Author Topic: My 91 BMW 318is repair log (Read 10400 times)

Wrench · « **on:** February 29, 2012, 06:33:48 PM »

Greetings... I'm a newbie to the forum but not to BMWs. I've owned a 3 series and I work on them for a living.

I recently found a red 91 318is that I bought not running. I have it running but with issues that I'm still sorting out. I love the 3 series, had been shopping for one and finding one of these was a stroke of luck. The guy just showed up at the shop one day wanting to sell it. We struck a deal.

I use forums and their data daily in my job to help speed up the diagnostic process, which is a process of elimination. Forums are full of what didn't work but that eventually leads to what DID work.

I've decided to share the journey with this 318is, from not running to eventually running perfect and what it takes to get there... so, here goes:

2-3-12 Purchased car for $500, not running
Car will crank but not start. Test for fuel pump, BAD.
Replace Fuel Pump, under rear seat cushion passenger side.
Car now gets fuel, fires erratically, but no start and run.
Check main relay, diagnose as faulty.
Replaced main relay in engine bay, middle of firewall, first one on passenger right
Car now almost runs, sputters and dies. Eventually it will run but hunt up and down in RPMs. When accelerator pedal is pressed, it will start to rise in RPMs then suddenly cut off, then at 1K RPMs fire and run again. Checked Air Flow Mater and found it to be used, top had been removed once. I removed the cover and it was all corroded inside.
Replaced Air Flow Meter with a new unit.
Car now runs the same, still stumbles, won't rev.
As most of you 318is owners know, the ECUs in these are known for going bad. So I found a used one.
Replaced Motronic control unit inside glove box, upper compartment behind removable plastic panel.
Car runs a little better, takes more RPM but still not right, stumbling.
Run "stomp test" turn key on press pedal to floor and release five times within 5 seconds. CEL blinked 1244, Cam Position Sensor
Replaced Cam Position Sensor front of engine on timing cover
Car ran the same, set the same code.
Disconnected O2 sensor and suddenly the engine ran smoother and it would accelerate. Plugged it back in and it continued to run fine for a while. I let it idle for about 20 minutes then it just died. It started and ran again for another 15 minutes and stalled. I took it out and drove it, it ran ok, good acceleration, a couple of misfires but ran OK, so, I drove it home from the shop tonight. It stalled once at a light, restarted ok but still has a miss here and there AND the CEL comes on steady every few minutes. If I let off the gas it goes out, then I can get back on the pedal.

UPDATE 2-4-12 Replaced TPS. No change. Still won't rev.
I connected a timing light to see the ignition signal. It does indeed cut out sometimes, but what's odd is, it will fall off RPMs most of the time and still show that it's firing.
I also checked t\he injector pulse. It goes up in value with RPM but it still falls off at about 2500 RPM and as it does, it continues to show a pulse.

UPDATE: 3-5-12 Replaced O2 sensor,
Replaced breather hoses, runs the same.

UPDATE: 3-6-12 Replaced all fuel injectors, runs the same

UPDATE: 3-9-12 Replaced replaced temp sensor, runs the same

This one is a puzzle

Opinions appreciated and please feel free to hijack this thread if you have issues like those posted here. We'll solve the problems together.

Cheers!
Dino

DesktopDave · « **Reply #1 on:** February 29, 2012, 08:08:31 PM »

Nice summary...that sort of methodical troubleshooting is required for older cars like these,. I had a lot of the same trouble with mine when I bought it. I use forums the same way. I'm no pro, but I get a surprising number of them right.

Have you tried to replace the O2 sensor? They can cause all sorts of havoc. Splice in a Bosch pre-cat sensor for a cheap fix. The Ford V8's used the same one we do with different leads.

This is a longer shot, but you know what M42 issue I've seen a few cases of...bad coil or coil driver. They tend to fail in pairs...if one is dying, it typically damages the other one. Swapping can work short term, but eventually the damage is repeated on the swapped part.

Unfortunately, I don't have a good way to test coils under load. Without diagnosis it's just a theory. I've verified this personally...pulled a DME apart and found driver transistors scorched despite the large heat sink. I'm still looking for a replacement part, but they're OEM Bosch. Dr. Robert tends not to share his specs.

Wrench · « **Reply #2 on:** March 01, 2012, 06:10:22 AM »

Thanks Dave... I ordered an O2 sensor yesterday thinking the same thing. I also forgot to include the Air Flow Meter I replaced in my above list. It's edited and corrected now. I thought about the coils as well. I may end up replacing them just to get everything up to date. This car has an odometer that stopped at 202547. I have a repair receipt I found in the car dated 2007 with that same mileage so there must be well over 250K on this car. It could use some refurbishing but over all a nice straight car.

Hope it gets me to work this morning. Guess I should leave early. lol

jscribble · « **Reply #3 on:** March 01, 2012, 11:14:12 AM »

Quote from: Wrench;110666

I thought about the coils as well. I may end up replacing them just to get everything up to date.

If you do replace the coils, look at our resident race team's coil-per-plug kit. rob_e30 can give you the info. Excellent deal, to boot!

locknload · « **Reply #4 on:** March 01, 2012, 07:31:34 PM »

One suggestion: Unless your replacement CPS was OEM or BMW, it may still be bad. Lots of problems with aftermarket CPS not having correct resistance value. Check with OHM meter to be sure.

Wrench · « **Reply #5 on:** March 02, 2012, 06:52:34 AM »

It's a Bosch part so no worries. Need to test the TPS today...

Wrench · « **Reply #6 on:** March 04, 2012, 06:54:06 PM »

See update 3-4-12 in first post...

Wrench · « **Reply #7 on:** March 04, 2012, 10:21:57 PM »

So after reading several posts describing the same symptoms as I have I still don't have an answer.

It seems it's ignition related, but I'm now taking a different approach.
I'm starting to think it's a wire harness issue. First suspect would be that damn junction box under the intake. Also, the wiring above the injectors. These both have crimped connections with a rubber cap on them. I've see this become an issue on Mercedes and BMW 6 cylinder engines causing misfires. Eliminate the crimps and solder those connections then cover with heat shrink tubing..... problem goes away.
It's worth a try. If nothing else, it's a good upgrade to have done.

After that I think I'll open up a spare ECU and replace/upgrade the injector and coil driver MOSFETs.
I'll probably post it as a project on my site, www.hackaweek.com

DesktopDave · « **Reply #8 on:** March 05, 2012, 01:17:56 PM »

I wonder if you have a power problem. The DME will compensate for voltage...perhaps you have a faulty compensator on the DME? It's pretty standard on RPM, right?

Maybe the alternator or battery has some issues with regulation? Bad ground somewhere? Just tossing ideas out, I'm no EE.

Wrench · « **Reply #9 on:** March 05, 2012, 01:25:11 PM »

Just replaced the old spliced in O2 sensor with a new OEM unit. No change in running condition.
Next item on the list is the injectors. I have a set of Bosch 4 hole injectors to install and while I'm at it, I'm going to solder all those crimped connections in the junction box and fuel injector harness.

I'll tear into that ECU one evening this week and poke around.

Wrench · « **Reply #10 on:** March 06, 2012, 12:50:43 PM »

Well I just finished replacing the injectors and it runs the same... like crap, rough idle, won't rev.
So now it's down to the wiring harness. I removed all the tape wrapped around the harness along the firewall and I found four big splice points in both positive and negative leads. I'll be checking these for weak or corroded connections. I have a feeling this gremlin lives in the harness somewhere.

DesktopDave · « **Reply #11 on:** March 06, 2012, 06:32:26 PM »

You checked all the obvious stuff right? Car timed right, fuel pressure high enough, CO harness disconnected at the firewall?

Wrench · « **Reply #12 on:** March 06, 2012, 08:18:40 PM »

Yep Fuel pressure is OK, new pump... connected gauge while it was trying to run and it had a nice steady pressure within spec.

It ran OK for a time with the new ECU with the old O2 sensor disconnected. The next day same thing so I think I have bad coils and they fried the drivers in the ECU. Ordered new coils today.

What CO harness at teh fire wall are you talking about??

Wrench · « **Reply #13 on:** March 07, 2012, 08:02:43 AM »

OK this is strange... started the car here at the shop this morning. Same thing... idles ok but would not rev above 2K. Left it to idle for a while... 20 minutes later went to check on it. It had stalled. I started it and guess what.. it revs fine now all the way to 6K with no misfires!?? WTF!?
It's cool that it runs OK but for how long and why? I hate gremlins like this one.

DesktopDave · « **Reply #14 on:** March 07, 2012, 10:25:58 AM »

It's more of a hot start problem...but it's worth checking out. Thread is here.

There's a CO potentiometer in the AFM that's enabled in non-cat Euro models (it's a lot like the old LH-Jet setup). Accidentally connecting that wire on "green" O2/catalyzed cars makes them run poorly.