AutoVE commanded AFR issue

[ttls1]

Hi, I am doing autove tuning on my turbo LQ4 and have been getting good results right down to 1.00 but have noticed that for some reason the commanded AFR actually leans out as I put my foot down and increase map. I have no idea why it is doing this as I have set all the commanded fuel tables as per the tutorial. I am running COS 5. The commanded AFR gets above 16:1 and the WBO2 is matching it under high loads. I would like to work out whats going on before i start on the BoostVE

cheers

[Woodchukka]

It would be a good idea to post your tune and logs. That will help diagnose the fault.

[ttls1]

Ok, here is my SDautoVE tune. Unfortunately the logs are on a different computer so I will load later. If someone could have a quick look over this tune that would be great

Edit: Have actually noticed that the B3647 table is commanding a leaner lambda at high map. I will change it to the inverse of these figures. Is this a possible typo in the tutorial?

Cheers

[Woodchukka]

B3647 is commanding a leaner mixture as that is what you are asking of it. Your units are in EQ ratio so less than 1 is leaner and greater than 1 is richer than the value you set in B3601 (Ratio of Air to Fuel Stoichiometry). Lambda works in the opposite manner which is what your B3647 seems to be set to. So make the figures as you say the inverse of what you have now for a starting point. Someone more knowledgeable with boosted applications will be of more use than I as to what to set them to.

You PE is also only at 1 so in PE mode it is likely to command an EQ of 1. Not the best idea especially if boosted. I would change it. There is a post that says to set B3618 to the 105kpa column of B3647 however I believe that was for naturally aspirated applications so am unsure if is applies to boosted applications such as yours.

Your VE table (B0101) has a few holes so the logs you used to change it would be helpful. In addition Commanded Fuel vs RPM (Normal) (B3647) has many values set to 1. This will set you to semi open loop (SOL) so your short term fuel trims will be active. You need ensure no values are set to 1. Set them to .99, .999, 1.001, anything but 1 so they will not affect your calcs for B0101.

I have only had a quick look as have to duck out however. Just wanted to start the ball rolling. I am sure someone else will chime in for sure. Most likely more experienced than me especially with boosted applications.

[Woodchukka]

Double post.

[Wheelz]

From my reading around here, boosted PE can be as rich as .77 lambda

[ttls1]

Thanks for the replies everyone, I made the changes and its behaving correctly now. I have the main VE basically finished and will start on the boostVE. I am going to autoVE the boost ve table but was wondering what happens with the commanded afr over 105kpa? I am looking for an afr of 11.5:1 at 12psi but as the commanded afr at 105kpa is 12.8:1 must I fudge this ritchening factor using the boost VE table? If so this will affect the BEN factor during autoVE wont it? Just not too sure how you all do it. Any info will be greatly appreciated.

[darcy]

You can use B3647 to command you non-boost fuelling, above 105kPa the 105kPa commanded fuel will be maintinaed.
You can then use PE fueling, and set to active at say 120kPa, to command a richer boost fuelling; or as you have suggested, you can fudge the boost VE table to give you whatever actual AFR you like.

[joecar]

...

Edit: Have actually noticed that the B3647 table is commanding a leaner lambda at high map. I will change it to the inverse of these figures. Is this a possible typo in the tutorial?

...In the tunetool, go Edit->Properties and set Fueling units to EQR (the default is Lambda)... then restart the tunetool and redo B3647 so that it is correctly rich.

When viewing B3647 (or any other table) take a close look at the upper right of the graph display, it shows the units.

[joecar]

If you're following the AutoVE tutorial, modify it as follows:
- if you have a serial wideband, for the BEN pid use CALC.WO2BEN from the Calc.VET thread (this uses GM.EQIVRATIO x EXT.WO2LAM1);
- set B3618 to suitably rich for boost if you run boost;
- set PE MAP enable B3613 to 110 kPa (i.e. so B3618 takes care of boost) <-- what darcy said above.
- set PE TPS enable B3616 to same as 2002 Camaro;
- set B3647 higher MAP columns to suitably rich for sub-boost;

Do not set B3618 to EQR 1.00.

[joecar]

Note that Cat Over Temp Protection (COTP) commands about AFR 11.7... so you should disable it if you have no cats.

[joecar]

Adjust the Boost VE table to make your wideband fuel same as commanded fuel.

[ttls1]

Ok we did some boost tuning today and I have posted the tune and some logs. I hadn't read the last few posts before we started so it may explain a few issues we had. The main one was the commanded afr would not change above 105Kpa. I now noticed that we had b3613 set at 120 but it still made no difference after that anyway. Could someone please have a look at the tune we used and see if we missed anything. B3618 had the first cells at 1.00, what would have this caused Joe? Anyway thanks again for the advice it is very helpful. We had to change the boost afr by fudging the boost VE as the PE didn't seem to work. I know its a bit ritch in boost but we can lean it out slightly. One last question, If the afrs in boost are fudged using the boost ve will the fuel trims try to lean it back to commanded afr in boost once it is back in closed loop?
We are wanting to have a crack at the timing so any info on that would be great too.