOT map. Perhaps there is some feature programmed in to do the equivalent of what carburettors do when a sharp increase in throttle is detected? Edit: Perhaps the issue is that there is no such feature?
DivineE
New member
This is why I'm so interested in finding a solution to this. Every highly tuned 944t I've been in has a slight cough as it comes on boost. Always somewhere between 2000-3000rpm (depending on the car). I want to know why no one can map this out? In fact thinking back they do this no matter whether on the standard AFM or MAF so it can't be the vittesse software
If that's the case then I don't think you can map it out. If the DME in the 944 simply isn't designed to do it then it must be re-programmed (not re-mapped) to do so. You would need the DME to detect when the throttle is 'floored' and increase the injector pulse width briefly to prevent the lean spike. Essentially this happens so quickly the DME needs to pre-empt the change and dump a load of extra fuel into the manifold. I'm sure this requires using the TPS as even a MAP sensor is essentially looking at the past. Edit: Is the TPS on the 944T just two switches, one for idle one for WOT or is it a 0-5v output? Thom certainly seems to have the exact throttle position in his datalogger but perhaps its aftermarket TPS?
Barks: 951 TPS is a potentiometer which has a voltage scale from 0 to 5V. Sorry to hear the fuel pump didn't help any Thom. I also saw your post on Rennlist. Let's hope someone can figure it out. I am running out of ideas. I can't do any testing myself until snow melts away. When it does I will test with Vitesse injectors so I can rule at least them out..
What we need to know is if the injector pulse width is increased when the throttle is floored. My guess at the moment is that the engine is sucking more air but the DME isn't telling the injectors to deliver more fuel as initially theres no change in MAF output and the system isn't programmed to change fuelling based on a quick change in throttle position.
TTM
Well-known member
There is the possibility to ponderate the fueling with the signal from the MAP sensor - I haven't done much testing yet but apparently it can measure vacuum as well as pressure. Might be an alternative. But surely, apart from Diver944, there must be other people using super responsive turbos and Vitesse equipment that must have gone through the same issue...
.
TTM
Well-known member
Very well, it's easy to use. The latest version has two extra input lines (mine has one) so you can log two extra signals in addition to the usual variables. I have the one with the wide screen which is I think the one that offers the best "at-a-glance" visibility. A friend has the round gauge, looks neat but it's too small I think.
FYI I did some experimenting in my turbo regarding your lean spike issue. When blipping the throttle on my car it spikes rich rather than lean, sometimes as low as 10 AFR. I am almost leaning back to thinking its fuel rail pressure rather than to do with the fuel injectors. Maybe you could use a pressure transducer on the end of the fuel rail to log the rail pressure with the zeitronix as well.
Posts made and opinions expressed are those of the individual forum members
Use of the Forum is subject to the Terms and Conditions
