406 2.1td starter motor funny business

[djoptix]

Having all sorts of fun with my (ahem) bargain 406. Some other stuff which I'm dealing with but I'd welcome some ideas on this one

The other day, it started making a very odd grinding noise. This started happening while driving. I wasn't going far so I got there, parked up and investigated. I quickly worked out that the starter motor was stuck on - as soon as I turned the key to the "running" position the starter would fire and the engine would start, without me having to push the key any further.

At this point I got sick of the thing and put it up for scrap (this was the latest in a long line of issues). Having taken some photos and posted it on removemycar.co.uk, I resolved myself to driving it home making a horrible noise, so I got in and turned the key... to find it was behaving normally again. I assumed it was an intermittent fault with the moving bit of the solenoid (i.e. it wasn't retracting when it should).

Question 1: I thought that modern starter motors had some sort of clutch in them so they couldn't get dragged along and burnt out in this scenario?

So, off I went, drove it around for the morning with no issues. Took it to the garage for an MOT. Came back to get it (it failed) and it was doing the start-as-soon-as-the-key-is-in-position-1 thing again, but this time once started, it sounded fine.

Question 2: this is where it started to make no sense. Is there any way that the starter could be engaging and running, but then retracting once the engine is started? The only way that makes sense to me is that a) the solenoid is both engaging the gear and passing current inappropriately, but then the ring gear gives it a "kick" when the engine starts and it retracts...? or b) the starter does in fact have a clutch arrangement which was previously seized and now isn't, and the starter is permanently engaged and running but the clutch is protecting it...?!

This afternoon I tried to trace this fault a bit further, as it had now occurred to me that the ignition switch could be at fault (or a relay, if there's one between the ignition switch and the solenoid "control" wire...)

Question 3: is there?

Anyhoo - I disconnected the solenoid control wire, turned the key to 1 (dash lights came on as expected), then to position 2 (nothing happened, as expected), then nipped round the front of the car - having made sure several times that it wasn't in gear - and touched the connector to the bolt on the solenoid - the car started, as expected. "Aha," thought I, "that means it probably isn't the solenoid itself, it's the ignition switch, or a relay if there is one - it's supplying 12v to the starter in position 1 when it should only supply it in position 2."

Question 4: ...but if that's the case, the starter would be permanently engaged with the ring gear, so why isn't it making a horrible noise?

So, I went and turned the ignition off, then went back round the front of the car to reconnect the control wire to the solenoid. I touched the connector to the post; SPARKS and the starter turned again!! I went and took the key out of the ignition, then returned; same thing happens. So now the wire that tells the solenoid to operate is permanently live (although it wasn't before I started this test).

Question 5: WTF?! I'm thinking that the sequence of events means that it's unlikely the solenoid is at fault. It has to be the ignition switch or a relay. What do you reckon?

[Peter.N.]

It sounds as though you have an intermittent short from the solenoid connection to the ignition live circuit which could of course be in the ignition switch. Not going to be easy to find I'm afraid.

Peter

[djoptix]

^ that doesn't explain why at first it was making an awful noise (the starter being dragged along with the engine) and now it's stopped doing it... but that does make sense otherwise, and the ignition switch seems a reasonable place to start. I'll do some loom tracing tomorrow.

[Peter.N.]

If the solenoid was permanently energised it would stay engaged with the flywheel - I would have thought. Hope you have success.

Peter

[djoptix]

Right, much poking around with the multimeter and I think I have an answer.

As far as I can work out (perhaps someone with a Haynes can confirm), the ignition switch mini-loom has six wires coming off it; red and blue in a black connector, two reds in a brown connector, and an orange and a white in a white (ish) connector.

The two reds in the brown connector appear to run the fans and various other stuff (i.e. connecting them makes the fans run)
The white and orange in the white connector are (I assume from the colour and some basic testing) ignition live to the rest of the car, and illumination or something.
Which leaves the red and blue. The red is (I reckon, going by the girth of the cable) a main incoming live, and the blue goes out to the starter solenoid.

I believe that the blue wire (which is actually red once it leaves the connector it goes into) goes direct to the solenoid, no relays in between. This means that the ignition switch must be at fault. Or rather, the variety of faults I've had (solenoid energised without the key in, starter running and engaged with the ring gear, starter runs in position 1 but then turns off once started etc) make me think that it's probably the ignition switch at fault. The only other possibility is that there is a short in the loom itself, between permanent live and the solenoid control wire (as Peter suggested).

I'm going to replace the ignition switch and its attached loom and see what happens...

[Peter.N.]

If you connect a meter to the solenoid connection and have it in the car where you can see it, you can observe whether the voltage disappears from the solenoid when you release the switch.

Peter

[djoptix]

Yep, done that (my meter has a continuity beep) but unfortunately that just told me that the problem was intermittent... With the ignition switch out of the dash, I can start the car with it and make the starter stick on, but it doesn't happen reliably the same way every time.

New switch should arrive today, watch this space.

[djoptix]

The plot thickens.

Replaced the ignition switch, which seemed to make things better, then they went back to being the same. If I disconnect the solenoid control input, everything behaves as it should. If I connect it, no glow plugs, starter runs at position 1.

Got fed up, and snipped the wire that goes to the solenoid in the ignition switch loom. Have that, yer bar steward! Now all behaves as it should, when I touch the bare ends of the wire together, the car starts. BUT! Haha! After ~30 secs of running, the starter came on and did its awful-bloody-noise-getting-dragged-along-by-the-ring-gear thing, without the wire in the ignition switch loom being connected. PROVING that it isn't the switch, and there is a short in the wiring somewhere.

Thus began the process of trying to trace the wiring. I've removed the airbox, air supply to the inlet manifold, battery tray (which I'm going to repaint while it's out), thing what holds the loom, lots of those plastic wiring guards, and a lot of LX tape. Before I had to come in and cook tea, I got as far as working out that the wire on the solenoid has continuity to four connectors in the multipin plug next to the preheat controller. Now that can't be right, surely...?

One final thing. I found a connector (a big heavy duty ring) flapping around in the engine bay. A bit of multimetering revealed that it was constant live(!) and after some head scratching and consulting the wiring diagram, I worked out that it should be connected to the alternator. The seller did say it had had a new alternator... obviously whoever fitted the new alternator doesn't know that (all together now) refitting is the reverse of removal.

Anyway... my theory is... that permanent live is presumably the one through which the alternator returns current to the system. If it wasn't connected, the alternator will be trying to feed current back into the system through one of the other smaller connectors - there is one on a small connector to the side, which I guess is the voltage regulator, but another small one on the big permanent live connection too. If the alternator was chucking charge through that, presumably that wire might get a bit hot and melt a) itself and b) others around it?

That's what I'm hoping. I'm also hoping that I can remember where all the bits I've taken off go.

[djoptix]

It's fixed.

Having dismantled quite a lot of car, I traced the fault to worn insulation on both the solenoid supply and the glow plug supply. They were touching, and so the car was running the starter when it should have been heating the glow plugs. The one time when it was on with the key out must have been either that permanent live touching the solenoid without me noticing, or possibly the post-heat.

It's taken me so long to get to the bottom of this that I'm going to have to pay for a fresh MOT rather than just a retest gah!

[Peter.N.]

Well done. That's strange one but you never know - I used to be a TV engineer.

Peter