Thread: tuning help with supercharger, odd problem

I'm having an odd problem trying to tune my engine with the supercharger using cells that are normally never used. I'm not exactly a tuning newb but this is the first time I've built a supercharged engine and I guess its a common problem with them.

As soon as I touch the throttle (3 TPS voltage), around 20 seconds on the chart, the engine stumbles and stalls (1 RPM). Looking at the datalog, the pressure from the charger is causing the engine to peak up to a really high load cell (4 Load, 6 Airflow).

As a consequence, the igntion is way retarded (7) and it also runs very lean (5). If I feather the throttle to keep the engine running then it settles back down and revs as normal ( around 22s on the chart), the boost drops (2) and everything else runs normal.

I just can't get rid of this peak as the throttle is first opened. What do you think I should do ? The options seem to be advancing the igniton about 20 degrees or pumping in a load more fuel. The AFR is really lean even when the injector duty is up...

Any suggestions ? Sorry for the size of the picture.

I see that you're logging, but what do you have to tune with? You mention advancing the timing 20 degrees... What would you be doing this with?

Basically I'm hoping you have some sort of fuel/ignition biggy-back or stand-alone EMS. What'cha got?

Do you have a picture of your spark and fuel maps?

sorry, should have mentioned, its an Apexi PowerFc with Datalogit. I've been logging and tuning these maps for around 6 years, but never seen these load cells get used before.

The basic map I'm starting with is for a turbo, so the progression is normally diagonally down the map because of the turbo lag. Now I'm jumping straight down to high load at very low rpm, thats what the problem is.

Heres the maps, the X axis is RPM, the Y axis (P) is load and is calculated from RPM and MAF voltage.

The 'P' plot (4) from the graph above equate to vertical cells 0-20. It starts at around load cell 3 but quickly jumps down to load cell 14 when the throttle is cracked open.

Injection map - the cell values equate roughly to 14.7/Cell = AF ratio



Ignition map



Rough idea of the map tracing, the green line is the normal turbo trace, the red line is the supercharger trace.

To be clear, you're worried about it going lean, and this lean situation is causing the engine to behave less than optimal?

Timing looks pretty good, but fueling is obviously way off.

Take the values in the fuel map at 3,600 rpm from 3882-25412 load and copy them. You'll want to apply them identically from 3,200 rpm all the way back to 400 rpm. This would be my starting point. You'll notice there's a odd lean spot at 3,200 rpm and 6706 load that this will take care of too. You might still want to richen it up some more on the three leftmost columns after you do this.

You really just need to mess around with the bottom left quadrant of the map until you get rid of the lean stuff. If this doesn't work, there's something else we need to look into.

cheers, I can't decide if the stumble is because of fuel or ignition. The fuel doesn't seem to lean out till way after the rpm drops and at the same time the injector duty is way up. I'm wondering if the lean condition I'm seeing is just because of misfire ?

I was thinking it might be because the ignition is retarding from 40 down to 18 almost instantaneously which is dropping the power and making the revs drop. Only problem is if I advance it any more its going to start detting.

I'll try richening up that area of the map and see what happens. At least if it doesn't help then I have some room to advance the ignition a bit without it detting.

Cheers for the help, I'll let you know how it goes.

If the ignition is truly too retarded then you shouldn't have to worry about detonation at all. Just increase the advance in the area you think it will most likely help and you should be fine. I'd say 18 degrees advance sounds about right for high load, low rpm. But that's on a SR20DE with 9.5:1 compression. I know the GTi-R motors run high base timing (by 5 degrees) so you might consider something like 23 degrees or so. Just if you make changes, make sure to "smooth" the changes into the map and apply the changes out to other areas close-by gradually. Don't just go stick 23 in a cell and be done with it. Make the cells around it be similar. You probably know this already.

Don't be afraid to experiment. I don't think detonation is something to worry about too much at those low rpms. You could easily be super retarded, causing the loss of power. Advancing will only help the situation. (That is if timing is your real problem, and not something else of course.)

Fix the timing map with what you know. Timing that low there will cause it to run lean and to misfire.

I think its definately the timing now, I tried to richen up the bottom third of the map and it didn't make any difference, ran worse if anything.

Out of time now, but during the week I'll try advancing the ignition, it got to be the fact its jumping 20 degrees, misfiring and dropping power.