More car trouble
Jan. 4th, 2004 09:19 pm![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
ewxAfter behaving for quite a while my car went back to its habit of refusing to start. The RAC engineer seemed more open-minded about the solution than previous ones (it's had two new starter motors to no effect...) and thought it was the ignition switch itself, evidence for this being that connecting the starter directly to the battery started it while turning the key continues not to.
While I was thinking of replacing the car anyway very soon I'd like to be able to reliably start it if only to get it to whatever place I finally part company with it, and conceivably for emergencies. Replacing the ignition switch would be expensive and not 100% certain (perhaps the fault lies somewhere in the relatively inaccessible region between the switch and the starter). But the engineer suggested a rather cheaper workaround.
The idea is to run a wire from the +ve terminal of the battery, via a 30A fuse and a switch on the dashboard, to the starter directly. (i.e. to emulate the green line on this diagram, not the red one.)
But I'm not exactly an electrician; does anyone who knows what they're talking about think this is a sane/insane thing to try?
(updated to link to diagram)
(no subject)
Date: 2004-01-04 01:31 pm (UTC)(no subject)
Date: 2004-01-04 01:36 pm (UTC)(no subject)
Date: 2004-01-04 01:37 pm (UTC)(no subject)
Date: 2004-01-04 01:38 pm (UTC)(no subject)
Date: 2004-01-04 01:42 pm (UTC)I'd consider the possibility that safety-critical systems may be separately switched on by the ignition switch (or by relays powered by the ignition switch)?
I'd wonder why it is that the ignition isn't wired that way, but uses a relay (presumably so that the massive current drawn by the starter motor doesn't go through the switch itself). I think you might have a job finding a switch that could reliably and safely handle the load a starter motor would put on it.
Might the "thing between the ignition switch and the starter" that you hypothesise is causing the problem, be said relay? In which case, if it's like various cars I've had unhappy dealings with, the relay may well be more accessible than the intermediate bits of wiring. Haynes manuals may be worth a browse?
(no subject)
Date: 2004-01-04 01:56 pm (UTC)Good luck.
(no subject)
Date: 2004-01-04 02:03 pm (UTC)It certainly does sound like the solenoid may be involved in the problem. <speculation type=wild> Perhaps there's damp getting into the solenoid? </speculation>
(no subject)
Date: 2004-01-04 02:12 pm (UTC)(no subject)
Date: 2004-01-04 02:27 pm (UTC)(no subject)
Date: 2004-01-04 02:34 pm (UTC)By "starter" I mean the entire starter motor unit, not the solenoid itself. More specifically, I propose to emulate the green line on this diagram (http://www.greenend.org.uk/rjk/junk/wiring.jpeg), rather than the red one (which doesn't include a switch).
I'm confident that the root of the problem lies outside the starter (three starters, of two models, have all exhibited it) though obviously this doesn't rule out a chain that ends within it.
(no subject)
Date: 2004-01-04 05:46 pm (UTC)The green wire first carries current directly from the battery positive terminal through the fourth position (start) on the ignition switch and thence to the starter-motor enclosure, thence to earth via the chasis. Inside which there is a solenoid or relay which engages high-current contacts to connect the starter motor across the battery. There being no latching, when the current to the solenoid is removed, the high-current circuit is disconnected.
Presumably the solenoid is in the same enclosure as the starter motor, and has been replaced each time, so probably isn't directly responsible for the failures. The starter-motor earth must be good, because activation with direct connection of the motor, seems to work when the system is failing.
If you have a period of reliable failure, you could perhaps check, with a multimeter and manipulation of the plug, if there's a voltage wrt ground at the starter-motor plug when you try to start (there should be, I think). You could further check, in a period of reliable failure, if shorting the battery against the solenoid pin (the end of the green wire) starts the motor. That would place the fault definately somewhere on the green route.
The part of the green circuit between the battery and ignition switch is, rather endeeringly, four-way-redundant, so I doubt that's the failure, and anyway, that also supplies the 'running' positions of the switch, so if that was faulty, you'd not be able to run the engine after jump-starting it. So it could be worth testing the switch output, with a meter, during failure, to see if the fault is in the switch or the end of the cable run. If you've a voltage on the switch but the engine doesn't start, then you could just replace the final part of the run of the green wire, between the switch and the starter. If it's the switch, you could either replace it or, cheaper, have a start switch shorting the input and fourth position. Then you'd need the key to operate the vehicle, to enable the running electronics, to start you'd simply use the key as before, sometimes pressing the switch whilst you are in position four (starting) if position four seems to do nothing.
The green wire first carries current directly from the battery positive terminal through the fourth position (start) on the ignition switch and thence to the starter-motor enclosure, thence to earth via the chasis. Inside which there is a solenoid or relay which engages high-current contacts to connect the starter motor across the battery. There being no latching, when the current to the solenoid is removed, the high-current circuit is disconnected.
Presumably the solenoid is in the same enclosure as the starter motor, and has been replaced each time, so probably isn't directly responsible for the failures. The starter-motor earth must be good, because activation with direct connection of the motor, seems to work when the system is failing.
If you have a period of reliable failure, you could perhaps check, with a multimeter and manipulation of the plug, if there's a voltage wrt ground at the starter-motor plug when you try to start (there should be, I think). You could further check, in a period of reliable failure, if shorting the battery against the solenoid pin (the end of the green wire) starts the motor. That would place the fault definately somewhere on the green route.
The part of the green circuit between the battery and ignition switch is, rather endeeringly, four-way-redundant, so I doubt that's the failure, and anyway, that also supplies the 'running' positions of the switch, so if that was faulty, you'd not be able to run the engine after jump-starting it. So it could be worth testing the switch output, with a meter, during failure, to see if the fault is in the switch or the end of the cable run. If you've a voltage on the switch but the engine doesn't start, then you could just replace the final part of the run of the green wire, between the switch and the starter. If it's the switch, you could either replace it or, cheaper, have a start switch shorting the input and fourth position. Then you'd need the key to operate the vehicle, to enable the running electronics, to start you'd simply use the key as before, sometimes pressing the switch whilst you are in position four (starting) if position four seems to do nothing.
<:)>Btw, always cut the green wire, never the red wire, or is it the other way around?</:)>
(no subject)
Date: 2004-01-04 07:01 pm (UTC)(no subject)
Date: 2004-01-05 12:41 am (UTC)I used to have a switch in a hard-to-find place connected to that wire so that even if someone broke in, they couldnt start the engine.
Use a push-to-make press button switch, rather than a rocker or "throw" switch.
Good idea
Date: 2004-01-05 01:10 am (UTC)There's a relay inside the starter motor housing
Date: 2004-01-05 01:16 am (UTC)(no subject)
Date: 2004-01-05 04:29 am (UTC)One benefit of a 5A fuse is that you could safely use a much thinner wire. (Obviously, the wire has to be a higher rating than the fuse.)
You need to tell your insurer you've done this to the car. If the car is insured against theft, they'll have the heebie-jeebies, so consider downrating to third party only, whereby they'd presumably not give a damn (but still need to be told). In practice, I don't honestly think it makes much difference to the security, since you still need the ignition key to unlock the steering and get the fuel ignition/pumping systems going. On that topic, I assume you've already thought to make sure all the other essential electrics do still work?
As for safety, provided you don't touch any live wires, I'd say the worst that could realistically happen is you blow the fuse. The more likely failure mode is simply that nothing happens, and it working is a more likely outcome still. Good luck. (-8
(no subject)
Date: 2004-01-05 05:06 am (UTC)