Following on from a post in a battery-light topic some weeks ago, I've been running an experiment, with surprising results.  I've done it on my Polo (see profile), but the parts and wiring are all interchangeable with a Fabia as far as I know. I think I started the experiment on the same day as that post, so several weeks ago now.

 

What I was hoping to demonstrate was that the battery light in the instrument cluster could be activated if a low voltage situation occurs even if the signal wiring from the alternator is broken. What I did was just unplug the two-way connector from the alternator, so that neither the blue exciter wire nor the brown (brown/red in some cars) DFM/Load sense wire is connected to the alternator.

 

I have an LCD display of battery voltage permanently showing, so I could see if/when I was causing charging problems, and be able to see at what voltage - if ever - the battery light was triggered.

 

The problem is... that the car doesn't seem to care at all.

 

No battery light at ignition on, as expected, but no non-charging situation, no other warning lights of any sort; not even any faults logged according to VCDS.  The car doesn't have ASR or any of that fancy stuff, so that clue to the DFM wire being disconnected isn't available.

 

On just two occasions (one this morning which reminded me to post this) the alternator has taken a few seconds longer than usual to start itself up, the displayed voltage lingering around 12V for maybe 15-20 seconds after start-up of the engine, instead of the usual fast ramp up to 14.5-ish within about 4 or 5 seconds.  Other than that everything has behaved totally normally. Charging voltage has been the same, car has driven the same, everything has worked. I don't even use the car every day at the moment 'cos I can cycle to work when it's not raining.

 

So my mission to investigate whether the charge light can come on even when the two signal wires are not connected to the alternator is a dismal failure, as I can't get the battery into a low state of charge, or the alternator to not be charging it.   I don't want to disconnect the main alternator output lead as I don't think running them open-circuit is advisable, nor do I want to leave lights on to deliberately get the battery into a poorly state, as that just seems silly, and unlikely to help my experiment if the alternator starts up and starts charging regardless of the blue wire's presence or absence.

 

Seems like some alternators don't really need the exciter connection, and some engine ECUs just don't care about the DFM/Load sense function.  Surprised by this. Disappointed that I still can't answer my own question about the battery light either.

Amuses me also that I spend so much time telling people on here to check their battery light at ignition on, and yet the broken wire which this indicates when absent doesn't seem to automatically correlate with any problems, on some variants at least. Might be a diesel/petrol thing, since diesels use so much charge on heating glow plugs and cranking? Maybe their ECUs do more with the DFM information?

 

Anyone got any thoughts? I wonder if I could feed a 'false' low voltage up the voltage sensing wire that goes from the battery fusebox  to the Onboard Supply Control Unit, and fool the car there's a charging problem that way?  Sounds like the next experiment...

 

 

I suspect the alternator is chucking out the correct voltage but you may well find no significant current and therefore no battery charging taking place, awkward to measure the DC current reaching the battery while the engine is running but you could probably cobble something together.

10 minutes ago, sepulchrave said:

I suspect the alternator is chucking out the correct voltage but you may well find no significant current and therefore no battery charging taking place, awkward to measure the DC current reaching the battery while the engine is running but you could probably cobble something together.

 

This is a good point but the battery and consumers in the car alone probably won't be enough to test it, you would have to find something else to clip on the battery to draw some real power. Then by seeing if the voltage drops from 14.5 you will be able to see if there is any difference between having the load sense wire connected or not

Voltage difference drives current flow; if insufficient current is available to meet demand, voltage dips.

Well I gave up on this experiment and plugged the alternator signal connector back in a couple of weeks back, and haven't got round to the other experiment I mentioned in the last paragraph of the 1st post.

 

Just now that experiment happened to me though. Battery light came on while driving, about a mile from home. 

 

I'd noticed a really strong smell as I left work ten minutes before, wasn't exactly sure what it was, or whether it was from the car or external, first thought was fuel, but it didn't seem quite right, then a minute later I thought "gloss paint, that's it" really intense smell of it.  Seemed to dissipate as I drove further.  I did happen to notice that my voltmeter on the dash was reading a little lower than I'd normally expect for the ambient/engine temperature, only just over 14V.   Aircon was on so I switched it off for a bit to see if that changed anything. Smell cleared after a mile or so, and I didn't look at the voltmeter again until the battery light woke me up to the (non-) charging situation about 3 miles later. At the point the battery light came on the voltage was just under 12.0.

 

Thought, Oh well, at least I can test the "can the battery light come on even if the alternator signal connector is unplugged" thing when I get home...

 

Turns out that @KenONeill is right about that; soon as I unplugged the two-way signal connector from the alt, the battery light disappeared, soon as I plugged it back in, there it was again. Engine running all the time, voltage showing just about 12V.

 

Before I got home, I thought maybe the smell was the aux belt melting on a seized pulley, but no, that was happily turning everything it usually does.

 

Letting the engine cool for a while before I start diagnosing, battery voltage now showing at 12.6, so nothing shorting it seems. Alternator strip-fuse looked intact at a glance.

 

Smell sound familiar to anyone? Is it related? What's happened? Watch this space...

 

 

Alternator internals giving you the "don't mess with me!" - ie a diode or 2,3,4,5,6?

Definitely alternator, not sure exactly what within yet.

 

Measured output voltage locally from output post to case casting and saw a couple of tenths of a volt less than what I was seeing at the battery posts, engine idling.

 

Just swapped in a spare alt and we're back to normal service. Nothing to see on/in the old alternator, but I won't dig down into that before the weekend, I should think. 137000 miles on the clock, for what it's worth. Spare was a year younger according to date code, can't remember if it was advertised as low mileage or not. Couldn't BA to check the brush length before fitting.

It's not a Valeo alternator is it?

3 minutes ago, SuperbTWM said:

It's not a Valeo alternator is it?

 

No, old and 'new' both Bosch; 037 903 025M

 

Edited 31 July, 20187 yr by Wino

Just wondered, had a golf around that year and you mentioning checking the brush length reminded me of the valeo alternator I took off it which had incredible commutator wear and the brushes were like new. A brilliant way to get you to buy a new one rather than just maintaining it.

 

Must of been a cost cutting exercise, the replacement from the dealer was a Bosch though.

 

Does the old alternator smell burnt?

Hehe that's a cheeky  'innovation'!

 

No smells generally under the bonnet, nor from the alt itself.  I think the paint smell earlier may have been unrelated.

@Wino - Cheers; I actually once had a control wire come off the alternator whilst driving. I noticed it when the battery light didn't come on next morning, and took it straight to the garage who made their minimum labour charge, and described work done as "replaced alternator control wire".

I seem to have 6 working diodes, 3 intact stator windings, 1 intact rotor winding. 

I guess that means the voltage regulator is at fault, though brush length seems ample.