Not exactly moving the project forward

[pyoorkate]

So, the Hazard lights were briefly working, I say this only for it's potential (but I still can't work out how) relevance.

I spent a few minutes removing, one by one, the connectors on the back of the hazard light switch, cleaning them, and putting them back. I did the same for the hazard light-tell tale, and I worked the switch a bit. To make sure that this was all nice and safe I disconnected the battery negative/earth.

Having finished I put it all back together, reconnected the earth and tried the hazards. Lo, all was flashing.

A little earlier I'd switched over the horn to a new one, since the old one didn't work.

Then, in a fit of foolishness I decided to make sure that the car would start for it's short journey up the ramps onto the trailer tomorrow (for it is tomorrow that it goes to the welder). Yeah. No. It's started fine and run fine every few days / once a week since it's been here (apart from an occasionally high-pitched tinging sound which seemed to go away - and which I'd not been able to find or place). Today, after reconnecting the negative (promise, I didn't take both leads off, so it can't've been that). It'd even started several times earlier today. When I came to try this time, there was (and I may have imagined this) what looked like a puff of white smoke from the voltage regulator, followed by the solenoid for the starter clicking briefly, and then the ign and oil lights going out on the dash. No effort from the starter at all.

Having pootled around and checked connections I tried the igniton again, no lights / no hazards / no horn (so even the stuff that's meant to work with the ignition off isn't working).

Since then it's come back to life once since, but my attempt to use the starter went exactly the same way - and having faffed around tidying up for 20 minutes the dash-lights still weren't on. So, uh, fundamentally; help?!

I can't imagine that it's got anything *actually* to do with what I was doing, and it's purely misfortune and coincidence - but I'm not used to cars that have anything between their starter solenoids and the battery...

Does that voltage regulator do something more clever? The fuses all appear to be intact.

[spunkymonkey]

There's nothing unusual in the (published) circuits on these, Kate. Assuming you have a multimeter available, it's just good ole electrical fault-finding I'm afraid.

Start with side lights on and check the voltage between the battery -ve post and the bodywork - if it shows ANY voltage at this point then the earth is bad - either the earth lead itself or the connection at one end or the other. If you do get a reading then, with the lights still on, check the voltage between the battery post and the clamp, then the clamp and the tag that's bolted to the body, then the tag and the body itself. Any that are faulty will show a voltage.

The sidelights only draw about 1.5 amps, so even a 0.1v drop having them on would become about 4 volts lost under starter load (assuming a 50A starter current). Much more than 0.1v on lights and you may even lose enough to prevent the starter solenoid working!

The next test is to do the same between the battery +ve terminal and the B+ terminal on the voltage regulator. Again, you should be measuring NO voltage between these points.

The B+ terminal on the regulator has the wire from the battery and another that goes to the fuse box. This is the main live feed for just about everything - make sure there's a good connection at the regulator between this wire and the one from the battery. You can check the voltage across this connection by sticking a pin through each wire (be careful not to touch earth with them!) and then using the pins as contact points for your meter.

For a complete electrical failure, there's a fair chance that the fault will show up by this point but, if not, post again and I'll run through the rest of the system

Joe


Added: in case you're wondering, I'm suggesting testing voltages rather than using an ohmmeter to check resistance because it's much more sensitive - in the example of a 0.1v drop across the earth lead, that would be caused by about 0.07 ohms resistance and a "normal" ohmmeter can't read reliably below about 0.5 ohms.

[daf44]

hi kate.

something BETWEEN battery and starter
the voltage regulator on my 44 is on a separate lead.
i have a lead straight from the battery to the starter solenoid.

shouldn't make any difference though if all the connections are good.

follow joe's advice- its exactly how i trace faults like this.
if you do not have a multi meter than a quick cheat is to connect a jump lead from the battery +ve straight to the starter solenoid and see if it works. at least this will tell you if the fault is on the +ve or -ve side.

let us know how it goes.

paul44

[pyoorkate]

Sorry, I meant between the battery and starter solenoid. On the minor it's not fused... iirc. Mind, not much is on the minor... Fuse the headlamps? Whyever would you do that

Anyhow, thanks for the advice - I had to stop yesterday 'cos it was 2030 and I'd not actually seen my partner for more than about 20 minutes since I got home at 1500, but today I got out there and - well, I didn't do a methodical fault search but instead did what I should always do with classics. Removed each and every connector I could get to, cleaned it, tightened the connectors and then put them back together.

The regulator was a complete backside. The connection at the bottom nearest the battery had got rusty, and the screw that holds it in at the bottom was similarly rusty. Took a good 45 minutes just to get the screw out of the bottom (at which point it was munged) and then the screw out of the regulator. Of course, none of this actually fixed the problem...

...and I'm assuming here that I have fixed the problem because we've gone through several start/stop etc routines and it's still working. The problem was (drum roll) that the connection between the +ve battery lead and the battery had gone. I was still getting 12V, but giving it a clean seems to have sorted it. I'm still rather surprised that nothing at all worked with it as it was... And I've no idea why it started and stopped working. But we'll monitor the situation and see how it goes

[spunkymonkey]

The reason for nothing working goes something like this, Kate (if you're interested for future reference):

Imagine the two circuits below. The first is the situation with everything switched off, so there's no load on the battery. The connection resistance, Rc, represents resistance caused by a bad connection such as a corroded battery clamp.



Now, if you connect a voltmeter between the "loose" end of Rc and the battery negative - by connecting between the clamp and earth - then a very small current will flow through the meter. Depending on the meter this will be a hundred thousanth of an amp or less ( say 0.00001 amps)

This current will also flow through Rc because it's part of the circuit and will produce a voltage drop across it.

The voltage across Rc depends on the current flowing and the resistance and is worked out using Ohms law: V = I * R

The voltage across the meter (which is what it will read) is the total battery voltage MINUS this "lost" voltage across the connection.

Say the bad connection is giving a resistance of 0.5 ohms (barely enough to measure accurately with most multimeters), then the voltage across Rc will be 0.00001 amps X 0.5 ohms = 0.000005 volts. The voltage across the meter will be 12 - 0.000005 = 11.999995 volts. That's going to read as 12v in practice!

Now think about the second circuit, where you have a load switched on (say, the sidelights). These are 4 bulbs, each rated at 5 watts. On a 12v system, that 20w total load will draw a current of about 1.7 amps. That current also must flow through Rc. In this case, the voltage calculations work out like this:

Rc = 0.5 ohms
I = 1.6
V = I x R
= 0.5 x 1.6 = 0.8v

So now, there is nearly a full volt across the connection and only just over 11 volts for the rest of the system.

If you turn the ignition on, it gets even worse. A typical 12v coil will draw around 4 amps when the points are closed. So, you have 4 amps flowing and 4 x 0.5 = 2 volts lost. Only 10v available for the electrics now.

Hit the starter and, even if the solenoid manages to energise, you'll have around 30 amps trying to flow through the motor. Only, 30 amps x 0.5 ohms = 15 volts across the bad connection and nothing at all left for the system. In this case, the connector will get very hot very fast and nothing at all will happen!

That's why, when tracing bad connections with a voltmeter, you should always have a current flowing in the circuit you're testing - a circuit that seems fine "off load" may have connections that will kill it as soon as it's loaded.

[pyoorkate]

d'oh!

Makes sense now

Thanks for that