S38 No Start - Solved

[MShimon]

Group, I'm out here auto crossing the '88 M5 and during a hard corner the power was cutting in and out. I believe that was because the fuel level was too low and the pump wasn't picking up. After the run, I was putting back to my spot and the car shut off. I have tried restarting it to no avail. The engine turns over hard and catches somewhat for a split second and nothing. I do not know the condition or age of the fuel pump. When jumping the fuel pump relay, there is fuel at the fuel rail. I haven't checked the pressure yet, but it looks to be about the same outflow velocity as the 528e. I have swapped main and fuel pump relays as well as the fuel pump fuse. I was also thinking that maybe I'm just plain out of gas, so I stuck a dowel into the tank. It came up with a believable level of fuel as compared to gauge. Any help is very appreciated.

[m535is]

Solder joints in the ECU, AFM, or really bad vacuum leak.

[mooseheadm5]

If it fires and then dies repeatedly, you should consider that it could be a bad engine speed sensor. It fires the first time from the TDC sensor, but must see a good speed signal to continue running. You can scope it or swap it with a known good one if nothing else works.

[m535is]

That is what I was thinking but that wouldn't explain the issues while cornering unless it really was a fuel issue.

[MShimon]

I was also thinking engine speed sensor. I have not tested this sensor yet, but it does not even fire initially. For a previous issue, which I have not solved, I did swap ECUs with Moosehead. This swap did not change anything which leads me to believe that the ECU is not at fault. I have opened up the AFM to do adjust the wiper arm on the track and did not notice anything awry then. A massive vacuum leak might be a culprit that I will look into as well. I would say that it is cranking like it is low on fuel though. It will quicken suddenly like it is going to start, but not. I would also like to note that I replaced the fuel filter that week.

[mooseheadm5]

There are not many places to develop a large vacuum leak, so it will be easy to test that possibility. Check the boots, the idle hoses, and the PCV hose.

[m535is]

It almost sounds a little like a cap and rotor, how old are they?

[MShimon]

It almost sounds a little like a cap and rotor, how old are they?

You sir, I believe, get the prize. I pulled those tonight and it appears that the new cap and rotor that I installed this summer has an interference issue. The cam bucket that's on the car right now has had some machining done to it, for sure, on the front of it for a crack. I think this resulted in the shim that was placed between the bucket and the mounting cover. I did not reuse this shim because I thought it was the cause of an oil leak. I now think that this shim should be there so that the rotor does not bottom out on the cap; as has apparently been happening. I will report back tomorrow when I get everything back together again to see if this is the real issue.

Thanks guys for the replies. It seems I keep getting into a jam with this car...

[Randomg]

It sounds like you shouldn't take parts off when you aren't sure what they do if you ask me... I know your pain though. My experience is that if you hold on, the car will become reliable again after you put in some work.

[MShimon]

I've been busy lately and haven't had time to work on it, but I have tested the fuel system and it works fine. The ignition system, however, is not so fine. Everything is in good condition except there is no spark off of the coil. There is continuity between the ECU and the coil. There is also continuity between all of the contacts on the cap and also all of the spark plug wires. The ECU was reconditioned, it had a few bad solder joints. I have taken off the speed and reference sensors and am now looking for the pin. I'm not exactly sure what I'm looking for, just a protruding pin? Previously I had to check the sensors and they were in working order as per Bentley, so there is basically one part left right?

[Philo]

Jeremy.., any updates. A corning loss of power has to be electrical ..... unless you've got a carburetor on that thin

* Ck connector to alarm system in the glove box, it a white 4 plug.
* Ck the relays ( I know you have already....).
* Ck the battery power lead to the ECU.., the red heavy gauge wire in the glove box, there's a large single pin connector that was corroded on mine.
* Ck fuses to fuel pump.., if the are moving slightly this could be an issue. I have to bend my fuse tabs to make them tight every six months or so.
* The coil could have an internal short that cuts out when you hit bumps or corner..., you need to do a swap with a good used one. I have one if you need.
* Ck the coil wiring to make sure they are installed correctly, the Bently manual have a picture showing this. It's designed to eliminate shorts if the wires move over time.
* Ck grounds..

Hope this helps.., just thinking over morning coffee...

[MShimon]

Philo,
Thanks for all of the helpful suggestions.

* Ck connector to alarm system in the glove box, it a white 4 plug.

*I'm not exactly sure how or what to check on this, but the ECU is getting battery voltage at pins 18 and 35.

* Ck the relays ( I know you have already....).

* They check out good.

* Ck the battery power lead to the ECU.., the red heavy gauge wire in the glove box, there's a large single pin connector that was corroded on mine.

* Battery power is being received at the ECU connector.

* Ck fuses to fuel pump.., if the are moving slightly this could be an issue. I have to bend my fuse tabs to make them tight every six months or so.

* Holders are solid with new fuses.

* The coil could have an internal short that cuts out when you hit bumps or corner..., you need to do a swap with a good used one. I have one if you need.

* We have an eta that I DD and have swapped with that coil a few times already.

* Ck the coil wiring to make sure they are installed correctly, the Bently manual have a picture showing this. It's designed to eliminate shorts if the wires move over time.

* The coil wiring is placed correctly as per Bentley.

* Ck grounds..

* I have gone through and cleaned up all of the grounds.
I have also checked the ground pins for the ECU, pins 10,16,17 and 19 according to the ETM and they all have continuity. I have check continuity from the reference sensors to the ECU, good. I've rechecked the amperes for the speed and reference sensors as per Bentley manual, they are all within spec. This however does not validate good and working sensors. Doug, Photopark, reconditioned my ECU and has been giving me tips to try. I had a notion that perhaps the raised pin for the reference sensor was missing since no spark is registering. The starter was taken off to get a good view and the raised square pin is intact. Doug sent me this photo to verify the proper arrangement. I'm not sure what the little hole directly below the square pin is for though, I couldn't feel anything behind my square pin.

Mine looks like this as far as I can tell. The coil checks out as per Bentley and receives 12v. We've checked continuity from the coil to the ECU, good. I tried swapping the sensors to see if perhaps one was bad, no start still.

I came across a thread about the OBC relay being bad, which will then disable the ignition is the vehicle security code is enabled (it's not in my car). Thread here. I haven't check this relay yet, but it seems that if I'm getting power to the ECU that this isn't the issue. I have also undone the negative battery cable for a long time to reset the OBC code.
I'm still at a loss as to what could be happening which would not provide spark from the coil. I should also mention that we checked the spark plug wires, cap and rotor which all had values within the Bentley's range.
Just to reiterate, the vehicle was cornering and I suddenly lost engine power, but it came back on when I pushed in the clutch. It then was intermittent for a little while longer and then quit completely with no refiring at all.

[johne]

Instead of replacing the relay you can connect the solid green wire and the green wire with a red tracer in the wiring harness going to the relay effectively bypassing that problem. My source at Century BMW in Greenville SC always does that whenever he purchases an E28/E30. I just did this on my M5.
FWIW John

[Philo]

Lets go back to basics.. If you have fuel pressure and an injector pulse and you have spark at the plugs, it should at the very least pop... Take a plug out and verify again that you have spark. At the same time turn over the engine with the plug out and you should get a strong whiff of fuel, which means you have an injector pulse.

It's gotta be one of three things to get the engine to pop and stumble. Air-Fuel-Spark. Let's work on getting it to run later. My point is.. even if there is something else wrong with the fuel injection system it should still at least pop. If we have fuel and spark and let's say the timing is way off it will pop and probably backfire. out the intake or exhaust. If it just cranks I don't think we are getting spark.

I have a good ECU if you want to try it...

[photopark]

Lets go back to basics.. If you have fuel pressure and an injector pulse and you have spark at the plugs, it should at the very least pop... Take a plug out and verify again that you have spark. At the same time turn over the engine with the plug out and you should get a strong whiff of fuel, which means you have an injector pulse.

It's gotta be one of three things to get the engine to pop and stumble. Air-Fuel-Spark. Let's work on getting it to run later. My point is.. even if there is something else wrong with the fuel injection system it should still at least pop. If we have fuel and spark and let's say the timing is way off it will pop and probably backfire. out the intake or exhaust. If it just cranks I don't think we are getting spark.

I have a good ECU if you want to try it...

This ECU was verified on my 88 M6 and ran great, it immediately started up after reflowing the solder joints on the injector transistor.

[MShimon]

OK! So I finally got around to messing with the car again. After double checking all the ignition components and grounds, I decided to look into the OBC security system since I wasn't getting any spark from the coil. I jumped the rd/gn and gn wires on the connector and viola! The car started up after a few tries. I believe the jumper wire isn't a very good connection so it took some fenagling, but it works!

[tn535i]

Somewhere behind the cluster is a pair of connectors that splice in the obc code relay. If you find the two pairs it's obvious where the alarm relay is patched in. I like to put them back as if the relay was never installed like on lesser equiped cars. One less set of connections and relays to worry about.

Glad you found it and hope that's the last problem(for a long while).

And back to the rotor. I've found the most recent Bosch rotors to be ill fitting and cause me trouble. One has bit me each on the M30 and the S38. On the M30 we had to remove the cam adapter to clean the broken bolts out of the threaded holes. On the S38 they just simply wiggled loose and eventually the rotor crashed into the cap. The early symptoms were so much like you described. Sudden power loss and then right back to life again.

My thinking is that Bosch has outsourced the screws now to some very poor quality source. In 20+ years of BMW motoring on similar engines I've never had this problem until the last couple years. I've switched to Bremi and all is well.

[MShimon]

Glad you found it and hope that's the last problem(for a long while).
Sudden power loss and then right back to life again.

This is what I'm experiencing right now as I took it for a test drive. I'll check out the rotor today.

[MShimon]

The car died again today on the test drive. It was running fine and would intermittently cut out power either at a steady speed or under acceleration. The last time it stumbled and then just briskly died while I was trying to get up an incline to my house. I don't believe it to be a cap and rotor issue, both were very secure and seemed like they would make good connections. There must be some kind of wiring fault somewhere. It's exhibiting the same symptoms as when it died in the first place. Stuck yet again.

[MShimon]

Ok, I believe that the in tank transfer pump is faulty as the vehicle will start and drive until I try to accelerate at anything more than a normal pace. The vehicle will suddenly lose power (as has been happening) and buck then catch itself again. Sometimes it would just shut off altogether. I'm gonna get a good pump in the tank tomorrow, hopefully it'll work.

[MShimon]

The new transfer pump seems to have helped with starting, but alas my issue is still around. I noticed that as I was accelerating, the car will always cut out at around 4000 - 4500 rpm. It isn't noticeable when the car is stopped and I'm just free revving. I still think it's a fuel delivery issue, I'm gonna do some pressure testing this weekend. I'm gonna test the FPR and also look at the pressures while driving, with a passenger of course. Any other ideas? I welcome everything.

[Philo]

hmmm.. 4500 rpm with light, medium or heavy throttle (load) ?

[Karl Grau]

I still think it's a fuel delivery issue,

How was the screen on the transfer pump when you replaced it, any sludge? Do you when your fuel filter was changed last?

[MShimon]

hmmm.. 4500 rpm with light, medium or heavy throttle (load) ?

From what I can recall right now, it was under moderate throttle and definitely heavy throttle. I'm not positive on the light condition, I'll try that tomorrow.

I still think it's a fuel delivery issue,

How was the screen on the transfer pump when you replaced it, any sludge? Do you when your fuel filter was changed last?

The screen wasn't bad, it was certainly old and discolored though. The fuel filter was changed about a week before the car had these issues. It was a brand new filter from BavAuto I believe. Perhaps it was defective?

[Philo]

Would a filter installed in reverse cause a restricted flow ?

Also.., when the engine starts to crap out at 4500, does the engine loose power smoothly and then when you let up on the pedal it comes back to life ? Or does it start jerking and bucking... If we are thinking fuel delivery, first is overly rich and second is lean. But if it's rich enough to kill power you'd be able to smell it !

I'm thinking lean, poor pressure at the injector rail. Look at Bentleys, there's a pressure test you can do. Just take the gauge to your windshield and go for a drive. The neighbors will love it !

[MicahO]

Would a filter installed in reverse cause a restricted flow ?

When I got my car and ran through tune-up items, the fuel filter that was on it was installed in reverse, and had a date-sticker from 1999. It had been like that a long time (though only about 15K miles). Didn't seem to cause a problem, though it caused me many "what-the?" moments while I was lying on the floor looking up at.

[MShimon]

Also.., when the engine starts to crap out at 4500, does the engine loose power smoothly and then when you let up on the pedal it comes back to life ? Or does it start jerking and bucking... If we are thinking fuel delivery, first is overly rich and second is lean. But if it's rich enough to kill power you'd be able to smell it !

I'm thinking lean, poor pressure at the injector rail. Look at Bentleys, there's a pressure test you can do. Just tape the gauge to your windshield and go for a drive. The neighbors will love it !

The engine most definitely jerks and bucks with immediate power loss. Also, before this ordeal, I've been battling a rich running condition. The O2 sensor will not properly correct the mixture, I've tried several new Bosch sensors and even had an old BMW indy shop try to figure it out with no dice. Needless to say, I've been running without the O2 connected for about a year now. However, I'm also leaning towards poor pressure at the rail as the rpm increases.

[mooseheadm5]

Get a fuel pressure gauge on it. Mine needed a new FPR because it would intermittently stick and the car would run like crap.

[MShimon]

Got the gauge hooked up, pressure is steady at about 41 psi across the rev range until I hit about 4500 to 5000 rpm. At this point the engine will cut out consistently. This happens either driving or out of gear in the garage(no load). When I slowly clamped off the return line from the FPR the pressure shot up to about 60 psi, Bentley says this indicates a faulty one. However, this cut out is so consistent it almost seems like spark is being stopped, almost like a rev-limiter. I have a stock chip in the ECU, so nothing with the programming should kill it. Perhaps a sensor is telling the computer to cut off spark or fuel?

I wonder if the timing has somehow been knocked way off, it might explain the terrible gas mileage. The crank sensor resistance readings checked out per Bentley. This also might explain the erratic behavior of my tachometer. I've visually inspected the flywheel and it looks good to me, but it's the only one I've seen. Here are some photos of what I think is the raised pin for the crankshaft position.

[Philo]

Lookie what I found... You HAVE to spend time on S14.Net. A wealth of info that directly applies to the S38.

http://www.s14.net/forums/showthread.php?t=33022