Running on

[rmaddock]

Otly's going but won't stop.

I've taken the +ve from the old coil and used it to switch a relay which supplies power to the new ignition system.

It starts fine but then when I turn the ignition off, the engine runs on. The first time it didn't want to stop at all but when I disconnected the original +ve (as above) it did stop. Subsequent runs have seen a delay of a second or two after turning off before the engine stops.

Any ideas?

[Oliver Sedlacek]

Most carburetted engines will run on under some circumstances. A hot exhaust valve can ignite a fuel-air mix, even without a spark. Many carburettors have a solenoid cutoff on the idle circuit jet to prevent run-on.

[rmaddock]

Yes, but this only happens with the EDIS setup. I put the dizzy straight back in again and it doesn't do it.

I'm guessing there's some sort of delay in the relay switching the power off.

[Oliver Sedlacek]

I would eliminate the relay as a potential culprit. A megajolt and EDIS shouldn't really take much more current than the old coil, so I would just wire it up directly.

[johnb]

My megajolt worked fine when I wired it up temporarily, but
when I wired it in properly it overran, and would have carried on
idling indefinitely if I hadn't stopped by engaging gear and letting
the clutch in. I noticed:
a) a red power indicator light that I put on the Megajolt was glowing
much weaker than normal;
b) the ignition warning lamp had come on.

The megajolt and EDIS were the only loads to ignition switch. The ignition lamp is quite big (probably a few watts), not the little glow worm that many cars have. I reasoned that the lamp must be conducting enough from the alternator to the switched side of the switch then to the ignition (At least the EDIS module has to be able to function on low voltages in order to provide sparks which starting). I put a small diode in series with the lamp, and it cured the problem. The diode allows conduction from the switch to the alternator but not from the alternator to the switch.

The tell-tale sign was the ignition lamp coming on, but this could happen with other causes of running on.

My idle speed was quite high; a lower speed might not give quite enough voltage for indefinite running but it might try for a few seconds.

Turning on some other load powered off the ignition switch, such as
a radio would probbaly also have killed it but I didn't have any other loads.

Looking on the Internet, some Ford wiring looms contain such a diode; some owners remove them but don't seem to understand what they are for.

John

[Oliver Sedlacek]

Is the feed to the coil pack switched through the ignition switch or is it connected to a battery feed?

[rmaddock]

Thanks for that John.

That sounds exactly like what mine's doing. I had guessed from looking at circuit diagrams that a bit of current might be leaking back through the ignition light. I've put a relay in my circuit so it stays on and running come hell or high water.

In the short (for short read long) term I've put an extra switch on the dash to turn it off manually....Her Majesty's armed forces had even left me a pre-drilled hole of the right size.

Somebody had suggested a resistor on the +ve coil wire to drop the voltage to the relay but the diode in the ignition light circuit sounds like a better solution.

Can you give any pointers about the type of diode please? My electrical knowledge is limited. I know what they do but not what sort to start with.

Cheers!

Robert.

[johnb]

Robert,

I used a diode I found in the scrap box which was a 1n4005 (1amp 600V) but a 1N4001 would do (1 amp 50V) or 4002,4003,4004. It just has to withstand the alternator reverse voltage of around 20V max and take the lamp current which is probably a good deal less than 1 amp. It is a few mm long and a couple of mm in diameter so could be soldered into the lead and covered with sleeving.

If you are in the USA you can go to Radioshack and have a good choice If in Britain (as implied by Her Majesty) where nothing so useful as a diode is sold on the high street, I could mail you one (I live in Oxford).

If the vehicle is ex-military does it mean that it has a big red ignition lamp that soldiers can't miss? That would explain the fact that it carried enough reverse current.

All this presupposes that the coil is wired directly to the battery; if the coil is wired to the ignition switch I doubt whether there would have been enough current flowing through the lamp to keep the ignition powered.

You can obvious check this theory by seeing if the engine will stop when overrunning if the lamp circuit is broken. It is worth finding out for certain in case it is something more insidious like a short or miswiring.

John

[rmaddock]

John,

I am in the good old U of K.

Although ex MOD, the Rover is actually ex-TA and civvy spec apart from the paint job. The ignition light doesn't seem too big.

I think my problem stems from my having used a relay to switch the ignition controlled by the original ignition wire. I've measured the residual voltage with the ignition "off" and it's about 2v. From looking at the circuit diagrams and from my sketchy understanding of car electrics the charge light is the only possible culprit.

I'll get a diode from Maplins,thanks. I keep ordering stuff hither and thither at the moment since volunteering to build a robot for my son. No honestly, just for him!

I'll try the "broken" lamp circuit test I think. Although I understood that the lamp's functioning was necessary to excite the alternator and get the thing started in the first place.

What fun!

Cheers.

Robert.

[Oliver Sedlacek]

to keep many 12V relays energised! Most alternators will start charging even without the ignition light as they probably have enough residual magnetism, so it's worth trying it with the bulb removed.