Overshooting load targets, low wgdc - what's wrong?

[mituc]

Ok, so I have this log I attached and I'm trying to figure out what's wrong with the car of a friend of mine.
Right now the load at each RPM point are about 0.1 higher than what I've set them in the tune and the WGDC, obviously, drops towards 0 while load is still maintained.

The only thing that seems to be off is the boost, and I suspect that a possibly faulty MAP sensor not reporting the right amount of boost is preventing the boost limit logic from kicking in, because there's no way the factory turbo will flow 250+g/s of air at only 13psi at 5500rpm.

Otherwise everything looks fine, the AFRs are about where they should be, so at this point I have no reason to suspect any boost leaks.

Mods on the car (gen1 speed3):
- Cobb SRI;
- TurboXS catless DP;
- Ultimate Racing catted TP;
- Ultimate Racing Catback;
- Autotech Internals.

[sheston]

Are you sure you updated all the load limit tables. Seems there are about 12 of them. I think if you search on load limits, you should find the write ups. You have to change more tables than you think

[g00s3y]

nevermind, you aren't using an AP to tune.

[mituc]

The tune has all things right (it's primarily an OTS map from the Versatuner with the boost limits lowered to avoid load cut) and the problems I'm trying to figure out are visible in the log:
- high MAF g/s with low boost;
- still 13PSI of boost with very low WGDC after the load limits logic kicks in, but the root cause of this one is probably the first problem.

[mituc]

Ok, so we're going to clean the MAF first and see where that gets us. There's a chance that the MAF is done and the ECU fuel compensation logic manages to keep up with the fuelling somehow, but to be honest I'm inclined to blame the other two things, the MAP and/or the factory EBCS or sticky wastegate arm.

I don't see this every day so I hope you guys don't mind if I ask for a little help: @Enki, @Lex, @Justin@Freektune, @phate, @Nliiitend1, @silvapain, @Tomas.

Thanks!

[Tomas]

Originally Posted by mituc Otherwise everything looks fine, the AFRs are about where they should be, so at this point I have no reason to suspect any boost leaks. .

Are you sure your AFR looks fine? Take another look at the spikes in AFR in lines 17 and 18. That may indicate a boost leak.
Where is your go pedal during all this? All the way to the floor or do you lift off anywhere? If its all the way to down then your throtle position looks a little wonky. There is ECU intervention going on where you see the AFR spikes.

Do a boost leak check even if you think that it isn't needed.

[mituc]

Originally Posted by Tomas Are you sure your AFR looks fine? Take another look at the spikes in AFR in lines 17 and 18. That may indicate a boost leak.

At lines 16-18 there's a gear shift from 3rd to 4th. the acceleration is lifted only at line 27. Until 27 it's WOT from like line 3 or so.

[Enki]

Try setting your closed loop load limits to the same as your transition limits and see what your WOT AFR does; I'm betting it will be rich. If it is, you need a mafcal and bad.

[mituc]

Cleaned the MAF - no visible change.
The MAF cal is the same I used when I had the Cobb SRI and it was spot on. Last time we checked it was pretty much ok on this car too.

If there was any boost leak I'd rather expect the WGDC to be high rather than so low.
The closed loop load limits already are about where the transition limits are. Also, the and OL PT and WOT AFR tables match in those areas for a smooth transition. But otherwise this tune is an OTS tune pretty much (with lowered throttle reduction boost limits) and even though I don't think it's relevant I attached it to this post.

[sheston]

Have you tried reloading the map to see if anything got corrupted?

With my load tune (hybrid), it seems to load limit fairly softly. I get a little over the load I cap it at. Are you just trying to run a very safe tune? If it was me, I'd try to raise the load cap a little and see what changes

[mituc]

Originally Posted by sheston Have you tried reloading the map to see if anything got corrupted?

Well, yes, this is the 3rd revision of the map... the first revision was basically the OTS map itself so that we can get a baseline of everything and we noticed the overboosting.

[Enki]

Now that I've got some sleep in me, I can comment better on this.
It's very likely that there is actually nothing wrong with anything, for the following reasons:

1. Load is calculated via RPM and massflow (g/s), and your AFRs are spot on (per your words) with the current calibration

2. 13 PSI isn't that much; with VVT deleted from the map on my car (when it was stock turbo), hitting higher than spring pressure wasn't entirely uncommon; increasing airflow reduces boost on the stock turbo, especially at higher RPMs.

3. It is trying to close the throttle due to boost overrun, but some of the log headers don't look to match up with column contents; for example, 15 KPA is actually only 2 psi; if you are trying for 0 psi, that won't work. If you are trying for spring pressure, you should just zero all the comps, and the WGDC table and see what the actual baseline is without trying to force it to go lower.

It's actually very likely that there's either a small hole in the WGA diaphragm or the factory preload is a bit higher on this engine; either way, it's going to make more boost than a "normal" car would. Also, elevation could be a factor here, as high altitudes tend to produce higher relative boost pressures with lower load levels while very low altitudes will produce higher load numbers with relatively low boost pressure.

[mituc]

Originally Posted by Enki 3. It is trying to close the throttle due to boost overrun, but some of the log headers don't look to match up with column contents; for example, 15 KPA is actually only 2 psi; if you are trying for 0 psi, that won't work. If you are trying for spring pressure, you should just zero all the comps, and the WGDC table and see what the actual baseline is without trying to force it to go lower.

Ah, I forgot to update the table header after I converted every pressure from kPa to PSI, and replaced MAP with Boost. It's the same for the Fuel Rail Pressure, it's PSI.

So the registered boost is not that much, it's 13-ish PSI across the board. But the air flow is a bit too high for only that, and given the fact that the AFRs are in target makes me think that the MAF or MAFcal is not the problem in this case.

I think if there was a hole in the WG actuator diaphragm the problem was getting the airflow for meeting the load targets, while in this case is like indeed the preload is a little too high or the WGA arm/diaphragm are sticking.

So now that we figured my table header mess, would you consider emitting another opinion?

Thanks!

[Enki]

The question of elevation remains. Adding timing to the map should see airflow increase with boost dropping off a bit; running lower timing will actually impart more energy into the hotside of the turbo as well. Have you disabled open loop fuel trimming to see if AFRs change at all?

I guess the thing I'm trying to figure out the most is why 13 psi and 250 g/s is actually a problem, especially on ~93 octane.

[mituc]

Originally Posted by Enki I guess the thing I'm trying to figure out the most is why 13 psi and 250 g/s is actually a problem, especially on ~93 octane.

Because at that RPM, load and similar elevation (35m/110 feet, so close to sea level, 14.6-14.8PSI BARO) it usually takes about 16-17PSI and 60-70% or more WGDC to flow that much air with similar or even better flowing mods.

As for disabling the OL FTs... I'm not sure how to do that in VT. As far as I knew they were actually disabled, in my case (similar euro 3MPS BK) when I had a small boost leak the AFRs were going into mid/low 9s anywhere above like 7-8PSI of boost...

[Enki]

There's a trim limit to them, but the tables you want are called "Closed Loop Load Limit" and default to 2.5 (which is the root of the "2.5 load issue" found elsewhere on the forum).

[mituc]

Originally Posted by Enki There's a trim limit to them, but the tables you want are called "Closed Loop Load Limit" and default to 2.5 (which is the root of the "2.5 load issue" found elsewhere on the forum).

Oh, cool!
Well, those are set to 1.063 in all 3 tables, same as the Closed Loop Open Loop Load threshold. So I guess we have no trimming above that load value and that's when the ECU goes in OL.

[Enki]

Correct.

You sure this car doesn't have a super stock turbo?

[mituc]

Originally Posted by Enki You sure this car doesn't have a super stock turbo?

No, it's a factory turbo and has like 5-6 years on the car (the original one was replaced after it started smoking in traffic and the previous owner of the car was a non-smoker, so the marriage had to break).

I guess we will have to lower the BCS DC base table values a bit more and see where that gets us, maybe that will compensate for a sticky or slow reacting wastegate arm/diaphragm - if that is the problem.

I'm also considering to send my factory MAP sensor to replace his because the boost is reported too low for that air flow even at sea level at low temps.

[Enki]

Easiest way to check that is to check boost KOEO.