Jump to content
soorooshg

3.6 M5 Alternator Trouble

Recommended Posts

Hi all,

Apologies in advance for the long post but I’ve had an ongoing problem with my car now that’s driving me crazy. So a while ago my battery kept dying overnight, I would have to jump start the car pretty much every day. There was a drain problem in the past but that was resolved, I checked the drain with everything off and it was around 0.05amps. When the engine is started, the alternator is only providing around 8-9volts, all electrics are very weak as you can imagine. Along with this, the ABS light on the dash stays lit. Once the engine revs hit 4k, everything comes back to life and the voltage jumps back to 14v as normal. This happens every time the car is started. Everything pointed towards a faulty regulator. I pulled the alternator out, and replaced the regulator with a new Bosch unit. While it was out I got the alternator bench tested and it came back all fine. I re fitted the alt with the new regulator but I still have the same problem. A few people on Facebook and some old forum posts told me to check the battery light on the dash as that has something to do with exciting the alternator. I pulled the cluster and the battery bulb is working fine, but there’s no power going to it. I also grounded the small cable on the back of the alternator (the exciter wire) to the same ground point (strut tower nut) that I use to jump the car, but still no battery light and still 8v on startup. I’m quite baffled at what this could be and was hoping someone here could help me out. 

Many thanks in advance

Share this post


Link to post
Share on other sites

In some cars, the voltage to the alternator used to go through the bulb. So if the bulb went, the alternator would not work. The E34 has a resistor in parallel with the bulb so even if the bulb went, the D+ terminal would still receive power.  The power to the bulb comes from fuse 17. It goes via the unloader relay to D+ terminal of the alternator. That terminal is negative when you turn the engine on so the bulb lights up.  When the alternator is charging, the alternator changes that so the bulb goes out. 

 

You don't mention the year of the car but I am pretty certain they all work in the same way. Diagrams are here:

http://www.wedophones.com/BMWManualsLead.htm

Share this post


Link to post
Share on other sites
17 hours ago, whiskychaser said:

In some cars, the voltage to the alternator used to go through the bulb. So if the bulb went, the alternator would not work. The E34 has a resistor in parallel with the bulb so even if the bulb went, the D+ terminal would still receive power.  The power to the bulb comes from fuse 17. It goes via the unloader relay to D+ terminal of the alternator. That terminal is negative when you turn the engine on so the bulb lights up.  When the alternator is charging, the alternator changes that so the bulb goes out. 

 

You don't mention the year of the car but I am pretty certain they all work in the same way. Diagrams are here:

http://www.wedophones.com/BMWManualsLead.htm

Sorry, mines a 1991. Thank you for the information, by the D+ terminal I assume you mean the smaller cable on the back of the alternator (exciter wire). I had a look at the link and on the diagram it shows 4 relays labelled as unloader relay in the main fuse box, all of which I’ve checked with a 9v battery and they all function correctly. I hate to be spoon fed but I’m quite a novice when it comes to electrics, what would you suggest I do next?

Share this post


Link to post
Share on other sites

I guessed '92 but they all work the same. Yes, that is the wire I meant. Fuse 17 should be OK - the engine would not run without it. And all the unloader relay does it cuts off power to stuff like the heated screen so it is not dragging the voltage down when you fire up. Good to know that the relays work though.

Maybe the first thing is to confirm what the ignition warning light is doing. Does it come on with the ignition and go off when the engine starts? Or does it remain off or on when you switch on and start? If there is no power to the bulb, you would not expect it to light at all. But it is always best to have a clear picture. You will have to forgive me if my approach is pedestrian :)  

 

PS Have you also checked fuse 28?  

Edited by whiskychaser

Share this post


Link to post
Share on other sites
1 hour ago, whiskychaser said:

I guessed '92 but they all work the same. Yes, that is the wire I meant. Fuse 17 should be OK - the engine would not run without it. And all the unloader relay does it cuts off power to stuff like the heated screen so it is not dragging the voltage down when you fire up. Good to know that the relays work though.

Maybe the first thing is to confirm what the ignition warning light is doing. Does it come on with the ignition and go off when the engine starts? Or does it remain off or on when you switch on and start? If there is no power to the bulb, you would not expect it to light at all. But it is always best to have a clear picture. You will have to forgive me if my approach is pedestrian :)  

 

PS Have you also checked fuse 28?  

Pedestrian approach is preferred in my case as I want to check everything methodically and make sure I don’t miss anything obvious haha 

 

as for the battery light, no it doesn’t illuminate at all, whether just with ignition, or the engine on (which it shouldn’t anyway), honestly I’ve never actually paid attention to it so I can’t tell you if it used to turn on before this issue came up, but I assume it must have worked in the past when the alternator was working properly. As I said I took the cluster out and took the bulb out and the bulb itself is fine, hooked it up to a battery and it lights up. I haven’t checked fuse 28 specifically but I have looked at every fuse and they are all in working order 

Share this post


Link to post
Share on other sites

Looks like you have two issues - no voltage to the bulb and the alternator will not charge properly.

 

I'd start with the cluster. There is power arriving at it because other bulbs light up and it is a common live. I would disconnect the cluster and using a meter on ohms, track it back to the live in at pin 7

Share this post


Link to post
Share on other sites
1 hour ago, whiskychaser said:

Looks like you have two issues - no voltage to the bulb and the alternator will not charge properly.

 

I'd start with the cluster. There is power arriving at it because other bulbs light up and it is a common live. I would disconnect the cluster and using a meter on ohms, track it back to the live in at pin 7

I’ve been led to believe that no voltage to the bulb *is* causing the alternator to not charge properly, or am I mistaken?

 

and thank you, I’ll have to pull the cluster again tomorrow and test the resistance. By pin 7 which one are you referring to? There were 4 multi pin connectors at the back of the cluster 

Share this post


Link to post
Share on other sites

Even if there is no bulb in there, the voltage needs to go through the resistor. Power to the bulbs comes in on pin 7 of X16 which should be 26 pin yellow. The sockets usually have the pin numbers stamped in on them. If you can't see the numbers, pick another bulb socket and test for continuity to the one for the charging bulb. If you pick the wrong side (negative), you will just get no circuit or a resistance. 

 

I know voltage from the bulb socket/resistor goes via the unloader relay and that may affect the ABS but it is probably best to rule out one thing at once

 

PS. It just occurred to me that when you tested and found no voltage, you were using a good known ground for the black probe weren't you? No disrespect intended

Edited by whiskychaser

Share this post


Link to post
Share on other sites
54 minutes ago, whiskychaser said:

Even if there is no bulb in there, the voltage needs to go through the resistor. Power to the bulbs comes in on pin 7 of X16 which should be 26 pin yellow. The sockets usually have the pin numbers stamped in on them. If you can't see the numbers, pick another bulb socket and test for continuity to the one for the charging bulb. If you pick the wrong side (negative), you will just get no circuit or a resistance. 

 

I know voltage from the bulb socket/resistor goes via the unloader relay and that may affect the ABS but it is probably best to rule out one thing at once

 

PS. It just occurred to me that when you tested and found no voltage, you were using a good known ground for the black probe weren't you? No disrespect intended

Thank you for the specific details, I appreciate it, I'll give that a go tomorrow for sure.

 

If you're referring to testing the battery light, I never actually tested it for voltage but just arrived at that conclusion based on the bulb itself being ok, but not illuminating when the ignition is turned on. When I tried grounding the D+ wire on the back of the alt, I did to to the strut tower nut which is what I always use to jump start the car 

Share this post


Link to post
Share on other sites

Yes, I meant the battery light. But rather than take the cluster out, you could just remove the unloader relay K61 and check if you have a voltage at pin 4. 

 

Nuts on top of the strut tower are usually a really good ground :)

Share this post


Link to post
Share on other sites
20 hours ago, whiskychaser said:

Yes, I meant the battery light. But rather than take the cluster out, you could just remove the unloader relay K61 and check if you have a voltage at pin 4. 

 

Nuts on top of the strut tower are usually a really good ground :)

Yep just tested that getting 9.6volts at pin 4, although this is with a weak battery so I assume it should be closer to 12 with a charged batt

 

edit:

also pulled the cluster and I am getting the same 9.6v at pin 7 with the ignition on. When I turn the key to ‘on’ position (but no engine start as battery is weak) the voltage drops to 8v

Edited by soorooshg

Share this post


Link to post
Share on other sites

There has to be something wrong with the supply. A battery is deader than a Monty Python parrot at 10.5V. It would be worthwhile testing what the voltage is across the battery and at the B+ terminals. (There is the one where you jump start and another in the e-box)

Share this post


Link to post
Share on other sites
37 minutes ago, whiskychaser said:

There has to be something wrong with the supply. A battery is deader than a Monty Python parrot at 10.5V. It would be worthwhile testing what the voltage is across the battery and at the B+ terminals. (There is the one where you jump start and another in the e-box)

Sorry I’m not following, if you mean to test the battery voltage then it’s currently at 10.6, it does need replacing at some point 

Share this post


Link to post
Share on other sites

Yes, I meant the battery voltage across its terminals because it struck me that you may be losing power between the battery and the B+ terminals. That is often caused by a failure of a fusible link. But you only have 10.5v to start with.

 

Most battery tables show 10.5V as being zero charge so there is not a lot left in it. Ideally, it should be 12.6v. I would be surprised if you could get the dash bulbs to light with only 9.6v. Will it charge up? 

Share this post


Link to post
Share on other sites
20 hours ago, whiskychaser said:

Yes, I meant the battery voltage across its terminals because it struck me that you may be losing power between the battery and the B+ terminals. That is often caused by a failure of a fusible link. But you only have 10.5v to start with.

 

Most battery tables show 10.5V as being zero charge so there is not a lot left in it. Ideally, it should be 12.6v. I would be surprised if you could get the dash bulbs to light with only 9.6v. Will it charge up? 

Yeah the battery issue is a separate one which I think could be partially related to the alternator, and partially because of short journeys and the fact that it’s died probably 50+ times in the past year. Strangely though I had it tested the other day and the guy said it’s all fine and doesn’t need replacing. If I leave it on charge overnight it does reach 12.6v but 2-3 engine starts later (with about 30-60mins driving time) and it’s back down to 10v and no start again 

Share this post


Link to post
Share on other sites

If the battery has been discharged that many times it may be shot. It may read fine when it has just been charged but it is best to let it stand for a few hours and check if it is still reading 12.6v.

 

Perhaps charge it up and continue testing in the direction of the alternator? There is a diode in the relay and another further down which seems to be only present on the M5. I'd certainly be giving them a close look

 

Apologies if this seems to be a little pedestrian but I am not one for throwing bits at cars    

Share this post


Link to post
Share on other sites
On 30 April 2020 at 18:19, whiskychaser said:

If the battery has been discharged that many times it may be shot. It may read fine when it has just been charged but it is best to let it stand for a few hours and check if it is still reading 12.6v.

 

Perhaps charge it up and continue testing in the direction of the alternator? There is a diode in the relay and another further down which seems to be only present on the M5. I'd certainly be giving them a close look

 

Apologies if this seems to be a little pedestrian but I am not one for throwing bits at cars    

Ah that's a good point I didn't really think of that tbh, I've never tested it that way but yeah it most probably wouldn't hold more than 12v for very long I assume.

 

In regards to the diode you mention, could you be a little more specific please? What exactly am I looking for?

 

Also I did a few more tests yesterday which ended up confusing me even more.

 

1. Tested pin 4 on the K61 unloader relay, and there is voltage there with the ignition on

 

2. Pulled the cluster, there is voltage at pin 7 on the yellow 26 pin, for the battery light

 

3. Connected the battery light directly to pin 7 with a spare piece of wire, turned ignition on and the light illuminated, then I started the engine but still same issue, low voltage until revved up, which is the strangest part of it all since I understand that if the light turns on, the alternator is excited and should be running correctly

 

One thing I forgot to do is to plug the cluster in, take out the battery light & holder, turn ignition on and see if there's voltage on the contacts of where the light is mounted at the back of the cluster. I take it if there is voltage in this situation I should start looking at the cluster? Since there is power going right up to the pin so if theres a broken circuit somewhere it must be inside the cluster? Very confusing situation here, and as I previously said I prefer the pedestrian approach as you say it's not ideal to go running round and replacing every part you suspect might be the issue. End up with a 4 figure bill and still have the original problem. Also no disrespect intended, are you 100% certain that pin no.7 is the one for the battery light? 

Edited by soorooshg

Share this post


Link to post
Share on other sites

Diodes only let voltage pass in the direction of the arrow. So if you put the meter on one way round, you will get continuity. But if you swap the probes round, you should get a reading that is so high it is almost infinity. There is a slight complication in that to test them, they should be out of the circuit. That is fine for the one in the unloader relay as you can take it out. Not sure about the one on the nearside inner wing as I have not seen one. But if the wire is disconnected from the alternator, you should be able to test in the same way.

 

Can't say I am 100% certain about pin 7 because it is a while since I worked on a cluster. But I think I would remember if the diagram were wrong. That pin supplies most of the bulbs in the cluster. So you could pick another one from the diagram and check for a common live. In your test, did you bridge from the warning light socket via the bulb to pin 7? And if so, the live side or the negative side?

 

The ignition warning light should be on when you turn on the ignition. When you start the car and it is charging, the alternator turns the light off.  

Share this post


Link to post
Share on other sites
39 minutes ago, whiskychaser said:

Diodes only let voltage pass in the direction of the arrow. So if you put the meter on one way round, you will get continuity. But if you swap the probes round, you should get a reading that is so high it is almost infinity. There is a slight complication in that to test them, they should be out of the circuit. That is fine for the one in the unloader relay as you can take it out. Not sure about the one on the nearside inner wing as I have not seen one. But if the wire is disconnected from the alternator, you should be able to test in the same way.

 

 Can't say I am 100% certain about pin 7 because it is a while since I worked on a cluster. But I think I would remember if the diagram were wrong. That pin supplies most of the bulbs in the cluster. So you could pick another one from the diagram and check for a common live. In your test, did you bridge from the warning light socket via the bulb to pin 7? And if so, the live side or the negative side?

  

The ignition warning light should be on when you turn on the ignition. When you start the car and it is charging, the alternator turns the light off.  

Re. Diodes, this may be starting to escalate to depths beyond my admittedly basic knowledge, I'll have to educate myself a bit and get back to that hopefully. 

 

Oh I see, I thought pin 7 was specifically for the battery light, I'll double check the diagram but I trust your information. And sorry I didn't explain very well but no, that is one test I forgot to do. By live or negative side I presume you mean on the bulb socket? Sorry if this sounds incredibly stupid but how does one establish which is which? I presume it doesn't matter which way round the bulb is inserted in the socket?

Share this post


Link to post
Share on other sites

Diodes are just like one way valves. They conduct the voltage in the direction of the arrow but not the other way round. So when you put your meter on them, they give continuity one way but if you reverse the position of the probes, you get a very high ohms reading

 

No, it does not matter which way the bulb is fitted. If the diagram is correct, the ignition warning light, ABS and handbrake lights all get their live from pin 7.  So you would expect to find continuity between pin 7 and one side of each bulb holder   

Share this post


Link to post
Share on other sites
On 02/05/2020 at 01:21, whiskychaser said:

Diodes are just like one way valves. They conduct the voltage in the direction of the arrow but not the other way round. So when you put your meter on them, they give continuity one way but if you reverse the position of the probes, you get a very high ohms reading

 

No, it does not matter which way the bulb is fitted. If the diagram is correct, the ignition warning light, ABS and handbrake lights all get their live from pin 7.  So you would expect to find continuity between pin 7 and one side of each bulb holder   

Someone on another forum pointed out that if you ground pin 12 on the OBD port, it should illuminate the battery light, I tried that and it did turn on the battery light, but unfortunately still bad alternator function. I take it the next thing to check would be the uncoupling diode which I've done some research on. I've attached the diagram for it which I found online, but I'm struggling finding the actual location of it. From what I can tell, it should be somewhere between the OBD port and the fuse box? There's very little information on this which is frustrating, I believe I've found the part number (61311375475) with an image attached, is this what I'm looking for?

Screen Shot 2020-05-07 at 20.06.10.png

s-l300.jpg

Share this post


Link to post
Share on other sites

Grounding pin 12 of the OBD port is just the same as grounding the same wire at the alternator. IIRC, you tried that and the light did not come on? Perhaps there was not enough voltage at the time? The diagram above is the same one I posted in my first reply. That file also gives you the location of the components - in the case of the diode V9501, it is supposed to be on the left hand side of the engine compartment below X20. The location of X20 is shown on page 7100-0-01. It might be worth testing the diodes to see if they block voltage in the reverse direction as mentioned above. But if the bulb lights, they are probably fine. 

The alternator is responsible for grounding the wire at D+. When it starts charging, it will change that terminal to a positive and the light will go out (The diodes are there to stop that voltage going in the wrong direction). I realise that the alternator was checked but it does not appear to be switching correctly  

 

 

 

Edited by whiskychaser

Share this post


Link to post
Share on other sites

Right, unfortunately over the weekend things took a turn for the worst, I’m probably going to have to book it in at my BMW specialist unless there’s something very obvious I’m missing. 

 

On Saturday, I started it up and as usual was on 9v, after letting the engine warm up I gave it a rev to 4k to excite the alternator but it didn’t do it like usual, it stayed on low voltage for a couple of minutes, then while driving it suddenly spiked to around 16.7!!! Which is obviously way too much, the lights were the brightest I’ve ever seen them. In a 5-6 min journey it was fluctuating between 9-16volts but it still drove as normal. 

 

Started it up again this morning to see if it had reset itself (I disconnected and charged the battery overnight) and in addition to the ABS light, the red brake warning light was also on, Speedo and rev counter was stuck on 0, I didn’t get a chance to measure the voltage but I assume it was well below 9 as the engine cut out after about a minute, was running very rough and would barely stay on. I’ve looked at everything I can, and will probably have to take it to Munich Legends which is about 45 min drive away so I’ll have to look at getting it towed as it won’t make it there in this state. 

Share this post


Link to post
Share on other sites

Something is very wrong. Did a bit of a search and it seems you should have a 14.1v regulator. No way should it reach 16.7v. It sounds like the regulator may be faulty.

With regard to the cluster, have you tried the instrument self test? With the ignition off, push down the right hand cluster button. Turn the ignition on and release the button.  The speedo usually goes to 80mph and the tacho to 3500rpm. Be fairly quick about releasing the button. Hold it too long and you may change the cluster language. 

 

No disrespect to said garage but maybe try a local auto-electricians? They may even come out to the car.

 

Share this post


Link to post
Share on other sites
5 hours ago, whiskychaser said:

Something is very wrong. Did a bit of a search and it seems you should have a 14.1v regulator. No way should it reach 16.7v. It sounds like the regulator may be faulty.

With regard to the cluster, have you tried the instrument self test? With the ignition off, push down the right hand cluster button. Turn the ignition on and release the button.  The speedo usually goes to 80mph and the tacho to 3500rpm. Be fairly quick about releasing the button. Hold it too long and you may change the cluster language. 

 

No disrespect to said garage but maybe try a local auto-electricians? They may even come out to the car.

 

Yeah definitely something very wrong here and it's eating away at me...

 

The regulator is what I originally replaced, its less than 2 weeks old. It's a Bosch OEM part from AutoDoc, I got the part no. from RealOEM, there seems to be a 115a and 140a alternator but both of their regulators have the part number anyway. Also when you say 14.1v regulator, before this new problem, when I was able to excite the alternator I would usually keep the OBC on the volt reading just to keep an eye on it, and it would normally fluctuate between 13.8-14.8 depending on what electrics are running

 

I haven't tried the cluster test, I will give it a go either tomorrow or Wednesday which is my day off. However I forgot to mention that everything did actually work fine, it's as I released the handbrake that all the needles fell down to 0, lights would still remain on (the red brake light also came on when the handbrake was released) I've been told the handbrake has some sort of ground to the cluster that may be causing this?

 

Well the garage I was planning to go to is probably the most well known BMW specialist in the country but yeah I fully understand your point, it would most probably be more worthwhile having an auto elec look at it, trouble is I've struggled to find a good local one specialised in these kinds of cars before. Good shout though, I'll talk to a few local E34 owners I know and see if they have anyone

Edited by soorooshg

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×