S38B35 long hard start - tech question

[HealeyBN7]

I will have to research the OBC cut off relay referenced by Dean. I didn't look for damage to the reference pin, or if it was broken, while swapping the sensors. I assume Dean that you mean the pin on the flywheel?

Yes. You can look inside the hole where the sensor mounts. You can use a cell phone camera and rotate the flywheel while you look for a pin. The m5 flywheel is actually a notched lug. The other sensor looks at the ring gear. Unless all of your teeth are missing you can ignore that one.

Please do search for the OBC cutoff relay. There is a way to bypass it. I didn't do that. I bought a new relay and it fixed my problem.

Dean

[Mab1957]

Well I received the used S38 reference sensors from Scott and both tested good in the 1040 range. I installed them but still no spark and still no vibration of the cleaned and reconnected original ICV. I swapped in another ICV that I have and it doesn't vibrate either.

Just checked and the M5 has a slightly different current path for the ICV. It's powered by the fuel pump relay and therefore it will not turn on with the ignition on. The engine needs to be cranking or running. To test this jump the fuel pump relay (pins 87 & 30) and see if the ICV now vibrates.

I suggest you AC voltage test the crank sensors now.

I didn't see this anywhere in your thread but did you electrically test the main and fuel pump relays?

I don't have this experience, but am willing to learn.
First - What resource are you using to "check"? The ETK from "a"'s signature http://wedophones.com/BMWManualsLead.htm 1988 635csi-M6 ?
Second - I'm guessing the voltage test would be to measure voltage at the loom side of the crank side of the sensor connectors on the firewall cover? I didn't see that test in the Bentley. Which pin(s)? I'll guess battery voltage at pin one. I see again in the ETK from "a"'s signature http://wedophones.com/BMWManualsLead.htm a section on Troubleshooting and Diagnosis, section 1210.4a includes the S38. Is that where these tests would be explained, so I am not guessing?
Third - I did not "electrically test" the main and fuel pump relays. I just swapped in new ones and when it didn't make a difference I pulled out the new ones and put in the old ones. What is "electrically test", looking for voltage when the key is turned?
Fourth - I'll look up a fuel pump relay (pins 87 & 30) diagram and (confirm this test please) insert a spade jumper wire to the loom connector at those two pins.

Thank you again for your help gents.

[demetk]

Download the 1987-1988 BMW 535i - M5 Electrical Troubleshooting manual. This one has the M5 wiring. Go to section Injections Electronics, page 1362-2 to 1362-3 and you'll see the ICV power coming off pin 87 of the fuel pump relay.

The AC voltage test of the sensors is done on the same pins as the resistance test. Simply set your meter to AC volts and then crank the engine. 4 volts AC for the speed sensor and .4 volts AC from the ref sensor come to mind for what you should be seeing. If you don't get any voltage from the ref sensor then it's time to verify that the ref pin on the flywheel is still there.

Edit: Forgot to mention what you need to test on the main & fuel pump relays. For the main relay confirm 12 volts coming off both 87 pins when the ignition is on. For the fuel pump relay confirm 12 volts is coming off pin 87 when the engine is cranked. I find a test light to be more useful as a test device here since a bright test light indicates both good voltage and good current flowing where as the multimeter will show you voltage but it can't tell you anything about the current.

cheers

[Mab1957]

Thank you again. I didn't see the M5 at the end of that one, and was looking in 1988 section. I'll read it now. I have a test light and can follow your lead. How useful and available is a diagnostic tool that can insert our diagnostic plug?

[demetk]

How useful and available is a diagnostic tool that can insert our diagnostic plug?

Pretty useless. OBD I cars (like ours) don't provide any useful information if any. You pretty much have to old school it on the e28.

[Mab1957]

Loom side with probes in 1 and 3 of the sensors. Each tested separately. My son turned the key to crank the motor over. I get no sensible readings with DMM on Vac. On Vdc I get .04 on the reference and .02 or .03 on the speed.

Loom side of the Main relay (relay is removed) the 30 pin is hot. [EDIT - 86 pin is hot, 30 is not] Neither 87 or 87b show voltage with the ignition switch in the on position or when cranking.

Loom side of the Fuel Pump relay (relay is removed and jumper between 30 and 87) the ICV does not activate when the ignition switch is put in the on position or when cranking.

Is the OBC relay over under the steering column under dash panel same as the e28 M30?

[gmb49all]

Loom side of the Main relay (relay is removed) the 30 pin is hot. Neither 87 or 87b show voltage with the ignition switch in the on position or when cranking.

Correct me if I'm wrong, but don't you need the relay in to see voltage at 87 or 87b, hence the purpose of the relay?

[gmb49all]

Fourth - I'll look up a fuel pump relay (pins 87 & 30) diagram and (confirm this test please) insert a spade jumper wire to the loom connector at those two pins.

Jumping with a spade connector was the most helpful during my no start. If you can start it with the voltage jumped, I think it rules out everything ignition related after the main relay (wires, coils, spark plugs, etc). Could be the speed and reference (especially if they are swapped, hate those wires, thank god for 1.3). I also think it calls into suspect the main relay, or something else upstream (OBC relay although I have never had a problem with mine), possibly bad ECU solder joints (but on more than one unit? doubtful) or ignition switch itself?

Another suggestion that helped me was to test the voltages AT the ECU. If you know what pins are supposed to have voltage, and one does not, that should clearly identify the problem. Kind of pain backprobing the ECU, but not impossible (I used some safety pins). Not sure how different the s38 is from the M30, but doing that led me to the main relay, which was my issue. My no start was intermittent as well.

[demetk]

Loom side of the Main relay (relay is removed) the 30 pin is hot. Neither 87 or 87b show voltage with the ignition switch in the on position or when cranking.

Correct me if I'm wrong, but don't you need the relay in to see voltage at 87 or 87b, hence the purpose of the relay?

Yup. You measure voltage from underneath the relay holder with the relay installed.

[ahab]

Is the OBC relay over under the steering column under dash panel same as the e28 M30?

Yes, and as you may know it does not look like a relay but it's a black relay "box" in where the chime is.

You can easily jump out the fuel pump relay by connecting a lead to the two terminals in the green circles. As soon as you connect them you should hear the pumps whir and fuel traveling through the rail. This is more of a test for the pumps themselves but it will bypass the relay too.



Here is a picture of what you are looking for on the flywheel.

[Mab1957]

Thank you again, gents.

So I've tried the Main Relay voltage test from behind the relay connector with a Main Relay in place. There is no voltage getting to the 87's. I thought I might be doing something improperly because when I test the relay in one of my other e28's it works, but I've tried the same test light in the other e28 and 87 lights up. I take one of the working Main Relays out of my other running e28s and I don't get voltage in the M5 on that test.

Thanks for the pics Ahab, I did that fuse box test and the fuel pumps work, I hear the fuel moving in the rail. The ICV doesn't vibrate.

I swapped in another OBC relay with no change. I then jumped the green wires of the OBC relay connector with no change. I swapped in my 3rd ECU with no change.

At this point I am going to re-install the original ECU, original OBC relay and original Main and Fuel Pump relays (which worked in the other e28). I will have it towed to a shop, put it up on a lift and try visual confirmation that the flywheel pin is good. Back to the diagrams...

[gmb49all]

Couldn't you just jack it up and look? If it were me, I'd rather do that than paying towing possibly to and from a garage...

[demetk]

There is no voltage getting to the 87's.

Trace the power on that green wire from the OBC relay to C103 and then to the main relay.

Just for giggles run a jumper wire from the battery positive to pin 86 at the main relay. Does the car start?

[demetk]

And as long as you're not getting the main relay to power up there is no reason to check the reference pin at this point.

[gmb49all]

Stupid question... is the main relay on an s38 the same as on an m30b34? Has anyone compared the schematics on the side of them?

[Mab1957]

Couldn't you just jack it up and look? If it were me, I'd rather do that than paying towing possibly to and from a garage...

Maybe. That is what I did when I replaced the reference sensors. It is on uneven ground on the street in front of my house so I'm anxious even with it chocked and blocked and a warning triangle out.

There is no voltage getting to the 87's.

Trace the power on that green wire from the OBC relay to C103 and then to the main relay.

Just for giggles run a jumper wire from the battery positive to pin 86 at the main relay. Does the car start?

I'll try it.

Stupid question... is the main relay on an s38 the same as on an m30b34? Has anyone compared the schematics on the side of them?

I worry that all my questions appear in that category. You question isn't. I'll compare the few that I have. I found only one option listed on RealOEM for the M5 (#8), but that's hardly a definitive technical analysis. #10 looks like it'd work, but if #11 is used as a replacement then there would be an issue. I'd guess that's why there are protrusions depicted in the casings of #9, #10 and #11, to distinguish them.



The one on the left is the original out of the M5, center a replacement spare from parts supplier I ordered for an M30B34, right an old spare from my parts bin, also M30B34.

RealOEM for 535i
08 DIODE RELAY SI537 04/86 ? 12631708646 ENDED
08 DIODE RELAY ? 12631708647

RealOEM for M5
08 DIODE RELAY SI537 04/86 ? 12631708646 ENDED
08 DIODE RELAY ? 12631708647

[demetk]

You could test the original main relay and see if it works. Apply power and ground to pins 86 & 85. Does it click?

And based on the wire colors at the relay socket you can map out the pin numbers. That way you can conclude if the replacements you have fit the socket pin for pin.

[Mab1957]

Loom side with probes in 1 and 3 of the sensors. Each tested separately. My son turned the key to crank the motor over. I get no sensible readings with DMM on Vac. On Vdc I get .04 on the reference and .02 or .03 on the speed.

Loom side of the Main relay (relay is removed) the 30 pin is hot. [EDIT - 86 pin is hot, 30 is not] Neither 87 or 87b show voltage with the ignition switch in the on position or when cranking.

Loom side of the Fuel Pump relay (relay is removed and jumper between 30 and 87) the ICV does not activate when the ignition switch is put in the on position or when cranking.

Is the OBC relay over under the steering column under dash panel same as the e28 M30?

In this morning's light I observed that the test I did the other night was reported incorrectly. The Main Relay pin 86 is hot, pin 30 is not. When cranking 87s do not go hot. So I back track and edit:

Loom side of the Main relay (relay is removed) the 30 pin is hot. Neither 87 or 87b show voltage with the ignition switch in the on position or when cranking.

Correct me if I'm wrong, but don't you need the relay in to see voltage at 87 or 87b, hence the purpose of the relay?

Yup. You measure voltage from underneath the relay holder with the relay installed.

With Main Relay in the 86 pin is hot, 30 pin is not hot. 30 pin nor the 87s get hot when cranking but I feel the relay clicking. When I jump the 30 pin to the battery (just for giggles ) it starts! So I'm back to the diagram to see where 4 RD C225 are. [EDIT - page 7000-11 Component views, C225 is behind the ECU. I'll check the battery cable clamp first] Must have some power supply issues there

[demetk]

Pin 30 is a direct battery connection. You are close now.

[demetk]

ETM says C225 - Under RH side of dash, above glove box.

[Mab1957]

ETM says C225 - Under RH side of dash, above glove box.

Found it in the ETM, thank you so much. Going out to the car again now.

[gmb49all]

I know it's on the street now, is it parked inside or outside? I know my sunroof on my 535 doesn't like to work on mornings that she's outside, I believe because condensation is seeping through my sunroof seal that needs to be replaced. Maybe your windshield gasket is seeping?

[Mab1957]

I started chasing the voltage, so pulled the glove box to get to the M5 battery lead that powers the Main Relay (brown one on the top of the bracket is the 3rd one I tried - they're all good ).

On the M5 the battery cables run under the back seat, one splits off under the car below the driver's seat and runs up to a power terminal forward of the A pillar while the other runs up the passenger side of the transmission tunnel and behind the glove box (that red one behind the AC line and harness) to a connector on the engine harness (white connector on the right):


It feeds pin 30 on the main relay, so since I don't have good voltage at Pin 30 I start here with a cleaning, some dielectric and confirm solid connection. I get 12.3+ volts at the connector. The light is weak.

Pin 30 is getting 12.3+ volts and the light isn't really bright. The car still doesn't start without a jumper between Pin 30 and the battery.

I don't see any break, moisture or issues for as far as I can reach behind the carpet at the firewall. So I go to the other end and clean the connector at the terminal.


It isn't bad, but a bit rusty/dirty/greasy so I clean it with a bit of white vinegar and some 180 grit paper and a brass brush. When reassembled there has to be good connectivity but no change in performance (weak light persists). I figure I'll start at the battery and work my way forward before pulling out the interior to inspect the length of the cable, it is wrapped after all .

The cable is shrink wrapped and has a flat spot just outside the battery box and that doesn't seem right to me.

Anyone else seen this on their car? The wrap is very old and the cables match closely in braided insulation and gauge. The terminal ends of the cable and extension are riveted to a piece of aluminum. Corrosion seems to have taken a toll on this connection.

When I jump beyond the connector the car starts and runs.

With the jumper I get a bright light at Pin 30. I sure wish I'd found this before performing all of those other activities! I offer my sincere thanks to everyone above who offered suggestions and provided guidance. I moved this M5 back into the garage. I plan to pull the seats soon so will [EDIT - NOT get under the carpet, loosen all the cable brackets and see if I can adjust the power lead to reach the battery as originally intended (without the extension) - BUT DO AS demtk SUGGESTS BELOW AND REPLACE THE FUSELINK]. This should be a simple fix now and in the process I've learned so much about the e28 m5 that I actually feel good about it.

[demetk]

Nice work. Does it look like a fusible link? If it is, it's odd that the ETM doesn't show one for the M5.

[Mab1957]

Nice work. Does it look like a fusible link? If it is, it's odd that the ETM doesn't show one for the M5.

Sorry that picture is so poor. The whitish flap pictured on the top of the connector is a thin piece of metal that had a sort of a insulator film on it and it is only attached to one of the rivets now that I've unwrapped and moved it. The metal connector appears solid though and I didn't look closely to see if the rivets are insulated from it. As I'm sure you've surmised, I've limited experience with electronics. I can't guess what the purpose of that flap may have been. Would that describe a fusible link?

[demetk]

Sorry that picture is so poor. The whitish flap pictured on the top of the connector is a thin piece of metal that had a sort of a insulator film on it and it is only attached to one of the rivets now that I've unwrapped and moved it. The metal connector appears solid though and I didn't look closely to see if the rivets are insulated from it. As I'm sure you've surmised, I've limited experience with electronics. I can't guess what the purpose of that flap may have been. Would that describe a fusible link?

Sounds like it. This is a blown fusible link on an e34.



Considering the connection is at the battery and runs the length of the car, it may be prudent to fuse it again. You just have to figure out what amperage to use. I would guess a max of 30-40 amps.

[Mab1957]

Ahh, I see! Likely what I observed isn't metal. Thanks again demetk, you know your stuff!! Time for more research.

[gmb49all]

That's just silly. If it were me, I'd just put an inline fuse in and be done with it.

[Mab1957]

Nice work. Does it look like a fusible link? If it is, it's odd that the ETM doesn't show one for the M5.

That's just silly. If it were me, I'd just put an inline fuse in and be done with it.

Doesn't show in RealOEM or on the ETM. I suppose these rivets could be drilled out an another fusible link riveted back in, but an in-line fuse would be ideal for future diagnosis and replacement. Then I need to figure out why this link failed. From E28 diagram:

I did find a discussion on this at Spec E30
with a RealOEM link
It shows 50amp fuse #14 PN 12411706111 applicable to e30 and e36 including M3.

[demetk]

Then I need to figure out why this link failed.

On my 89 e34 both fusible links had visible cracks in them. They didn't blow, they just cracked presumably from age and vibration. It looks like these old style fusible links are susceptible to cracking.

In your situation an 8 or 10 gauge 40 amp inline fuse holder would work fine. Just cut the existing wire to clean wire and then solder it in and then go out and enjoy your ride.