Here's the situation. I've converted my '85 2.8 v6 camaro's ecm to a '165 w/ $32B. I've copied as much of the stock v6 '302 calibration (about 80%) as I can into the '165, but I just can't nail it down. It behaves like stock, but it's down on power everywhere. I believe the problem may be related to the v6 maf sensor I'm using. I've copied all the '302 scalars and MAF tables, unchecked the 'analog maf' switch, and changed the LV8 scaling factor. What could I be missing?

 

SA? Check the data log for the BLM & INT values. For the most part they will tell you whether the fueling (MAF tables) are close. Spark timing is torque control. Not enough and the engine is weak and runs hot. Too much and the engine is rough and makes no power, but can 'feel' good.

With MAF the LV8 defines the load for SA. A smaller engine has a lower LV8 for the same load.

RBob.

 

Spark tables are stock out of the '302. BLM's stay fairly close to 128, INT flies all over the place from 128 to pegging at 75. I'm getting a maf low code, not sure if the ecm wants the maf grounded at the ecm or the body, tried it both ways and didn't seem to make a difference. Another thing I forgot to note is that the tcc never locks, is this related to the MAF scalars or the LV8 scaling factor?

 

A MAF code will cause the ECM to revert to n-alpha mode. This can also happen before the code or SES is even set.

The MAF input should be on B6. Which is different then the analog MAF normally used on the 8's.

RBob.

 

Hmm, that's interesting. I wasn't sure if it was in safety mode or not, wasn't sure what to expect from this ecm. The datalogs show it's reading the maf even though the signal's on pin b12, I'll hook it up to b6 and see what happens.

 

But is the ECM reading the MAF? Or, is the ALDL reporting the n-alpha generated airflow value?

RBob.

 

Good question, wish I could tell but my guess is that it was reading alpha-n. Not that it matters anymore since I switched it to pin b6 ,which was the correct pin. No more maf low code. But now I have a new problem...

I F'ING HATE OLD COMPUTERS!!!

-My laptop took a crap. Now it stalls in closed loop when I come to a stop and I don't have a way to datalog it to figure out what's going on. I almoast chucked the laptop out the window while I was driving, but since it's snowing I figured I'd better not end up in a ditch.

O.K. time to cool down, I'm goanna go sit in the snow for a while...

 

Got the laptop going again. It only stalls below 5mph with the wheel cranked. I've got the power steering switch hooked up at b9 and the 'use psps' switch checked, not sure if it's working though. Idle eventually corrects, but it takes ~1/2 second or so to respond which makes me think the stay alive function is what's doing the work. Good news is it feels up to stock power levels. Does the digital maf have a burnoff period, and if not will it cause an issue?

 

The PSPS logic may be cal'd out. I don't believe the TPI cars used it.

IIRC only the hot wire MAFs had a burn off. I think yours is a film MAF, which wouldn't have a burn off cycle.

RBob.

 

That's what I thought, good thing I set the burnoff period to 0. Haven't got a code for it yet so should be fine. So any idea what the problem is with the tcc? I think it's load related, I'll play with the enable values to see what's going on.

 

Alright, starting to get this dialed in. It's hesitating at throttle tip-in, probably related to the fact that there's no throttle follower function in $32b, need to tweak the "pump shot" settings to get rid of that. TCC is locking, had to set the 'road speed coast' and 'min load limit hysterisis' to 0 to get it to lock. Slight chugging at low throttle with the tcc locked, noticed that with the original ecm but I'm going to work on it anyway. Still has a slight hesitation above ~60% throttle (before pe turns on), not sure what to do with that. It will run off specified pw tables but I had to cut the original values in half. Stock calibration had ridiculously long pw's, the highest was 20 msec's or so!

 

Oh, there's no power steering logic either. Fixed that by raising the stall saver settings to 500rpm on, 600 rpm off. Chugs a bit with the wheel fully cranked, but atleast it won't die. Keep in mind that I live in Colorado so these issues might not effect anyone living at lower altitude where the air is denser, thus making more torque.

 

You can run more idle spark advance to help the stalling/low idle, and especially so at higher altitudes.

There's no real baro compensation in the 32b and 6e, as far as I know/remember. Typically, I see more IAC is needed, even for cranking, even though it gets less air (a pressure differential issue).

Torque converter lockup may have an issue with the smaller engine, high alt, and the corresponding LV8 thresholds. Reduce them to whatever you find as your free rev LV8, and TPS. Hysteresis is just a way to prevent quick on and off and on and off. You should keep a few bits of hysteresis in it.

You did change the cylinder number, right?

 

Yep, changed the cyl # and the fueling mode . I've alredy added some spark around idle, gm put a little 'dip' across 600 rpm for some reason. Another one of those things that don't seem to make much sense to anyone but the gm techs, so I took it out. I've messed around with the "pump shot" settings a little and reduced the tip-in stall a little, but haven't been able to get rid of it. I took out the low pw offset just to see what would happen and tip-in stall was considerably better at low speed which makes me think it's still rich at low load. I noticed a little bit of idle hunting when I started it up too. What exactly does the low pw offset table do anyway?

 

I've tried to get rid of this tip-in stall but it seems like nomatter what I try I can't get rid of it. I completely took out the "pump shot" and it still stalls. I took out the accel enrich offset and it stalled. I lowered the bpw and it still stalled. So this must not be a fueling problem. I lowered the 'min delta lv8 for accel enrich' value thinking it's not responding quick enough and that didn't seem to have an effect either. Suggestions?

 

Tip in stall is 99% of the time LEAN. You've successfully removed all the fuel that is intended to help that. Put'r back in.

 

Funny thing is I already tried richening it. I doubled the "pumpshot" and that didn't help either. Maybe it just isn't long enough.

 

On a MAF engine (and no MAP) the MAF tables contribute a lot to driveability, tip in, over-run, etc. How close have you tuned in the MAF tables? And, if the MAF de-screened that doesn't help. You may need to play with the MAF tables to get the tip in correct.

RBob.

 

Thanks again for all the help. The maf still has the screen and the tables are stock so they should be pretty accurate. I took a while to dig into the old threads and found this 'MAF 6E Tip-in/AE (Stealth Ram)' thread. I followed what was suggested and the tip-in stall is much better, only noticeable from a stop now. It needed a pretty big adjustment to all AE related tables. Once I get the part throttle stall worked out this swap should be ready for everyone.

 

I've been leaning it out quite a bit recently. I haven't run into detonation yet, infact it still smells rich off idle but I'm starting to wonder just how short a pulsewidth I may need. Right now it's idling at ~1.5 msec and it's still showing 110 Int (blm's locked). Anywhere above 5% throttle and it's dead rich. I have the max pw set at 3msec and it still shows up rich! Egr, CCP and pe are all off. Injectors are good, haven't verified psi yet. No stall at part throttle but this doesn't seem right.

 

Could be a couple of things:

Injectors firing at the wrong rate (o-scope or DVM that measure freq)

Incorrect injector constant

$32 mask/code does the MAF to airflow calculation differently

Some CCP system are NO, which means the line needs to be pulled and blocked. Although after it has been purged it usually drives the BLM up.

Maybe unlock the BLM and see how far it is off. It may just need some slight tweaks to the MAF tables, or it may be way off.

RBob.

 

Does it go dead rich if you're driving it and go above 5% throttle? If not, then I'd guess the purge is open.

If the purge canister gets full, like from very long idle, not driving in forever, or even filling the tank too full, the canister gets very full. Even a slight amount of purge opening can let the engine run with no injection at all, if the load is low enough, and the throttle is cracked open enough to uncover the port.

Best bet is to block/pinch the hose to test this.

Also, if you accidentally switched the purge line with the return line, the pressure in the tank will push fuel up that new purge line, and fill the canister with liquid fuel, overflowing it, making a huge mess and hazard. Don't ask... wasn't my fault... but I did know why, almost immediately, because it followed a fuel pump change. The return line is supposed to be the submerged line in the tank.

 

Just had another thought on this. The $32 code can create the injector PW via 2 different methods: first is a table lookup, the second by a claculation via the airflow value (directly from the MAF).

For the PW table it uses the LV8 variable. Which is calculated from airflow, RPM & a scalar. If the scalar is incorrect, then the whole table is off. And, the SA table is also incorrect (along with a bunch of other stuff).

Even with the correct LV8 scalar the PW table probably needs tweaking.

There is an option bit to select between the two methods of PW creation.

RBob.

 

Haven't had a chance to mess with it much lately, it's just me and the head mechanic at the shop now and it's the busiest shop in the state... during the busiest time of the year... and he's a nut ! LOL.

Only goes rich above 5% throttle. I'll plug the line and get some readings soon. Right now I have it doing the calc'ing it's self. I put the v6 load scaling factor in and it ran like crap so I went back to the tpi scaler and it runs much better (aside from being rich and the odd pw).

 

Turns out the stock pulse width table I disregarded as ridiculously long runs well with the v6 scaling factor. Not sure how the scaling factor had such a huge effect, but it's running leaner just about everywhere than with the short pw table I came up with using the tpi scaling factor. I plugged the ccp and it leaned out a little also. What puzzles me is that the idle pw on the table is nearly 5msec, but I have 128-132 blm's at idle! Enough guessing with the stock nb, I'm getting a wb so I can see just how far off it really is.