MAF Tuning Problem

[1963novawj]

Hi guys,

I'm using the tuning school's GM beginner book to tune my nova with a 5.3 (cammed with 224/224 110 LSA). I tuned the VE table and my LTFT were a respectable -3 to 0. I moved on to MAF tuning and it's been a bumpy ride so far. During testing and datalogging the car runs okay in gear but upon return to idle it often stumbles (my wideband shows a very rich condition ~ 10.0) and sometimes dies. I'm attaching my tune file and log file for my last test. thanks in advance for the help.

Wes Stock 63 Nova MAF ONLY Cam 1 Tune 14.1 gs VE tune smoothed MAF tune 2.hptmaf tune 2.hpl

[jimmyblue]

I always like to ask what makes you think the MAF is
inaccurate. It's probably more accurate than fuel trims
at low RPM with a high overlap cam (and LSU4 wideband
meters suffer the same from exhaust stream residual O2).

If the MAF is virgin, in a straight tube and not right up
against the TB face (so you have neither bend airflow
biasing, or manifold reversion to mess with the calibration)
I'd put this messing-with at the back of the list.

Running rich (in actuality, as opposed to by the meter)
will make idle response to IAC / DBW TB step-air soggy
and the stock loop PID params can't deal with it. A
stoich meter reading with a high overlap cam will show
false lean, and "baking that into" a new MAF (or VE,
or worse yet, both) table will put you rich, badly
perhaps.

[1963novawj]

Thanks for the response. I'm not sure about the accuracy of the MAF but i do know the car was much happier running with no MAF. I have no doubt that this is probably user error as I'm a novice. Any suggestions as how i might address the issue?

[dermotw]

The MAF is much more critical than the VE, and, assuming you want to use a MAF, its much more important as well. A minor MAF error will turn it into a donkey, but the VE can be all over the place and it'll still start & run...

If you read the actual info on how these ECU's operate the MAF is the DOMINANT part in all fuel calcs, both high AND low speed. The VE is only there to improve MAF behaviour in transients (and as a failure backup of course). I don't know why you hardly ever see this written down!

Your MAF curve is at least the right shape, but it looked too rich low down to me. You can tune it in the same way as VE basically, (I assume you have created a suitable table etc.) and copy/paste "x50%" etc., but at idle areas its hard to 'hit' the cells, but for example by using the scanner controls to deliberately set idle speeds higher and lower to capture enough data, but even then, the very bottom few values you would likely have to guess, err I mean extrapolate.

[jimmyblue]

The problem lies in how you are going to "adjust" it. What
do you think you know, how do you know your reference
instrument is more trustworthy than the thing you're
trying to calibrate? Neither NBO2 nor WBO2 readings can
be trusted at idle with any kind of cam overlap. And if
you've got long tubes forget using STFTs to tune idle
mixture by MAF or VE tables.

"Never try to teach a pig to sing. It wastes your time
and annoys the pig" - Somebody or other

[IARLLC]

"Never try to teach a pig to sing. It wastes your time
and annoys the pig" - Somebody or other

Does this just apply to rich pigs????? or lean pigs as well?-)

No, really. Good point on the reference instrument or sensor question. Even good sensors have a bit of variety to them too. I have a few MAF laying around that on OEM applications fell outside of the (P0101) lines but still work. In my pre-HPTuner days it was interesting to compare one MAF to another on a Tech2.

[dermotw]

Amen to both the above ^^

Below e.g. 1200 rpm or so (varies..) you are kind of on your own anyway, maybe you will get sensible data, up to a point, maybe not!

Its also "interesting" to compare any kind of logged MAF data with an OE file in detail. My experience is that the OE ones are far more precise. More precise than I believe we could ever get by just logging... Possibly GM used log data and then massaged the curves, because they are absolutely precise...

[1963novawj]

Thanks guys. Whether or not the pigs are lean or rich they're flying way above my head .Interesting thing is it seems to return to idle after moderate throttle before it gets up to temp. Once I'm above 200 degrees it has trouble finding the target idle speed. It basically goes to idle then falls down to 500rpm and quickly lowers then dies.

[dermotw]

Once you have changed cam etc. you likely will have to massage various idle/startup tables as well, not only MAF/VE...

In your log, the IAC is stuck on 310? It should be moving and somewhere quite low, under say 80... Also, the idle spark is all over the place, which is a sign usually that the setup needs adjusting somewhere, it MIGHT be only the actual idle spark control, but just as likely its idle setup. In a good setup the spark at idle will only vary say 4 degrees or so. You can test whats happening roughly by using the scanner controls to 'lock' the idle spark to some value (e.g. 22 degrees or whatever) and see what happens during idle. If your tables are close, then nothing much happens. If they aren't it either jumps up in RPM or dies (or close to it)..

[1963novawj]

I don't have a IAC as it's DBW. I'll look at the other idle setup stuff and report back.