Alternator problems

[tvrolet]

Struggling to work out what's going wrong here. The car ran a small block Chevrolet motor for quite a few years - the alternator was off a Citroen of all things, and the charge light was an LED with a parallel resistor as part of a set of dash lights. Everything was fine - charged fine, always a steady voltage and the charge light worked perfectly.

I've now put in an LS7 and there was no way the old alternator would fit, and hunting about the only 'regular' alternator that was short enough was from a Ford Explorer. First off it seemed to get real hot real quick - this is on a car that's only running a (big) fuel pump, ECU, injectors and data logger/dash. There will be more load if the radiator fans kick in, but even with them off it heats up real quick.

But the bigger problem was the charge light would flash intermittently. I could see voltage spikes on the logs and in an effort to cure I ran another earth direct to the block and rewired the live feed through its own fuse direct from the main power. It didn't cure it, but with nothing else to do really and as it didn't seem to be doing much harm I ignored it. But that alternator packed up in a very short space of time - and it was new, not a recon unit. Regulator knackered apparently, and I got a new replacement unit.

Ran the car again yesterday for the first time after some other engine work with the new alternator, and the same thing - voltage spikes. This is a trace from a lap (and yes - taking it easy on the revs for the first time out).



So while its mostly at 14.1 volts, every so often it spikes up to 16.5 volts and drops to 12 volts, and at this time the charge light flashes (as before, LED with resistor in parallel). The last thing I want to happen is to break this alternator.

So as an alternative I had a Denso mini-alternator I had always planned to use, so last night I rigged up some brackets.

Different problem now. With ignition on (so power to the alternator Ig feed) and engine off the charge light doesn't illuminate at all (and the alternator is wired properly - I checked many times!), but when the engine is running the light glows dimly, but the logs say its giving a good charge. So the charge light circuitry on the Denso alternator is obviously different to the Ford Explorer (and previous) alternator.

So - 2 choices. Either, any suggestions as to what's causing the voltage spikes in the original alternator and if I can address it I'll use it, or why does the Denso alternator not show a charge light when it's not running, and then shows a dim charge light when its actually charging?

[Pumaracing]

Not really my area at all but could you have a bad connection between the battery and the alternator?

[stevieturbo]

Just how modern is this alternator ? and how many connections ?

Lots of modern stuff are ecu controlled, so not just a simple install.

[Steve_D]

Are you still using the TVR 100amp fuse between alternator and starter?
These corrode due to being mounted on the chassis rail.

Also

As a test try taking your LED etc out of the circuit and replace with a standard bulb and see if that makes a difference.

Steve

[stevieturbo]

Should the resistor not go in series ?

[nsa]

I come at this from a very low level of experience, but on my old Porsche 924 the alternator circuit relied on a good connection actually at the dashboard light. Think I had voltage spikes, then eventually low charge voltage. The dashboard light worked intermittently then failed to come on at all. I replaced the alternator and regulator, and fiddled around with a lot of different things before checking the bulb connection. Cleaned it up and reseated, and all OK now.

Confess I didn't understand your OP fully, so forgive me if you've tried to eliminate the bulb already.

[tvrolet]

Thanks for the suggestions folks, but still no further on really. Although the folks who make the mini-dash with the [previously working fine] charge light suggest bridging the wire from the charge light to the alternator to power with another resistor. I'll give that a try, but it doesn't seem to address the intermittent nature of the issue. Worth a try though.

Pumaracing said:

Not really my area at all but could you have a bad connection between the battery and the alternator?

Checked and re-checked. Even took off and re-connected the cables just in case.

stevieturbo said:

Just how modern is this alternator ? and how many connections ?

Lots of modern stuff are ecu controlled, so not just a simple install.

Just 3 wires - main battery power, switched live and charge lamp. No idea how 'modern' it is, but it's new.

Steve_D said:

Are you still using the TVR 100amp fuse between alternator and starter?

There is a fuse between the alternator and starter, but not the 'TVR 100amp' one. I wired up the car from scratch myself so any similarity with a TVR loom is purely coincidental

Steve_D said:

As a test try taking your LED etc out of the circuit and replace with a standard bulb and see if that makes a difference.

That was on my list of things to try; if the 'resistor' doesn't work then it really is the last thing to try I think.

stevieturbo said:

Should the resistor not go in series ?

I'm just going by the wiring diagram that came with the dash/charge lights. I'm guessing putting the resistor in parallel reduces the resistance of the circuit, and in series would increase it. I don't really know what is 'correct' though. But also googled LED charge lights and most of the digrams have the resistor in parallel.

nsa said:

I come at this from a very low level of experience, but on my old Porsche 924 the alternator circuit relied on a good connection actually at the dashboard light.

Sure sounds like a culprit, but I've been over the wiring so many times and checked out voltages and resistance and everything looks fine.

The only thing I can't really do is check while the car is on the move in case vibration or braking/cornering is making something short or disconnect. But I've replaced the live feed just in case and been over the charge light wire so many times I would surely have moved anything that would be a problem.

[annodomini2]

Voltage spikes is probably a fubarred regulator. You can put a 2.2uF paper capacitor (in parallel) on to smooth it out if you want, but ideally it needs to be fixed.

[Pumaracing]

Suggestion I have had from a friend on another forum is that (as Stevieturbo has already said) an LED requires a resistor in series to limit the current through it and one in parallel would not do that. He suggests that as many alternators require the warning light to be working properly to make the alternator work properly that you rig up the exact same warning light as the alternator would have had as standard before condemning the alternator itself.

[Sardonicus]

Pumaracing said:

Suggestion I have had from a friend on another forum is that (as Stevieturbo has already said) an LED requires a resistor in series to limit the current through it and one in parallel would not do that. He suggests that as many alternators require the warning light to be working properly to make the alternator work properly that you rig up the exact same warning light as the alternator would have had as standard before condemning the alternator itself.

<This most alternators need the bulb/resistance to excite the voltage regulator and allow charging as mentioned already though many late alternators are controlled via the engine management inc free wheeling front pulleys, and can I take it you do not have one of these fitted to the Explorer alt because when they start to fail they can cause all kinds of charging abnormality's its just a sprag clutch after all and they can slip (erratic charging) when dying check all electrical avenues 1st though.

[LordLoveLength]

An LED isn't ideal for the job of a charge lamp on most older style alternators.
It won't pass enough current to the alternator to get it working properly, and would need to be bypassed by a resistor in parallel(which you say it is).
But an older alternator is quite capable of passing current *out* of the charge terminal when the ignition is switched off - hence the flash of the charge light as the engine stops.
This will kill your led by reverse bias - unless it has another LED reverse connected across the first, and they need to be 12v LEDS.

It may be that you have a faulty LED, or resistor. The best and easiest way to test is, as suggested, do away with it and temporarily wire a 5W 12V bulb between +ve and the charge terminal on the alternator.