1997 R170 Charging issue

[Borris1954]

As it was such a lovely day yesterday the SLK was taken out for its first real run of the year. As it's not used greatly (especially in the winter) I fairly regularly charge the battery which I last did about two/three weeks ago.

"Senior Management" and I had a lovely drive in the country with the top down and enjoyed a great pub lunch before returning home. Parked on the drive I went to close the roof - absolutely zilch happened. No light on the switch, no noises from anywhere - nothing at all. As the engine was still running I thought I would turn it off and restart and try again. Turned the key - a small click and the starter motor struggled but failed to turn the engine over - battery as flat as pancake. Given we had just been driving for a while that was not what I expected to happen. So I jump started the car and with the leads still in place closed the roof. I connected up my battery charger (not really expecting it to do much) and went off to look at battery prices. The upshot is that the battery charged normally and held charge overnight and starts car fine today. So not the battery!

A voltage check reveals the battery voltage is about 12.2V until I start the car at which point it falls under load of running engine to about 11.4V. My attention next turned to the K56 crash relay. This appears all good with same voltage on both the G and battery terminals and an ohmmeter check reveals the relay is definitely closed when the engines running and open when it's not - all good.
So it looks like the alternator is defective.... BUT......

What I don't quite understand is a symptom I should have notice much earlier and may have been there for ages (possible even since I bought the car 3 years ago). When I turn the ignition on not all the warning lights come on and stay on until you start the engine as you would expect - one being the battery light! Also some flash on very briefly and then go out - the oil pressure being one. I'm wondering if they are being extinguished when the K56 relay closes.

I have searched and googled but can't find this behaviour described anywhere - has anyone seen it before?

I guess the next step is to get the car up in the air and take a look at the alternator but I am just a little nervous the issue may still be elsewhere.

 

[Borris1954]

Actually thinking about it a bit more the oil light was definitely working OK when I did the last oil change so it's something more recent.

 

[Borris1954]

I decided to swap out the alternator so bought a cheap (£26!) used one from eBay which was advertised as good working order. When it arrived I quickly checked it out by bolting it in a vice, connecting a 12V bulb to its output and spinning it up using a socket on the pulley nut and an electric drill. All good.

There's plenty of descriptions on the Internet on how to change it but there's no substitute for grazing your own knuckles. There's not much room and getting the old unit out and the new one in will be a shed load easier next time! The biggest issue I had next to lining up the bolt holes on the replacement was getting the drive belt off. The tensioner needs an external torx socket according to the writeups I read but I got nowhere with that. No room for a ratchet and trying to hold the socket in good old "mole" grip locking pliers failed too. Then I noticed the tensioner wheel had a plastic cover on the front face which was becoming unclipped. Taking this off revealed a nice standard hex bolt head that a ring spanner fits onto nicely - problem solved!

Anyway the upshot is that once the replacement was installed I fired up the engine and revved it up a bit and the battery voltage rose to a more normal 14.2 volts. My plan now is to properly overhaul the original alternator with new bearings, slip rings (and voltage regulator if needed) and put it back in the car at some point.

However, the battery light on the dash still isn't working so I need to investigate if it's just the bulb or something else.

On that subject I'd be grateful if some who has a pre facelift R170 SLK can confirm the behaviour of the oil level light. Mine lights very briefly when you turn the key to position 2 and then goes out before the engine is started. The coolant level light does exactly the same. Is this correct? I was previously thinking of it as being more like an oil pressure indicator light of old which therefore needs the engine running to get oil pressure whereas a level sensor doesn't. Also given the manual says it's an oil level warning light and the sensor is actually mounted on the side of the sump rather than in a pressurised part of the system I am now thinking this behaviour is correct. Hopefully somebody can confirm one way or the other.
Cheers




Sent from my SM-T710 using Tapatalk

 

[Borris1954]

I have now found the relevant bit in my copy of WIS that confirms my instrument cluster bulbs aren't working properly. It says all indicator bulbs should come on when key is turned to position 2 and that only the SRS malfunction indicator bulb should go out after approx 4 seconds and in some countries (not UK) the seat belt reminder after 6 seconds.

So I have another problem to sort out! Anyway at least I now know why I hadn't spotted that my alternator had go AWOL. In terms of the battery light not coming on I thought I would check what was happening with the signal from the alternator that feeds the instrument cluster. As expected this was at zero volts with engine off and just over 14 volts when the engine was running. So all good.

This morning I removed the instrument cluster and started checking all the indicator bulbs - every one tested good and there are no obvious issues with the cluster itself that are visible to the naked eye (no broken tracks around the bulbs etc) but I did not attempt to dismantle it. I have also checked every fuse I can find in the car - all good.

So I am now a bit stumped and everything would seem to point to a problem in the circuitry within the Instrument Cluster itself. Reading WIS it seems quite a complex beast and apparently there is even has a microprocessor in there somewhere.

Any bright ides anyone? Anyone know if I can just bung in a suitable second hand one and expect it to work (albeit with the wrong mileage displayed) or if I will need some coding magic done?

Normally Google is my friend in these situations but so far I have only been able to find one thread where a similar fault was reported. Sadly that was someone looking at a car to buy which I think they just walked away from so there was no problem resolution posted.

Any advice welcome.

 

[Borris1954]

Instrument cluster is now on it's way to Cluster Repairs UK (of Wheeler Dealers fame) who can hopefully sort it for me. I will keep you posted.

Sent from my Nexus 7 using Tapatalk

 

[Borris1954]

The issue is now fully resolved. Cluster Repairs UK advised that the cluster wasn't repairable and I should replace it. So I purchased a second hand one and put it in and full of expectation turned the key. Low hand behold no change whatsoever! Back to square one. The ebay seller (a very keen R170 man) was a great help in talking through possibilities and even connected up his laptop and read out the fault codes for me. Still nothing obvious so it was back to first principles and a lot of studying of the WIS wiring diagram. The key to the issue seemed to be to find out why the battery charge light wasn't coming on at all with either instrument cluster.

First I needed to make sure the bulb had power. I identified that if I could make the bulb failure light come on by disconnecting a headlamp bulb that would prove power was available at the at battery charge bulb. Logically the other side of the bulb should earth out through the alternator when the engine is off. As I had already measured the voltage at alternator and new it was zero with engine off I felt I had an issue inside the clusters or a wiring fault.

By applying 9v from a battery between the right pins on the both clusters I could make the charge light come on OK. So both clusters looked good. Measuring the resistance to earth via the alternator at the cluster end showed essentially an open circuit. Measured at the alternator end it was just a few ohms so a cable fault was obvious. Removing the connection at the alternator again it all looked fine. However, a sharp tug on the crimped on connector revealed it not to be crimped on at all. The wire was broken at the crimp point but all was held in place by the hardened rubber sleeve and no electrical connection was being made.

I crimped on a new terminal connector and put it back on the alternator and the fault completely cleared and my non-repairable cluster is once again working perfectly!

With hindsight the cause of the fault is blindingly obvious. What fooled me is that some lights were coming on and staying on until the engine was started. This was giving me false assurance that the signal from the alternator was getting through. But reading WIS a bit more shows that the engine running state is calculated from both the alternator feed and by the engine revs being above a certain threshold.

Very pleased to have got there in the end and learned a load on the way. Just a shame I didn't change that crimped on terminal first....

It also begs the question as to what Cluster Repair UK think is the the non-repairable fault on my original cluster. I must drop them an email and ask.

Sent from my SM-T710 using Tapatalk