Springster

1986 2.8L w/MFI I'm not the owner of this car but trying to help others figure it out. It had been in storage for 3 years and most likely has plenty of varnish in the fuel system. Five gallons of premium were added to what was in the tank and the car ran at highway speed ok but the idle was rough. It was tuned by a shop and driven a bit and the cat was glowing and the idle was still rough. The next day it would not start. It could be made to start and run (backfiring) by continually pumping the gas pedal. Otherwise it will fire a few times and die when cranked without touching the gas pedal. Fuel pressure is correct. Pressure regulator is not shooting fuel out the vacuum port. Has new plugs, wires, cap and rotor. I'm thinking the tank and lines need to be de-varnished using thinner and gas mixture.

What I don't understand is why it was running then after sitting overnight it will only fire for 1/2 second then stop. I'm wondering if the varnish laden fuel gummed up the injectors when they cooled off? I'm also wondering if varnish effects other parts. Thanks in advance for any advise!

Springster

After reading a bunch more on this awesome site I think the issue is the ignition module. Seems to make more sense than fuel injectors plugging up over night.

Gumby

a glowing cat, is a clogged cat, probably melted down on shut off as no gasses were forcing passages open.

Springster

That sounds like any easy thing to test out. Just loosen the cat and crank. Thanks! Makes a bit more sense. I suppose the varnish doesn't burn in the engine but gets passed on to the cat and can't burn there either and gums it up?

Springster

Welp. Wasn't the cat. I'm thinking ignition module again. I've had more time to look at it, since I bought it . From what I understand, the ignition module has 2 modes. Crank and run. Seems like the crank mode works because it does a normal start and sounds perfect. About 1 second later the motor fires oddly and slows down and quits. If I crank again it fires up fine again and dies a second or 2 later.

I pulled the module and plan to bring it in for testing.

Springster

Ignition module tested ok, several times.

I watched the fuel pressure and it stayed at 45psi while it ran for a second then died.

Any suggestions on what to check next? I am thinking of watching the pulses at the ignition coil with my oscilloscope.

Springster

Scope shows even pulses to the ignition coil for a while then they spread out and the engine dies. Like the computer was trying to achieve a slow RPM. Shop said they tried new TPS, MAP and MAF.

Springster

Well a bit of progress. The car starts and stays running with the MAF sensor disconnected. After it's been running a few seconds, connecting the MAF sensor will make it run a bit better. I've never had one go bad so I'm gonna check into how to test one. My old Autoxray screen died so I can't monitor it.

So I'm learning that unplugging one sensor at time can gets some results. A slow process of elimination.

I'll post some pictures later. Shiny with no rust, super clean interior, 65k, nice tires, for $2000 (as is not running). I'm feeling pretty pleased now. The down side is that I'm 6'4" and my head hits the ceiling so my wife will be driving it... Sigh... I am sooo envious. Makes a good 25th anniversary present though. She loves it. And a I scored a few points.

Maverick H1L

iTrader: (8)

Simple easy way to test a MAF is to remove it and look in the engine end. If the orange film in the middle of the board (the actual sensing element) isn't completely flat when a light is reflected off of it from the other end, it's bad. If the film is dirty (get MAF cleaner), it won't read right. The film needs to be completely flat to read properly.

You can test the MAF without a scanner provided you have a multimeter that will read frequency. Our MAFs are oddballs that put out a frequency signal instead of a voltage signal like everything else. Anyways, you read the signal at the brown and black wires. The frequency should be 32Hz at idle.

The really lazy way to test it is to hold the engine speed at about 2000 RPM, if possible, after a cold start and until the temp gauge reads over 100*. If you release the accelerator and the engine is willing to idle, your MAF is bad. IDK why it works that way, but it does. I had 2 MAFs that were defective do this.

If the MAF turns out to be bad, GET A GM MAF. You'll regret it if you buy aftermarket anything.

FYI, any shop telling you they replaced the MAP sensor on that car is ripping you off. There isn't one. It's not carbureted, TBI, or 90-92+.

Springster

Thanks for the MAF test info. I used an oscilloscope to read the MAF and got about 50hz. Looks like MAF frequency follows air flow. Seems to be working but I have to get another issue fixed first, i think. Very clean inside the MAF but the foil is not flat.

Unplugging the MAF will cause the engine to stay running after startup. But it has a hesitation issue that might be why it doesn't stay running when the MAF is connected at startup. Once it's running I plug the MAF back in and it runs but will almost die when I open the throttle any amount. After about 5 minutes of idling the RPM starts increasing and will go back down if I give it a quick tap on the gas pedal.

My old AutoXray screen died and I'm waiting for a replacement so I can see what is going on better...

Just double checked the TPS. It was at .8 so I set it to .5v. Bottom screw was loose... No change to the startup issue though.

Maverick H1L

iTrader: (8)

50Hz means the MAF is bad. Period. It's over-reporting the air flow by about 25%. Again, the MAF signal at idle should be 32Hz. And no, a MAF putting out higher than normal frequency WILL NOT ALWAYS THROW A CODE 33. It creates other issues such as hard starting and running rich (code 45). My last aftermarket MAF was putting out 48Hz. No code 33 until just before I replaced it, only code 45. Replaced MAF with a GM 25507668 MAF (Firebird MAF, Camaro P/N is 25507557) and no more code 33 and no more code 45. All other sensors within spec.

Springster

[quote=Maverick H1L;5784828]50Hz means the MAF is bad. Period. It's over-reporting the air flow by about 25%. Again, the MAF signal at idle should be 32Hz. quote]

Thanks for the info! That lead to an idea to confirm the out of spec MAF. Since it's reading high why not modify the air flow to lower the frequency? So I put a chunk of black tape in front of the foil element (on the MAF screen/air straightener). After trying a few spots the engine stayed running! It was not a very good idle but at least it started and stayed running. Tweaking the tape location and size would help but it's time for a new MAF now that I know the MAF is for sure bad and not some other device, like the ECM. If I were made of money I could have just bought one and tried it. Instead, I had to go and learn something new... LOL Thanks for the input Maverick!

So this made me think that it would be cool to have a MAF simulator. Seems like all you need is square wave oscillator controlled by the TPS? At 0.5v it outputs 32Hz and at 4.5v the output is whatever the upper Hz value is (anyone know what it is?). Maybe won't work at start up? I notice the MAF puts out 10Hz at no airflow. Thoughts? Maybe only used once every 1-5 years?

Maverick H1L

iTrader: (8)

IDK the MAF signal at KOEO (engine not running with ignition on). And have no way to find out since I sold my factory replacement MAF last year and run a F0rd MAF now having cut out the factory MAF connector. Before you ask why, I'm running a MegaSquirt and want the system to be as precise as possible.

I actually had this idea for what you call the "MAF simulator" but in reverse so I could keep my factory MAF with either my MegaSquirt or a 165 ECM, where it converts the frequency output to a voltage signal. I suppose that if you got creative enough you could make one that does both, frequency to voltage and voltage to frequency.

Keep in mind that the TPS output voltage key on and throttle closed is supposed to be .55V. Upper MAF frequency is 150Hz.

Springster

New MAF didn't make a difference. It's remanufactured so I suppose it could be bad out of the box. The film is pretty flat. I think my next step is to borrow some from a junkyard that owes me a favor and try them out. Or try a different computer. The garage shop tech said he swapped MAFs also. Is it possible to have that many bad MAFs?

I can't get the car to idle so I watched the frequency during start up. It peaked at about 50hz. At key on it is 11 hz.

Lets say the MAF is working fine. Fuel pressure, plugs, wires, ignition module, timing, TPS and EGR are fine. What would you check? I have an scanner and it shows all sensors working fine. Something interesting I just thought of. The MAF does not show a constant reading like the other sensors when they are being monitored. I even blew some air through the MAF to get different frequency outputs and the scanner didn't seems to monitor it very well. I usually got a low value or a higher value that seemed to only update sometimes after I cranked the engine. Is this normal or should the value always be updating even when the car is off but the ignition is on and MAF powered? Is the computer not reading the input? That would look like a bad MAF...

I'm gonna pull the computer and look at the board. I looked at the signal at the connector and it matched what was at the MAF. Maybe I'll get lucky and find a bad solder joint...

Springster

Using the scanner I monitored the new MAF and am now getting some readings. 3gm at key on and up to 9gm when starting. I'll have to try the old MAF again just to double check that I wasn't always getting readings.

I've read in some posts that you should be able to start and limp home with the MAF unplugged. This is interesting because it doesn't happen with this car so can someone confirm?

Another bit of info I didn't mention: After the car has been sitting for minute, I crank and it fires up fine and dies. I try again the same thing. After about 3-4 attempts it will no longer fire up and cranking longer will not get it to fire. (fuel pressure is still up at 42) Wait a minute and it's good again. I've seen this sort of thing with a faulty ignition module in my Iron Duke powered car. I did take the Camaro's ignition module to a store and had it tested about 5 cycles with no issues. I suppose it could still be bad. I don't see anything on the scanner to check the ignition status.

So that makes 2 questions:
Should the car always start in limp mode with MAF disconnected? (with no other problems).
What would cause it to stop firing after a few crank and die cycles?

Thanks!

rusty vango

as mentioned, DO NOT BUY a re-man MAF.i would bet you ended up with a "A1-cardone" total junk right out of the box. many many of us have fought that battle.only to have to buy a new one. I have a TPI car ,and my new one set me back $450. there really isn't anything else you can do. as you are finding out. sorry buddy, and good luck with your project

willexoIX

Yea, when I went to grab a new MAF when I was having problems with the stock MAF, they gave me a reman even though I asked for new. Put it on the car and had the same problems. Lo and behold I looked at the film on the "new" MAF, looked like waves from kids splashing in a pool. Not exaggerating either. I returned it because the film on the old one was much better looking, but still not flat.

I drove mine like that for about a year and a half until I decided to get rid of the MAF altogether and go with a MAP powered Megasquirt ECU. Best thing I ever did for the car.

Moral of the story, do NOT buy a reman MAF, your better off using the broken stock one FWIW. And at the prices even the reman ones cost, its not worth it by a long shot.

Gumby

there is part of the problem, returns go right back on the shelf.

same with starters that wont hot start, it will bench test fine cold and go right back into a box.

willexoIX

Yup, I have told the chain stores that the parts are bad and to send them back, but I watch them put it right back on the shelf. Same spot it was when I came in for the part to begin with. Some stuff they mark defective, but not all of it.

I even watched one of the employees put " customer didnt need" on the return slip on a defective part as well. Sigh..

Springster

I ticks a guy off that junk is being sold as good stuff. Adds to the frustration of troubleshooting. The new MAF has a light wave to the foil. The old one is cleaner and straighter. Both are GM. The old was it not a rebuilt. PepBoys didn't ask for a core while other stores do so I at least can take the new one back easily.

I hooked up a square wave generator in place of the MAF and set it to 32hz. Didn't make a difference in start up. Maybe it isn't that easy. Maybe the computer is expecting a transitioning signal. It was easy to do so I tried it.

If your supposed to be able to start with the MAF disconnnected and my car doesn't start I'm wondering why.

I unplugged the 4 wire connector on the distributor and cranked. It started and quit same as always. I'm gonna learn what that's about next. But I'm still not gonna say the MAF is good. I'm just confused why it doesn't start in limp mode. There could be 2 problems.

Gumby

people always yell that its not or wasn't needed but 10+ years ago I got one from Autozone, they said you must unhook the battery both cables, on install or they would not warranty it. Always done that when it needed removed. [ now with extended wires, I can just move it out of the way easy]

also trace the MAF wires, you will see they go to an open air junction block in the nose, maybe its to crusty.

Also have ya pulled the ECM and opend it up and looked / smelled for any burnt sections? [using your nose finds bad spots better]

rusty vango

I don't advocate the shotgun approach.but im thinking that you may well have a faulty memcal(chipset) and/or a faulty computer.since you have done everything logically. it seems likely at this point

Maverick H1L

iTrader: (8)

Um, unplugging the 4-wire dizzy connector guarantees a no-start. The ECM isn't going to get reference pulses with it unplugged so there won't be any injector pulses. As I've said before, "No spark, no squirt". If the ECM doesn't see the reference pulses, it doesn't know the engine is turning over, so it won't fire the injectors.

You're "running" off of the cold start injector. If the engine is puffing, you have spark, which means that at least one part of the ignition module is working. But there is no guarantee the other part is UNLESS the ECM is outputting an engine speed signal in RPM to the scanner.

After I replaced the REMAN AZ ECM, I never could get my engine to run in limp mode caused by unplugging the MAF. I had to leave the thing plugged in and run the engine above 2000 RPM until the temp gauge got off of its butt and started showing engine temp...

Oh, and the "reman" places just LOVE to shove the "reman" parts back into GM housings...

Springster, Camaro or Bird? The Camaro MAF is 25507557 and the Bird MAF is 25507668 (possibly have new numbers now but I doubt it). I just looked on eBay and so far there's 2 Bird GM MAFs for under $80... If you want to try another ECM, I have one I can't do anything with right now...

rusty vango... Our V6s use the 1227302 ECM. No MEMCAL.

Springster

I checked the rpm with the scanner and got 450 to 500. Bad ign module? It tested ok at AZ. I gotta get some sleep before work otherwise I would hook the oscilloscope up to it.

It's a Camaro with the 7557. More putzing tomorrow. Good thing I don't need it to get to work.

Gumby

car ever have an aftermarket tach? those quick connects they come with will cut the signal wire strands inside the insulator.

Stewie

The ICM can cause some strange symptoms. You can have a bad one even if it tests good. As I recall, I asked the parts store to test it several times in a row to get it warm. Sure enough, bad ICM. I also bought a spare to keep in my bird because one can go bad at any time. And, it takes only 15 minutes to replace.

Good luck solving. I just found a cheap to fix gremlin ($11) in my bird that took me several months to find.

Maverick H1L

iTrader: (8)

Next time you get a chance and it's not running, unplug the front injector harness connector. Use a test light hooked to BATTERY POWER (clip the clip either to the battery, alternator BAT post, or some other constant power source) and probe the blue and green wires in the engine harness side of the injector harness connector. The test light should flash while cranking the engine over IF the ECM is both receiving reference pulses AND the injector driver is working to pulse the injector circuits.

Springster

Maverick, did some reading and learned about cold start injector. This is my first car with it. I agree it's running on it. Because after a few start attempts that injector stops squirting due to a thermal circuit heating up to prevent it and the engine stops firing. After letting it sit a bit it will fire again. Thank you.

I will do the injector test tomorrow. I scoped the ICM and found the ECM is controlling spark advance. Though I didn't test the output to the coil, just the output from the ECM to the ICM. So it seems like the ECM is getting dizzy pulses. I am learning, slowly. Lol

I know that the injectors can work because I have had it running, poorly, and used a stethoscope to listen to them.

Maverick H1L

iTrader: (8)

Okay. In detail, the CSI works until the engine temp reaches 100*. In theory, the CSI stays open until that point, but I don't see how because when the key is released from the START position, the power supply to the switch is turned off. Unless your ambient air temp is getting over that and the engine and the CSI switch are being heated above that, the CSI shouldn't be shutting off.

There are 2 fuses in the fuse panel for the injectors. FP/INJ 1 and INJ 2. Check them if you're not getting any power to the pink wires in the injector harness. You apparently have power to the fuel pump (which is powered off of the same circuit as the ECM constant power circuit for whatever reason).

Springster

All checked out good there. I also scoped the ecm ign pulse output (white wire) and the icm (black/white tach wire) and they matched. Should mean the icm is good.

Maverick H1L

iTrader: (8)

What's the injector resistance per bank? Measuring at the front injector harness connector on the 1st and 3rd and the 2nd and 4th terminals, the resistance should be close to 4 ohms. 6 or more is a problem along with 3 or less...

You have spark, you have injector pulses, you have fuel pressure. Only thing left is an injector issue.

Gumby

Hope you’re not getting over technical, had a friend like that, it looked like an ER with scopes n graph machines and take hours to do very simple task by over complicating it.

Maverick H1L

iTrader: (8)

Well, at least he's sure the ECM is working now...

Springster

Lol. I'm only doing it because I have time to on this car. I know Iron Dukes real well but this is my first 2.8 mpfi to figure out. : ) And I REALLY appreciate the help. : )

Using my cheap meter I got about 5 ohms for both sets of injectors. Same results must mean they match and are ok?

Just to rule it out, I checked timing by pulling #1 plug and made sure timing mark was at 0 degrees and dizzy was on #1 plug wire at TDC. Was fine but the plug was wet. I think this matches what the maf is doing (reading 25% high). It was plugged in for the testing so the ecm was injecting a rich mixture. I know I gotta get a good one and maybe I'm wasting my time till I do, but shouldn't it run with the maf unplugged? Seems like I have 2 issues or is the 2.8 a bit fussy? Lol

Maverick H1L

iTrader: (8)

Well, now you need to take off the upper intake and check the individual injector resistance. Each injector should come out to NO LESS THAN 10.4 ohms.

Springster

Due to overtime and fixing other cars I haven't had time to pull the intake yet but I was talking to a coworker about it and he mentioned that the ethanol in todays fuel could have ruined the injectors. What happened to him is that he was driving a 1990 GM (didn't remember the model) and it just quit on him on the highway. The mechanic told him this was a common problem for the erra. They replaced the injectors and all was good again.

So I did 5 minutes of research on this site and started finding there are issues with ethanol eating injector coil insulation. One of the times I was able to get the engine to idle I used a stethoscope and listened to each injeector. They were all clicking but if the coils are partially shorted then the injector(soleniod) isn't opening fully, I imagine.

I'll let you know some injector resistance measurements some time this week...

Edit: Later I recalled that this car had been in storage for "only" 6 years but I believe we ethanal added to fuel before 2008. If it were to have issues it should have showed up then? Just a thought.

Gumby

That stuff has been in our gas local since late 80s, never seen it cause a problem.

Springster

All injectors ohmed ok. 16.2-16.5 ohms

Doesn't mean they are not plugged. Time to take em out and inspect and clean unless there is something else to check?

I REALLY like working on this car! Most stuff is easy to get at and easy to take apart.

Springster

Three plugged injector screens... All on the same side of the fuel rail. Seems weird? The screens were full to the top of a crystallized stuff which fell apart to a fine sand when I poked at it. Anyone know what it is? It was also stuck to the rail where the injector sits but no where else.

rusty vango

man this is more common now than ever.im finding that in some cases, its deteriorated plastic.sometimes its like super fine rust particles.and SOMETIMES its when customers buy super cheap fuel filters.they come apart.i would flush out the fuel lines and the fuel rail.replace every last o ring . and if you don't already ,buy WIX fuel filters. who ever thought itd be a good idea to add ethanol to our fuel supply. may they get poision ivy on their ball$.may their wives hate them forever .and when they die and get to the pearly gates , may ST. pete be there working on his Z-28. with your very same problems. amen.

Springster

I cleaned the injector screens and still have the same problem. To prove the injectors where flowing I put the pressure gauge on the rail and let the pump go for its 2 seconds to get 42 PSI then pulsed each bank using the connector pins under the throttle body (disconnected from the ECM). The gauge dropped so I figure at least 1 of the 3 are opening. I may have to try one by one next but no time today...

I brought a current clamp home from work that plugs into a scope to make sure the ECM is putting the expected amount of current through each bank. So 16 ohms in parallel is 5.3 ohms. 12v/5.3 = 2.25 amps. I was getting around 1 amp pulses but the current probe is confusing so I'll maybe try another. In the mean time I put the scope on the blue and green wires going to the injectors to measure the voltage and it looked normal. There was 12v which was getting pulsed to near zero every 3rd ignition coil pulse for about 8mS. But I forgot to test that the 12v feed to the injectors was staying high...

Another thing I played with was the timing. I loosened the dizzy and moved it around to try to improve starting but it was ok where it was. I had the ECM timing plug disconnected. I also watched the spark output from the coil to the block via a modified spark plug with a huge gap and it seemed pretty strong.

One more thing I did, just for the heck of it, was to try a MAF from a running car. No difference.

Oh, and I moved it into the garage and noticed the exhaust stinks which is weird.

My head hurts... Mostly from a cold but this problem isn't helping. sigh

Gumby

did ya back flush the whole system??? more junk could be incoming.

Maverick H1L

iTrader: (8)

Check the resistance of the CTS and IAT. The IAT doesn't matter as much as the CTS but they should be relatively close without the engine having been run. Keep in mind the CTS is the coolant temp unit with the round plug in the intake manifold. If the CTS is open then the ECM will be reading crap for a coolant temp and the engine will run like crap because the injectors are being fed bad signals due to the ECM not knowing engine temp.

Springster

I let a bunch of fuel flow into a glass container (sounds safe ) and there was nothing floating around or sitting on the bottom. (I had a "Gumby" for a roommate in '84, we called his '76 Monte "Pokey" )

The Autoxray showed about the same values for IAT and CTS. About 3 degrees difference at "room" temperature.

So I wasn't sure which fuel line was feed and return and took them both off the rail and added hoses to get to my container. Later I tried blowing through the return line and couldn't do it with my mouth pressure and didn't want to break anything with compressed air. Didn't seem normal but I'm a newb to the F-Body. So I did some searching and found some related problems but not on the 2.8L. My tank is 1/4 full. Should I be able to blow through the line with my mouth? Some other threads found the same white powdery stuff plugging things as I found in the injector screens. Is there something to plug up in the tank where the return line dumps the fuel back in? Though I also read that if the return line is plugged then the rail pressure will be high. Mine is 42PSI.

Thanks guys. I appreciate it!

Maverick H1L

iTrader: (8)

I've put 90-100 PSI through the return line to blow it out, from the port under the FPR diaphragm to the fuel tank. Shouldn't be any issue there. But that doesn't sound like your issue. Although it wouldn't hurt to make sure there isn't any crud remaining in either line (disconnect the pressure line at the filter first and then attach a hose to it). Have you replaced the fuel filter yet? May also try a spare set of injectors as well since the screens may be plugged in a way you can't clean. There is a place somewhere that sells a kit with a tool and new injector screens, but I don't remember offhand who or where...

Gumby

same junk is probably clogged in there as the fuel circulated.

Springster

Well, I think I remembered backwards... Cause I took it apart under the hood where it bolts to the rubber hose and I was able to blow and hear bubbles in the tank. And I turned on the key with the return pipe disconnected and clean fuel flowed out while still getting 42 PSI at the rail so regulator is plugged. It was probably the feed line I was trying to blow through a while back. My bad.

Next thing I'm gonna do is take off the cold start injector and check it out. Before taking the rail apart the engine would fire up quick then peeter off. After putting the rail back together it seems like the cold start injector isn't squirting as much. Some of that powder might be plugging it.

The fuel filter was changed after they took the tank off for cleaning but I have another and will put it on after flushing with the current filter off.

Springster

The cold start injector screen was pretty clean and it squirts. So I'm checking that off the list. What's strange is that after I cleaned out the other injector screens, now the engine does not fire up the same. It seemed like the cold start injector used to dump more fuel. But maybe what is failing it just getting worse...

chevythunder.com has a lot of good reading and tests. I found a cool way to test that the ECM is pulsing the injectors. Tap 12v into the 4 way dizzy plug(purple/white wire) and the noid light flashes because ECM thinks dizzy pulses are coming in. I also had the ignition module tested again at a different place and it was still good.

I still cant find anything wrong yet. The only thing currently weird is the exhaust smell. It doesn't smell normal. The raw gas smells fine.

I sprayed some starting fluid in the air cleaner and that made it fire more but not stay running. So it must be fuel starved or have a poor spark? Is there something I can do to make it run richer? I know the MAF controls the fuel amount but I've tried 2 new ones and one from a working car.

I also tried pulling the PROM to force it to use the calpak values (I think the MAF is not used then also). It fired a bit different but still no start.

What should the O2 sensor read when cold using an Autoxray? I'm getting about 480mV.

Would it be helpful to post all my Autoxray readings?

Thanks again for the help. My kids '86 Century wagon with an Iron Duke sounds like a rock tumbler (bearings shot) so I'm able to spend more time working on the 'maro so my wife can drive it and my kid gets her old car.

Gumby

another odd isue thred not long ago, was a bad cat not letting it run.

Springster

Pulled the cat but still no start.

Did a compression check (but not with throttle wide open and all plugs removed like your supposed to), I did one plug at a time. All read the same at about 140psi.