Longshot08

So I've been powering through the host of issues my GTA has had for awhile finally. After replacing the struts, shocks, endlinks, fuel injectors, cleaning plenum, fixing some brake issues, and etc I'm ready to move on to re-building my A/C compressor (detailed schematic with the shop manual).


Start car up leaving work last week and it immediately dies... Crap. After 6-7 cycles of startup and immediate shutdown at idle, I manage to keep her revved up at 2000-2500 RPM and limp my way home. I check the TPS voltage and it reads .65-.66 at idle and smooth transition to 4.5 at WOT. Little out of tolerance on the idle but it looks good. Disconnect the MAF and car idles perfectly. Ok so I replace the MAF with a Duralast piece. The element looks different but it's $110 with a lifetime warranty so I give it shot. Reset the computer and car runs rough at first but smooths out after a 20 min drive. Next time I drive it the car throws a code 15, 33, and 36. Ok. I replaced the coolant sensor 2 years ago (Duralast), so that's confusing, but the fans come on at 210 degrees( wired the fan relays in series because the A/C does not work), so I can ignore the code 15 for later. Re-adjust the TPS to .55 volts at idle, reset computer and the code 33 disappears, BUT the 36 is still there after re-start.


Either this is a result of the Duralast MAF, my burn-off relay is actually bad, or the wiring between MAF, relay, or ECM is bad.
With the code 15 being tripped when the coolant gauge reads 220 and then disappearing once fans come on (Duralast coollant sensor), and this code 36 craziness with the Duralast MAF I'm inclined to believe its Duralast's fault (even though I bought them and put it on). Anyone else used a Duralast MAF and experienced the same/similar issues?

Longshot08

So I performed the first 3 steps of the code 36 trouble shooting guide. I determined the MAF Burnoff relay was probably faulty because one of the circuits on it was pretty high and variable resistance wise. Like it would jump from 60-160ohms. When I actuated the relay, the other circuit read steady at 2-5 ohms. The MAF power relay read the same on both circuits. So replaced the burn off relay but I still saw a 36 after a computer reset, drive, and subsequent restart. I'm kind of stumped right now. The only thing I can think of is the Duralast MAF element being physically different from the stock Bosch is playing a role. Still this shouldn't affect the burn-off function right? Car drives fine, I just don't like staring at a code 36 check engine light every time I drive. Any pointers?

Longshot08

For reference the top photo is what my old MAF looked like. The one I have on the car now is the second. I went went with the second one because it's new not remanufactured and cheaper. The size of the whole unit, and connectors are the same. The element is the only discernable difference.

Vader

The original MAF uses the hot wire type sensor. The replacement MAF uses the thick-film resistor type sensor. Either will work just fine. I've had that same type replacement MAF on my '86 for about 15 years with no problems.

The big difference is that the TF element does not require a burn off since there is no hot wire. The replacement MAF must have a circuit which simulates a burn-off feedback to satisfy the ECM. Early versions of this type replacement had the necessary circuitry to accomplish this. There are questions regarding some of the later/more recent versions of this replacement MAF as to whether they can perform this burn-off simulation. However, all connections and wiring must be intact for the unit to function even if it is the correct type.

This might be informative:

https://www.thirdgen.org/forums/tech...ck-my-maf.html

Longshot08

Well the Duralast MAF does have 5 connectors but whether or not the circuitry is present to satisfy the ECM is another story.


I want to confirm I'm performing the 3rd step correctly. This is after unplugging the burn-off relay and confirming it has a constant 12 volts on the orange wire. The 3rd step says to unplug MAF and using a jumper wire ground circuit 900 (the solid black wire). This means running a wire from that connection and applying to any ground in the car like the battery for instance (not confident I performed this step correctly originally)?

Longshot08

Thanks Vader that was informative. I'm back on top of this issue now after getting my A/C functioning again.


Where I'm at now and total discrection here...lol. I took a good functioning fan relay (secondary one) and put it in the place of the bad (allegedly) MAF burnoff relay and bought a new relay from AZ and put that in the old fan relay location. Still have the code 36 after reset.

Longshot08

I bought another relay from AZ and am going to replace the burn-off relay AND MAF power relay and see if that takes care of the code. If it doesn't, I'm going to take the car to AZ and have them scan it and then replace the new MAF under warranty and see if that cures the code 36 problem.

Longshot08

The car seems to run fine. Only discrepancy I've noticed is wandering idle when I first start the car for the day. Will idle at 800 ish or so in park and then wander up to 1100-1200 and then drop down. Will do this for a few minutes then disappear by itself. If I it put in drive or reverse while this happens it idles normally but stumbles very slightly under light acceleration until coolant temp starts to get around the first mark (170?). After that problem is completely gone.


Probably need to clean and adjust the IAC per the procedure on the forum in the near future just haven't done it yet because I've been focusing on this code 36, fixing the A/C, and a cooling issue that I had to rectify in the last few weeks.

Longshot08

I shouldn't have waited so long to pick up a thirdgen. I'm having way too much fun fixing all these little ankle biters. Once she runs perfectly, and is painted, will be time to drop a T-56 and start to get smart on programing the computer in preparation for some aftermarket heads and "slightly" more radical cam.

Longshot08

Update:
After changing my starter and re-connecting battery and taking car for a long spin the code 36 returned. However there is a twist. After car sat and idled for about 5 minutes (playing with phone) the light disappeared (what the crap right?). Took it home because its late and shut car off. Heard the MAF burn-off and MAF power relay kick on. Waited about a couple of minutes and started car up. Light re-appeared. Still getting an intermittent code 15 to but disappears after about 4-5 minutes but stays stored.

I'm thinking this is all related to the Duralast thin film MAF sensor. Should I go to trouble of swapping it out with another (warranty replacement but it has to be shipped in which means car is down while I wait for arrival with the old one turned in) or just wait till I get a full on data logging setup/chip burner setup and figure it out then?

Longshot08

I hooked up thin wires to the MAF sensor connector and plugged it in to try and see what the voltage readout is during the MAF burnoff cycle. I get no volts on my DVM when the car is shutoff after entering closed loop AND hear the MAF Power and Burnoff relay click. This was with the leads on the solid black and white stripped with red or blue wire.


I did get a proper reading from the green and solid black wire at idle. MAF .8 ish volts (don't remember exactly) at idle with a linear increase to higher volts with higher RPM.


I am going to redo the check again to confirm no volts during the burnoff but it appears this Duralast MAF is not supplying the dummy load to the ECM on shutoff. May get it replaced.


More to follow.


EDIT: This is for the Code 36 the SES is still flashing. The SES still comes on after 1st restart following an ECM reset and never goes away.

banksGTA

iTrader: (3)

Anymore lead way on your issue?

Longshot08

Not yet. I just got my scan cable but the car wont start because of a shot fuel pump. Currently waiting for new fuel pump to arrive.

GeneralDisorder

iTrader: (1)

Your fuel pump arrived on your Porch ~20 minutes ago BTW.

GD

Longshot08

Indeed it has. It's a nice looking kit.

GeneralDisorder

iTrader: (1)

Swap pump tomorrow? It's a steaming pile of fun.

GD

RBob

iTrader: (1)

The MAF power relay can also cause a code 36. The NC contacts are used during burn off.

RBob.

Longshot08

The funnest part is the pulling and rotating tank repeatedly for 30-45 minutes while asking your wife "is the filler neck going to clear this time?". Good times.
Half done just got to put everything back together now.
Did you re-use your old pump isolator or did you make a new one GD? Or did you just not use one period?
The wiring was a bit different but nothing cosmic to figure out, and I ditched the pulse dampener. Read it is was an unnecessary point of failure. The old pump is not factory and a fuel strainer was never hooked up which explains the car dying in sharp turns after a 160ish miles a tank. Also fixed my old fuel level sender. Guess the PO wasn't lying when he said he had fuel pump replaced however it was installed without a strainer, AND the bottom inlet was clogged probably because of lack of strainer.

Looks like its easier to get tank in then out. Truth?

GeneralDisorder

iTrader: (1)

I think I just made something work for the isolator. I had to buy a whole new tank, and pickup assembly in addition to the pump. I probably modified whatever came with the pump, or the pickup assembly.... or something. It's often the case with aftermarket pump "kits" that the isolators aren't a very good fit. Seems all too common. Usually have to break out the utility knife and clearance them a bit.

I remember having to tilt the tank quite a bit to get the filler neck into place. I did it on a lift though.... I remember the most difficult part was compressing the foam tape I used for the straps. I got some adhesive camper foam tape and it was pretty thick. The job got done though and I had to modify the fill cap to fit the aftermarket tank threads and the aftermarket sending unit reads almost full when it's full. So hooray for aftermarket parts! Pump works great. I believe the damper wasn't included in my replacement pickup and the new turbine pumps don't have the pulsation problems they old pumps had so not needed from my experience.

GD

Longshot08

Oh ok so slightly different situation so different solution. I have the fuel tank strap material worked out. The old pump is wrapped in a black rubbery foam like material. Probably going to use that and bend the fuel pump assembly arms around it to ensure its a tight fit. The old pump was zip tied to the fuel return line and there was a lot of room between the pump and the fuel pump assembly arms where pump actually sits.


If I knew the type of material to buy that won't melt and break up being immersed in fuel into hard little pieces that will float around in tank forever I would just get that and make a little fuel pump wrap myself, but my search skills have failed me so far.

Longshot08

Buna-N or Nitrile Rubber will work apparently.

GeneralDisorder

iTrader: (1)

Buna-N and Nitrile are the same thing. Both are basically natural rubber - pretty much the same family as Chewing gum. LOL.

GD

Longshot08

That's why I said "or" to indicate two names for same things.


Not really the same as chewing gum, unless having the property of being flexible and chewy is what qualifies as being pretty much the same thing.


Can you immerse chewing gum in gasoline and it not deteriorate over time?


I was a lazy bum yesterday and didn't touch my car when I got off work yesterday. Today's the day though...

GeneralDisorder

iTrader: (1)

Surprisingly close to the same thing. They make gum from Butadiene Styrene (Buna-S). Buna-N is more resistant to chemicals and oils, yes. But in a pinch I've seen Grape Bubblicious used to seal hydraulic fittings. It will work for a little while.

GD

Longshot08

Was that before or after chewing? Work long enough to autorotate?


I just realized how off topic this thread has become. Apologies to the mods. I will haze myself appropriately and get back on original topic once car is running.

Longshot08

Update: Replaced Duralast MAF with another new unit (another film element MAF not hot-wire). Engine runs fine however, still throws a code 36 every time engine is started. The burn-off relay functions (hear the click on engine shutoff, see the reference voltage when tested with a voltmeter). So apparently this unit does not simulate the burn off function for the ECM so I still get Code 36. Code 15 and 33 are gone.

At this point I either stare at the light until I program the trouble code out (if possible have to research that) or figure another way to get the ECM its feedback signal to give it the warm and fuzzy that the burnoff function is performed. How? Open to suggestions.

Tuned Performance

iTrader: (92)

Code 36 can be programmed out of the tune. The mask or I’d you are running is called $32b. You can get tunercat or tunerpro and a burner as well as a moates.net g1 and program it out.

GeneralDisorder

iTrader: (1)

If you are running 6E, just make note that the burn-off function has an internal failure counter and will only trip the code after like 3 failures (or was it 5? 6?) in a row.

You can either tune out the code, or you could fake the computer by giving it the voltage it wants to see during the burn-off cycle.

Also make sure your burn-off and power relay's aren't swapped. That's what caused my code 36. They will work and it will run with them swapped but you will get burn-off failure code. Took quite a while to figure out that foolishness since I'm running 6E and it had to cycle the burn-off function multiple times before the code would trip.

At any rate - if you are using one of the film style MAF's - I don't believe the "burn-off" function is even a feature of that type of sensor. AFAIK, no modern MAF has that capability, nor is it needed if you have good filtration.

GD

Tuned Performance

iTrader: (92)

How would one fake the burn off is it a 5v or 12v back to the ecm ?

http://www.chevythunder.com/maf_code_36.htm