I started a new thread for this since it is sort of a different issue than before. I did a tpi swap on my 84, and I finally have it running really well - as long as the maf is unplugged, that is. The maf I am using is a duralast unit from autozone. The first one I got was reading a little too high of a reading and was causing a code 33, until it crapped out on me sunday and quit reading at all. I unplugged it and the car ran better than it ever has - before I would sometimes get hesitation and stumble on takeoff and had trouble getting it to rev past 4500 rpm. I had replaced the injectors a couple weeks ago since some were leaking, but it still ran the same. I drove the car a few days with the maf unplugged waiting for the replacement to arrive, and it always runs very well, with no problems and revs to 5000 rpm no problem (revs past that, but no power as is expected) Today I recieved the replacement maf and installed it, and all the same problems are back, except the code 33. I unplugged it again, and it runs great again. I do not have the equipment to datalog, but I have been watching readings on my scanner, and can't really pick out anything obvious as far as what's happening. I was hoping you guys would have some suggestions on what to look closely for to see why the maf would make the car stumble and hesitate on takeoff, occasionally die at idle, and not rev past 4500 rpm. Right now I basically have a 150 dollar hose coupler with a lifetime warranty, lol.

BTW - when I pulled off the maf that failed, it had a piece rattling around inside of the sealed part at the bottom.

 

I'm wintering in Florida and my car is in upper MI so cannot just go and take a look. But, any chance the MAF could be installed backwards - like the air is flowing through it backwards?

Interestingly enough, on my 1986, when the MAF sensor is unplugged, it will barely run.

Might want to be sure the MAF wiring is correct - right wires in the right connector locations.

Not much help but thought it might trigger some new thoughts for you.

 

Depending on what car you took the TPI and computer out of, you may have to tune the computer a little bit to get it to run right. If you took from a car with aluminum heads, the aluminum heads have higher compression that a 1985 would, but the 1984 has the same compression ratio as a 1985.

Did you change the cam? If you didn't, that would also require a tune.

There's a lot of things that come into play here.

 

What have you done to implement the MAF burnoff and MAF power?

I have heard a lot of bad things about Duralast MAFs. Apparently their version of a rebuild is to take it apart, clean things, put it back together and put it on the shelf. I guess they don't have a good method for actually testing them...

 

..... The first thing that comes to mind is that you may have a lot of air entering the intake downstream of the MAF ... possibly through the PCV vent , the IAC , 9th injector , EGR , or a large vacuum leak like a brake booster ... if air enters without going through the MAF the ECM will not be aware and will try to run the engine based on what it sees ........ What is your scanner showing for MAF grams/sec at idle ? ..... Who did your wiring harness and pin-outs for the ECM ? ... which ECM are you using ? ..... What is the broadcast code on the Prom ? .........

 

MAF calibrations vary, so plan on some tuning for any replacement MAF.

You can use your closed loop BLM values vs flow as a relative indicator of a given MAFs calibration (with fuel, injector, O2 and other vehicle influences lumped together).

You can also use the BLM vs flow data to compare different MAFs to each other and work from there.

For a more robust solution, check out Blowerworks.net or any of the many MAF threads that I've posted in related to this topic, or drop me a pm.

 

Thank you for your replies.

@ DanielRicany - The wiring harness and ecm are out of an 86 corvette (7165) Mrwillys burned me a prom to work with my iron heads and the cam that is (supposed to be) in the car.

@ Cliff Harris - The maf power and burn off relays are hooked up normally, and are working properly. (lots of testing to confirm) The duralast mafs look nothing like an original bosch, even looks and feels cheap. The first maf I had was reading sometimes as high as 65 grams/sec at 2000 rpm with low throttle (cruising) and was constantly setting code 33. It finally quit altogether and set code 34 and was stuck at 4 grams/sec. When I pulled it off it had something rattling around inside of it. I put on the new sensor and checked the readings and it looked pretty normal, followed the computer's maf reading really closely.

@ C409 - I have done checks for airleaks, and I do not believe I am pulling any air after the maf. The grams/sec were normally around 6 at normal idle. The new maf reads more normal and has not exceeded the 45 grams/sec limitation while cruising. I did all my own wiring and swapping. I have a 7165 ecm (got hold of two bad ones, third time was a charm) and I used the entire 86 engine and ecm harness, doing all the required wire swapping at the dash harness connectors and the driver's side bulkhead connector. All the wiring has been extensively checked and rechecked while going through the two bad ecm's, and then again when the maf failed. I have a prom burned by Mrwillys to work with my iron heads and camshaft specs

@ tequilaboy - I need to get the stuff to datalog, right now I can only take 125 frame "clips" and it is a little frustrating trying to "catch" an event properly, and then having to write down any info you want to keep because the new clip always overwrites the old one. My closed loop blm's were around 110 the last time I checked. I probably need a tune, and I am less than certain that the cam I was told is in the car really is.

Thanks for the help guys, I am going to try to get a record of maf values and all the rest going so I can share a clearer picture of what is happening

 

...... How does it run with the OEM Prom ? .........

 

I do not have it anymore, but it did not cope well with cold starts. Other than that it seemed to run the same as the new chip.

This weekend I hope to use the scanner with and without the maf and compare the two readings during certain conditions. I have a new mystery noise at the rear suspension, so I may end up messing with that all weekend.

 

..... How did the engine run BEFORE the TPI swap ? ..... It sounds like you might be in limp home mode ... is the EST wire plugged in ? ........

 

I got the crossfire running really well, then tried a single throttle body setup, and then went to the tpi because I came across it cheap. Wouldn't I be seeing that I was in backup spark and fuel mode on the scanner if I was in limp home mode? I have been driving the car for a week now with the maf unplugged, and it is the best the car has ever ran, even better than the crossfire (except the crossfire had a few more usable rpm's on top since it was an x-ram setup, but only a couple hundred rpm) The main reason for all the swapping is boredom and not liking the way the crossfire looks, lol. The only time I have any issues is with the maf plugged in. The est is plugged in, and base timing is at 6 deg. btdc (with est unplugged). When watching the scanner before, I was originally suspecting a timing problem, but the advance was not doing anything weird and the knock sensor was not causing any timing to be pulled. The only other thing to watch would be inj. pulsewidth, right? I plan on watching that the closest, along with differences in timing when comparing maf to no maf readings, and I will also write down any differences in the maf and airflow readings. My fuel pressure is at stock specs, but I plan on putting on an adjustable fpr along with a new diaphragm. The old diaphragm looked ok, but not perfect. Today I will do some driving and write down values before looking at the rearend.

 

Did some driving and recording today, nothing jumped out at me as obvious, but I will share what I saw.

At idle (625 rpm) with the maf plugged in, tps .58, I saw maf and airflow @ 7 grams/sec, BLM of 116, base inj pw of 1.8 mS, load of 57, and IAC of 25, with O2 lean/rich

during stumble (1100 rpm) maf plugged in, tps 1.44, I saw maf and airflow @ 24, BLM of 128, base inj pw of 3.9, load of 103, and IAC of ?

cruising at 2000 rpm (70 mph) with maf, tps .86 I saw maf and airflow @ 33, BLM of 134, base inj pw of 2.8, load of 75, and IAC of 68, with O2 lean

cutting out at almost full throttle @ 5400 rpm with maf (does not cut out at 4500 since replacing mufflers) tps 3.88, I saw maf and airflow @ 210 and 205, BLM of 128, base inj pw of 8.5, load of 186, and IAC of 93 with O2 rich

Maf unplugged -

At idle (620 rpm) tps .58, I saw maf and airflow of 5 and 254, BLM of 149, base inj pw of 2.0, load of 43, and IAC of 18, with O2 lean/rich

cruising at 2000 rpm (70 mph) tps 1.10, I saw maf and airflow @ 36 and 254, BLM of 138, base inj pw of 3.7, load of 89, and IAC of 87, with )2 lean/rich

almost full throttle @ 5500 rpm, tps 3.96, I saw maf and airflow @ 162 and 254, BLM of 128, base inj pw of 6.3, load of 138, and IAC of 99, with O2 rich

When the car stumbles at takeoff with the maf plugged in, it can be very severe, and takes a while just to pull itself at all, but when it gets to about 2000 rpm's it will take off like a rocket. At the topend with the maf plugged in, it is like a switch is flipped and it will not go past that rpm, and it sounds like it is missing or something.

With the maf unplugged, it never hesitates or stumbles, and will keep revving till I get out of it. You can tell it is running out of breath, but it's not like the sharp cutoff with the maf hooked up. In the middle, between 2 and 4 k rpm, it does seem a little more crisp and responsive to throttle with the maf hooked up, but the hesitation down low makes it suck to drive.

 

..... What size injectors ? ...........

 

The blue 24# injectors from FIC, installed a couple of weeks ago. The prom is programmed for them by Mrwillys, and the same symptoms were present with the old injectors as well (24# also).

I need to test again to confirm, but it seems like the problem with the maf plugged in gets worse the longer the maf is running. It doesn't have anything to do with engine temp - I can drive the car all day without the maf, stop and quickly connect the maf, and it seems to take a minute or two before it starts messing up, and it gets worse as time passes. I can then stop, quickly unplug the maf and go, and all stumbles are gone. I have never plugged in or unplugged the maf while the car was running.

 

..... The MAF has to go through some ignition cycles (I don't know how many) and meet other criteria for the ECM to request a MAF burn-off ... the burn off function cleans the sensor wire so that it more accurately measures the airflow ... perhaps you haven't achieved a full burn off yet and that's throwing a curve at the ECM airflow-wise .... honestly , its because of quirks like this that I switched to a Holley 950 controller and stopped struggling with the over protective factory ECM .................

 

The maf burn-off happens at shut down, and the ecm has done the burn-off while the maf is hooked up. Thing is, the car has run this way with two brand new maf sensors (although of questionable quality). This already has me wanting to go speed density, but I want to get this figured out.

 

From your data, its clear that the MAF (when plugged in) is running quite a bit richer than the default airflow calculation (when unplugged). It about 30% richer under heavy throttle at 5500 rpm, for example.

Have the MAF tables already been edited in the bin? Compare the MAF table and scalar values to a good stock bin. This may explain the difference right there. Go back to stock values if applicable.

If you still have stock MAF table and scalar values in the bin, could be that the MAF itself is reading a bit on the high side. You can reduce the MAF table values to compensate.

I'd split the difference between the actual sensor and the default airflow calc, by reducing all MAF table values by about 15% (multiply by a factor of 0.85), then re-evaluate.

 

I'm thinking at this point that at least part of the problem may be that I had MrWillys burn me a prom to go with a cam that probably is not in the car. I ran the stock prom for a very short time, and although it ran badly(rich) at cold startup, I cannot remember any issues with stumbling and backfiring - but the ecm I was running was having issues (communication problems).
I was thinking of asking MrWillys to re-burn the prom for a stock cam and see how that works, but I feel like I should get the stuff to do some real datalogging first to aid in the tuning. Guess it's time to quit thinking about it and order the cable, lol.