Welcome to our resource for MG Car Information.



MG parts spares and accessories are available for MG T Series (TA, MG TB, MG TC, MG TD, MG TF), Magnette, MGA, Twin cam, MGB, MGBGT, MGC, MGC GT, MG Midget, Sprite and other MG models from British car spares company LBCarCo.

MG MGB Technical - Anti run-on circuit

I've just replaced the ignition switch on my US-market 1978 MGB with an aftermarket switch. Accessories, ignition, and starter all work fine - engine turns and almost catches but won't start. From the smell of it, it's fuel starvation.

The problem that comes to mind is the anti run-on circuit. The old ignition switch had a slate (grey) wire from it to supposedly an inline fuse to an oil pressure sensor to a fuel cut-off of some sort. With the new ignition switch, I applied +12v (always on) battery to this grey wire and fitted a push-button in the line. I expected to push the button after shutting down the engine to complete the circuit and stop any run-on.

The car does not start with or without the button pushed - in case this fact helps in diagnosis.

I am now wondering if there is something else I should have done to the grey wire. Perhaps more importantly, how do I disable the anti run-on circuit to see if the car starts? I have no idea where the components are located (fuse, pressure sensor, fuel cut off - assuming I'm correct that these are the components).

This is a continuation of another thread about ignition switch wiring, which I've retitled because the ignition switch now works "fine" I think except for the anti run-on. Or at least the anti run-on is the first place to look because it's the only thing changed with the ignition switch replacement.


Hi Daryl.

I don't think your problem is related to the anti run-on valve.

Does your fuel pump tick when you turn the ignition on ?. (or make the appropriate noise if it is an aftermarket pump).

Could this be something to do with the seat belt switch circuits, which I believe are ignition-switch related ?.


Don, the pump does not click. But it would be strange coincidence for that to fail right after the switch replacement. Sooner or later I will need to check fuel delivery. (Can't currently do that for reasons too complex to explain.) There is plenty of fuel in the tank.

Can you help me further with the seat belt matter? With the old ignition switch there was a purple wire attached to its side that had something to do with the belts and key-in warning buzzer (I think). Now it is attached to nothing. Maybe it should be attached to . . what? A ground? Unfortunately won't be easy to get at anymore, but if that's the problem, I must.


I should have said that the pump clicked when I first switched on, but never since although I've turned over the engine enough since to draw fuel.

Hi Daryl - I think your logic of using a push-button to operate the anti-runon valve is correct, if the push-button is of the 'normally open' type and closes the contact when you push it. To disable the anti-runon valve temporarily to see if it is that which is preventing the car from starting, simply leave the grey wire disconnected, or in your case don't push the push button :o)

With a different type of ignition switch maybe its contact arrangement is different in other areas as well, and it isn't putting out 12v to the ignition coil when it should, have you checked for voltage with the key in the 'run' position and when cranking? Have you checked for a spark at each plug when cranking? Does the fuewl gauge register? After cranking for a while take out a couple of plugs, if they are wet they are flooded, and I'd expect this if cranking for some time with choke and no spark, you say 'from the smell of it' you suspect fuel starvation, but there would be no smell with starvation, only flooding. If the plugs are bone dry with no fuel smell then there is nothing getting through. They should be dry but with a strong fuel smell if all is well in the fuel department.

Other problems with the anti-runon system can cause fuel starvation, to eliminate these as a cause remove the pipes from the float chamber overflow pipe - is is the valve applying suction to this pipe that sucks the fuel out of the jet and stops the engine in normal use, and can also prevent it starting if it is still sucking during cranking.

It was the earlier North American model that interrupted the starter circuit if seats were occupied but belts not fastened, but it interupted the cranking circuit, not the ignition, and the 78 didn't have this system. There should be a pink (not purple) wire on the ignition switch as well, and this carried a ground whenever the keys were in the ignition, which in series with an additional door switch sounded a buzzer if the door was opened with the keys in the ignition. However again this doesn't affect either cranking or ignition.

With the exception of taking the pipes off the float chamber overflow just treat it as a normal non-starter and go through the 'first-principles' check-list ignoring the fact that you have 'only' changed the switch and so it 'must' be that that is the problem. Check for spark, and at the right time i.e. at the top of the compression stroke and not the exhaust, plug lead order (1 3 4 2 *anti*-clockwise), and fuel.
Paul Hunt 2

Thanks, Paul. It is still 6.30 am here so no car work yet.

No smell of fuel at all, thus the inference of fuel starvation.

I did check that all terminals of the ignition switch put out voltage in the start position. With just ignition on, the dash lights do go on, and the fuel gauge registers. Headlights, etc work.

Will check for spark and fuel delivery later.

Plug leads etc must be in order. Car worked fine before the switch change-out. I did not go under the hood in the process of changing it.

Hi again.

Sound advice from Paul, as usual.

If it turns out that the pump isn't getting power :

The fuel pump takes power from 'the white wires' via an 'inertia switch' according to the Haynes diagram 10.34.
I don't know where this switch is, maybe it is possible that you pulled a wire off it while wrestling under the dash ?.


I certainly did wrestle under the dash. Does anyone know where the inertia switch is and what it looks like? From previous experience, it's always what you just worked on that causes the problem you had right afterwards.

Incidentally I could not remove the steering wheel. Hub puller wouldn't grab on the rubberized and funny-shaped wheel. Tried banging on the shaft (with nut loosened of course) while pulling, but did not want to break anything. Any ideas on steering wheel removal in the event I have to go right under the dash? (Please no one tell me to remove the shaft!)

Reading the archives, looks like a good chance that hammering on the steering column to try to get the wheel off could have tripped the inertia switch. Anyone know exactly where the switch is found, so I can press de button?

It's right above your left Knee as your sitting in the seat. It's been a while since I have looked, but it's roughly 4 inches long, circular, has button on the top of it. I thinks it is a grey color. I can check when I go home if you still haven't found it. I think the button pops out to deactivate, but it may be pushed in to deactivate.


Thanks, Jayme - with your help I found the inertia switch, And thanks all.

Problem solved.

Wasn't the inertia switch. Bypassing it made no difference. Later the car ran fine with it in the citcuit again.

Wasn't the anti run-on valve. (When I got the car going, it stopped beautifully using the new push-button arrangement, by the way.)

It was most likely the plug cable boots that need replacement and got shaken loose when I banged on the steering column to try to remove the steering wheel. I'm pretty sure of this because I got the car going but running on only a cylinder or two - which then remedied itself when I pushed the cables on tighter. Then it started and ran just fine.

Incidentally, it was a bear fitting an aftermarket universal ignition switch in the same place as the manufacturer's switch. Too little space, new attachment method needed, and it's possible to foul the steering column at the place where the lock is. But it is rather nice to have the key go in the same place as the original.

Thanks again for all the help.

This thread was discussed on 28/04/2006

MG MGB Technical index

This thread is from the archive. The Live MG MGB Technical BBS is active now.