OLSD fueling: 30% off then spot on between runs without any changes!?!

[Pekka_Perkeles]

This is one of the strangest things I've seen. For some reason, this car needs 30% more fuel occasionally and I haven't find a reason why.

To put shortly, when I tune the car in dyno everything works. Then, let's say after 20 write calibration only -changes, the car does not start. I reflash once or twice, and then it starts again. I already got used to it last year, but now it's getting worse.

As an example, I was able to nail down idle and low-speed driving at dyno. It was just about perfect, considering the engine is heavily modified with F2, boost-referenced fuel system, 160 lb injectors and so on. Of course the tune is heavily scaled.

Now, after dyno tune I flashed the final changes and drove the car out from the dyno. And it did stall as soon as I hit the pedal. Ok, there was perhaps 15-30 minutes while the car was unstrapped. So engine was bit cooler, but not much.

To get the car running, I had to add 30% into the VE table. Sounds crazy, but that's the way it behaves.

Now the funny part. Three hours later, I started the engine and it was running very rich, so I flashed the tune from the dyno session. And it worked perfectly. Now, RAF may not be fully dialed in at cold temps, but anyway, why an earth it needed 30% more VE after leaving a dyno. And not anymore after three hours?

Next day when driving with original dyno tune, everything is perfect.

Day after and after driving and having a 10 minute break and engine shut off, as soon as I start the engine, it needs 30% more VE.

I have hard time to understand why all this is happening. I mean, fuel pressure stays the same and other things should as well. But for some strange reason, there's this 30% change. Last year it only did it occasionally after flashing, now it seems that you don't need flash, just wait for a next day to see what it wants.

Attached is a dyno tune and about 30% added VE tune to get the car running when it decides it needs more.

One final comment. This is a -98 Vette with 2001 PCM. VIN code has not been changed.

Please help.

[jimmyblue]

Are you seeing large swings in IAT? That goes directly
to air mass. And of course there's the whole question
of just which "air" you really want the temperature, of -
NA it's pretty simple, boosted I have no idea what if any
location makes actual sense.

I guess question #1 is, do you see the same air mass
for the same conditions (RPM, MAP, boost pressure)
across this difference, or is air mass moving big time
for mysterious reasons?

[Ben Charles]

Bet its the IAT, hate OLSD

[blownbluez06]

Pekka, are you logging the commanded AFR? Is it changing? If it's changing, these guys must be right. I see you changed your stoich to compensate. I don't recommend doing that unless you change your PE and BE values because it'll throw your commanded WOT fueling off. Change the VE instead. I also can't believe you haven't changed PCM's. You got a lot of money tied up in the goods. Too much to have a slow, weak 98 PCM in that car.

[Pekka_Perkeles]

Pekka, are you logging the commanded AFR? Is it changing? If it's changing, these guys must be right. I see you changed your stoich to compensate. I don't recommend doing that unless you change your PE and BE values because it'll throw your commanded WOT fueling off. Change the VE instead. I also can't believe you haven't changed PCM's. You got a lot of money tied up in the goods. Too much to have a slow, weak 98 PCM in that car.

Bret, it has a 2001 PCM.

Yes, I agree, changing VE is way to go. But while tuning VE, I want to avoid going too lean and as I'm tuning this car based on AFR error and not trims, it's not a problem if commanded is too rich at WOT in those first pulls. Anyway, I like to use lower stoich mostly during non-WOT tuning.

[Pekka_Perkeles]

Are you seeing large swings in IAT? That goes directly
to air mass. And of course there's the whole question
of just which "air" you really want the temperature, of -
NA it's pretty simple, boosted I have no idea what if any
location makes actual sense.

I guess question #1 is, do you see the same air mass
for the same conditions (RPM, MAP, boost pressure)
across this difference, or is air mass moving big time
for mysterious reasons?

I don't have enough logs yet to prove anything, but at least I have something, albeit just screenshots from the owner.

Now, isn't there a direct relationship with Cylinder Airmass and Dynamic Airflow. I mean, you can calculate one from another, if you know number of cylinders and so on? Too bad I don't have Greg's books now with me.

Below is two screen shots with first one having correct AFR (or close enough) and the later with the same tune, it goes lean. Both are taken when the car starts moving. They are not identical cases, but perhaps close enough. IAT's are quite close to each other (36c is 97 fahrenheit and 38c is 100 fahrenheit). What strikes me is that in both cases Cylinder Airmass is identical (0.19 g/cyl), but Dynamic airflow is considerably different (20.29 g/sec vs. 13.54 g/sec). How is this possible?

[jimmyblue]

The dynamic airflow seems to follow the RPM difference,
straight up. g/cyl is no different however, indicating no
significant difference in motor efficiency (VE) across a
50% rise in RPM - that to me seems unrealistic especially
with any kind of decent cam (VE table slope w/ RPM
ought to be significant as you come up out of overlap
affected low end).

Now since the motor really -should- have a VE table slope
and there evidently is none, maybe that accounts for the
fueling swing?

[Pekka_Perkeles]

The dynamic airflow seems to follow the RPM difference,
straight up. g/cyl is no different however, indicating no
significant difference in motor efficiency (VE) across a
50% rise in RPM - that to me seems unrealistic especially
with any kind of decent cam (VE table slope w/ RPM
ought to be significant as you come up out of overlap
affected low end).

Now since the motor really -should- have a VE table slope
and there evidently is none, maybe that accounts for the
fueling swing?

Yes, I agree with that. The VE is not fully dialed and I think that in dyno I didn't touch yet the 80 kpa range near idle regions. That I think explains the lean condition above.

What I still do not understand is that for example at idling 900 rpm and at 55 kpa cell and engine fully warmed, it either wants 53 as a value in VE cell, or after re-start it wants 76 as value, and vice versa. That's a lot of difference, although the real value in both cases is probably calculated between values in 800 rpm and 1200 rpm cells. In fact it would be either 58 or 82 based on 900 rpm idle. It means 40% increase. That's quite a lot I would say.

I'll try to have comparable logs so that it's easier to see and find out what is changing and what is not. I would also like to thank everyone so far. I don't think I can solve this by myself. PayPal is also an option, just PM me.

[Pekka_Perkeles]

Well, it seems that the problem has been found.

It is related to fuel injectors and how they behave. If you ever face similar problems, just PM me.

Anyway, thanks for everybody involved.

[blownbluez06]

Oh crap! what fuel are you running? ethanol or methanol content?

[Pekka_Perkeles]

No ethanol or methanol, just pure gas.

Edit: actually there is a bit ethanol in the base high octane gasoline here in Finland: 5% or less. Should not be a problem.

[jimmyblue]

So what about injector behavior changes? Or, is it just
that the reponse to something else puts them onto min
PW and you get the fuel shot you get, and it happens
to be 30% more than you wanted?

[Pekka_Perkeles]

So what about injector behavior changes? Or, is it just
that the reponse to something else puts them onto min
PW and you get the fuel shot you get, and it happens
to be 30% more than you wanted?

I still don't have evidence, but what I've heard from local colleagues is that with identical Injector Pulsewidth, the fueling changes drastically, i.e. just like I have described in this thread. So in other words, I don't change the tune, but fueling changes. All this happens typically after the car has been standing still for perhaps half an hour after driving and then it needs 30% or so more fueling. And with cold engine, it works perfectly with the original tune all the way up to the normal temperatures. Then when you let it cool for 15 minutes, it's again out of fuel.