MG-Cars.net

Welcome to our resource for MG Car Information.

Recommendations

Parts

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 - Oil pressure switch for anti-run-on

Ok, very frustrated. All the replacement switches available locally as "exact fit" seem to be "normally closed" which results in a drained battery. Is there an alternative wiring approach to allow use of these switches, or is it necessary to pay considerably more for a "normally closed" switch?
Thanks for any info/insite,

Pat
'73 BGT
Patrick Callan

Your statement/question appears contradictory.

Do you want normally closed or normally open?
Dave O'Neill2

Dave,
Believe it is supposed to be "normally open" but all the switches I have found so far have been "normally closed," even though their catalogs indicate an exact fit replacement. Guess I may have to go to Moss instead of local parts suppliers.
Patrick Callan

Oil pressure switches are normally closed, to light the 'low oil pressure' warning lamp until enough pressure has built up to open it.

The North American anti-runon switch is a special that is normally open, and closes with oil pressure. When the ignition switch is turned off with a running engine a normally closed contact on that operates the anti-runon valve in series with the closed oil pressure switch to earth. Then when the engine has stopped and oil pressure has died away the oil pressure switch opens to release the valve. With the 'wrong' switch the valve doesn't operate until the engine has stopped and oil pressure has died away, then operates the valve and flattens the battery as Patrick says.

In theory you could use a normally closed switch with additional circuitry, like a relay, but then the relay will be operated all the time the car is parked. Less drain, but still not ideal. You may be able to reduce that still further using electronics. There is also a different approach whereby you could use the reduced system voltage once the engine has stopped and the alternator has stopped charging. Or a time-delay circuit that cuts off the valve (and itself) after 10 secs or so.

But if your 73 stops OK anyway, you don't need it. It's only from late 1976 when there was a wiring change including an error that meant the anti-runon valve was the only thing that stopped the engine, not the ignition switch, that it could be a problem.
Paul Hunt

Paul,
Thanks. Great info. I started down this path because I was having starting problems, and then cut-out problems. I found a write up on the Moss web site indicating that the run-on oil switch might be cleaned, so I did that, and all seemed well until the cut-out (even while driving) started happening.
I changed the switch, but it turned out to be the "normally closed" type, and drained the battery quickly. Replacing the old switch did not seem to solve any problems, but I am now pretty sure that it was not the source of the issues.
Now going through the LT wiring. I have good fuel delivery and spark (most of the time) but it seems like a random loss of power (tach drops to zero). If I am rolling, it will seem like a "spit" and then return, but if I am at a stop, it dies and is difficult to re-start.
More investigation in my near future.

Pat
Patrick Callan

Check the condition of the small braided copper wire inside the distributor that runs to the vacuum advance plate. These often become frayed, over time, and give intermittent current resulting in no spark. RAY
rjm RAY

" random loss of power (tach drops to zero). "

That shows current pulses through the coil have ceased. Could be anything in that circuit i.e. ignition supply, coil, points, or any of the connections, including the cloth-covered earth wire inside the distributor that RAY mentions.

The points plate continually twists back and fore as you move the throttle pedal, so this earth wire and the wire to the points are being continually bent back and fore. They are designed to be very flexible, more so than ordinary wire. Do you have electronic ignition? There is a question mark against some types of these, in that the wires aren't as flexible as the originals and they can fracture inside the insulation.

If the ignition warning light comes on as well, then it's the ignition supply from the ignition switch that has failed.
Paul Hunt

I'm going through the wires and connections next. Ignition light does not come on when this occurs.
Before all this started, I had a problem where the engine would crank, but not start, and the ignition light did not come on. That turned out to be the small brown/yellow wire at the alternator - it had come loose. Re-connected that, and it started, with functioning ignition light. Took it for a drive, and that is when this new problem began.
Have a number of other projects going, so I'm not sure when I will finish checking all the wires/connections, but I will post if I can find a "solution."
Patrick Callan

"That turned out to be the small brown/yellow wire at the alternator "

That will stop the ignition warning light coming on, and means you would have to rev it to about 3000 before it would start charging, but it shouldn't prevent it starting. You can unplug the alt altogether and it should start.

Incidentally a short on the wire between the coil and the points can also cause this, e.g. a short-circuit condenser.

If it remains intermittent you can hook up a voltmeter where you can see it in the cabin, connected between the black/white wire at the tach, or at the coil if that's easier, and an fused ignition feed somewhere, but not the coil positive. That should show about 7 volts when the engine is running normally. If it shows 14v when the problem happens it's a short on the coil to points circuit. If it continues to register about 7v but the tach has stopped registering, then the 12v from the ignition switch through the coil has failed. If it drops to zero then the circuit through the points has failed.
Paul Hunt

Ok, notes between other projects:
Found a partial break in the lead to the plus side of the coil. Repaired that, but no change in behavior.
Found that the wiring diagram in the Haynes manual looks more like the '75 at the coil, and more like the '74 at the tach. Confusing.
Tried disconnecting wire to the the oil pressure switch, and the brown/yellow wire to the anti run-on valve. Also disconnected the pipe from the anti run-on valve to the manifold (although, come to think of it, I did not cap it off).
Engine started, and kept running, but kept trying to stop, then re-energizing. May be because I didn't cap off the line to the manifold?
Anyway, making some "progress" although very confusing.
Back to other major projects.

Pat
Patrick Callan

Floor the throttle as you turn the engine off - no running on when you do that.
John Prewer

4-clyinder ignition changed from unballasted 12v to ballasted 6v with the change from chrome to rubber bumpers, i.e. at chassis number 360301 (roadster) and 361001 (GT). The former should have a single white on the coil +ve and the latter two white/light-greens. Both should have two white/blacks on the coil -ve, and the tachs should be the same i.e. a green, a black, and the white/black. Prior to any PO interference of course.

The oil pressure switch should have a slate/yellow to the anti-runon valve, and that has a slate/red to an in-line fuse, and a slate from the fuse to the ignition switch. Brown/yellow goes from the ignition warning light to the alternator, definitely not the anti-runon valve, prior to PO interference as I say.

If the engine is Dieseling when you switch off i.e. a very rough running with the engine perhaps going backwards before it stops, then you do need the anti-runon system. Removing the line from the inlet manifold to the valve disables its function - if the valve is operating. In fact almost any blockage or leak in the emissions plumbing can disable anti-runon. Not capping the line from the manifold lessens running-on, if anything, as letting air into the inlet manifold is what after-market anti-runon valves do, although it needs a lot of air to work.

Flooring the throttle, like stalling it in 4th, is a really naff way of stopping running-on in my opinion. Get it right and you don't need to do either.
Paul Hunt

Another update:
Wiring from the anti run-on valve to the oil pressure switch is, in fact, slate/yellow - looks like brown from years of fading and dirt, but very different from the brown/yellow at the alternator. The + side of the coil has a white, and the - side has the two white/black wires (one of which goes to the distributor. Tach wires are as Paul described.
Today, I disconnected the oil pressure switch, the anti run-on valve, and the pipe going to the manifold from the ARO valve (capped it off this time).
The car started, but I had to hold the accelerator all the way in to get it to kick in.
Once running, I was able to get it to idle. Thought maybe things were good, when power and the tach dropped to zero again. The fuel pump was still running, but the ignition light came on.
So, now I'm going through the white wires and connectors - so far no luck finding anything unusual.
I guess it may also be time to take the charcoal canister apart.
The ignition switch is relatively new (but I know that doesn't rule it out).
Anyway, back in the hunt.
Patrick Callan

Got five used ones on the shelf, all work. tell me where to send one.
GH VanHorn

GH,
Thanks. If I can ever figure out which part is causing all this I may get back to you. Do you have extra ignition switches, or anti run-on valves?

Pat
Patrick Callan

On a North American spec 73 if the fuel pump continues to run when the tach suddenly dropped and the engine died, that tends to confirm that power is reaching the 4-way bullet connector in the mass by the fusebox. This is where the main harness joins the rear, and power goes though that connector to get to the pump. There should be two white wires coming out from the main harness, one will be the feed from the ignition switch, and the other is the feed to the fusebox then the coil. If the bullet on this wire is making a bad connection with the 4-way connector, you will get this problem.

The fusebox is simply used as a junction point, power does not go through any fuses to get to the coil. There should be two white wires at the fusebox, and ordinarily I'd expect them both to be in the same spade connector, so that even if that connector is removed from the fusebox, it will kill the gauges and all fused ignition circuits, but the engine will continue to run.

However if any work has been done such that the two whites have separate spade connectors, each on their own spade on the fusebox, then power does have to go via the two spades on the fusebox to get to the coil. In this case as the gauges are dying before the engine stops then it could be the white carrying power from the 4-way bullet connector making a bad connection with its spade.

Of course, this doesn't take account of any wiring changes a PO might have done.

Several times in the past when faced with intermittent electrical failures I've positioned a voltmeter where I can see it while driving, and connected it to a certain point along the circuit in question, to see if power fails at that point when the problem occurs. If it does, then I move back a bit towards the supply. If not, then I move forward a bit towards the load. In your case, you can really only test the two ends of the wire between the bullet connector and the fusebox.
Paul Hunt

Paul,
Checking all the white wire connections you mentioned. The 4-way connector looked good and tight, but I cleaned it anyway. The two white wires going to the fusebox were still on the single spade, but one of the wires had been "spliced" by a PO (possibly to connect something), and that splice seemed a bit loose on one side. Have removed the old splice, and placed a better connector on that wire. Voltages and connectivity seem to be good across all these connections, but then again, the problem is a bit random.
Now waiting for some better weather, to see if there is any impact.
Patrick Callan

Getting more frustrated by the day!
Wires and connectors all look good - inside and outside the distributor. Wires and connectors to coil look good.
Points, rotor, and cap all look good.
Have purge all the lines to/from the gas tank.
Cleaned all the battery connections.
The car starts, even idles, but then randomly cuts off, with ignition light coming on an fuel pump still running.
Symptom occurring with two different coils.
I guess I could try a new condenser - the one currently in use is almost new, but I understand even new ones can be faulty.
Patrick Callan

>>I guess I could try a new condenser - the one currently in use is almost new, but I understand even new ones can be faulty.<<

Patrick I think if you'd have put this at the start most poster would have that a newly fitted modern made condenser is a likely cause of any trouble

do not fit another newly made condenser unless you get it from the US equivalent of the Distributor Doctor

if you have an original old one fit that instead or a NOS from years ago or a old s/h one - they last decades in use, well the old ones do, new ones, minutes days, weeks if your lucky
Nigel Atkins

Replaced condenser with an old one (working when I replaced for an overall tune up).
Same symptom. Starts, runs and settles to idle, then stops.
Must be something "in the loop" that looks good, and even tests ok for voltage or continuity, but randomly fails.
That's two coils, two condensers, same result.
Going crazy. Looks like a gradual ignition parts replacement is in my near future.
Patrick Callan

Today, one step at a time. Replaced the wire from the coil to the distributor - no change.
Replaced the rotor - no change.
Replaced the dist. cap - no change.
Guess the points are next, but I would think that they would either be ok, or not - as opposed to ok at first, then failing.
Patrick Callan

well the good thing about replacing or testing one thing at a time is that usually it should show up a culprit, unless it's a combination of faults, even then hopefully you should see some incremental improvement

keep going
Nigel Atkins

So the engine stops, but the fuel pump is still going.

Does the fuel gauge continue to register? Or does that drop as well?

Does the tach drop to zero the instant the engine starts to die, or does it only slowly drop as the revs drop?

Does the ignition warning light only come on when the engine has finally stopped? Or as soon as it starts to die?

Clear answers to those questions should localise the problem.
Paul Hunt

Engine stops, fuel pump keeps going.
Believe the fuel gauge continues, but will verify.
Tach drops immediately.
Ignition warning light comes on immediately.
Tried a new (old, but working when changed with tune up) points yesterday. Now it won't even start - have to re-check the points adjustments, and the post where the condenser wire connects. If that doesn't work, will replace with previous points that at least allowed startup.
Patrick Callan

If the ignition warning light comes on while the engine is still spinning, it should then flicker and go out as the engine comes to a halt. Does it do that?

Or does it only come on as the engine stops spinning?

The first scenario indicates the ignition supply has been lost from the ignition switch, but then that would affect the fuel pump as well as the gauges.

The second case indicates that something else is stopping the engine, if the tach drops immediately that could be something in the ignition LT circuit i.e. coil, points, condenser, or an HT or fuel problem, but in that case the fuel and temp gauges, and all the other ignition circuits including the fuel pump, would continue to operate.

So there is still a mismatch in symptoms if the ignition light comes on immediately, but the fuel gauge continues to register.
Paul Hunt

Paul,
Tried to verify the fuel gauge today, but now the car won't even start. Went back to the set of points that were allowing it to start before, and it still will not start. My gradual parts replacement appears to have hit a snag.
If I remember correctly, when the engine quit before, the ignition light would come on and stay on while the fuel pump kept running, until I turned the key to off. I do not remember it flickering or even going out when the engine stopped.
Right now, it is acting like no spark. I checked the points gap, and the post that requires isolation. Both looked fine.
I would try changing the ignition switch if it were not so expensive. Would like to find a way to bypass the switch when in the start and run positions - maybe with a temporary toggle switch, just to verify if the IS is the problem.
Thanks for your continued following of my mystery.
Patrick Callan

Patrick,

I was thinking the same as you about bypassing all the switches you can. If I were working on it, I would run a known hot wire (like maybe from the fuse box) straight to the coil feed side (side that does not go to distributor). I would remove the existing feed when I did this. Then I would see if it would start and run, or at least has spark. I think your car takes a full 12V to the coil, but I am not sure. So, don't run it long like this unless you can verify that it is suppose to run a full 12V.

If you said what kind of fuel pump you have, I missed it or forgot. If you still have the original type of SU pump, it should stop clicking after the carbs are full. So, if you have the SU pump, and still hear it running when the car dies, I would check to see if you have any fuel in the carbs.

Charley
C R Huff

Patrick,
You said earlier that you had good fuel delivery. Does this mean that you disconnected the fuel line and checked the flow?

David
David Overington

Charley,
Aftermarket fuel pump, so it does keep running.

David,
Yes, I did disconnect the fuel line to check flow. I originally suspected the anti run-on system was causing fuel starvation because the pump was doing fine.
I seem to be chasing an elusive culprit.
Patrick Callan

Check you have assembled the points correctly. If you have the type with the stud and nut the tags from the coil and the condenser go side by side, between the two stepped insulators. The stepped part goes into the holes in the tags, i.e. the lower one faces up, then the two tags, then the upper one faces down. Finally the nut.

If you have a voltmeter or test-lamp check to see that you have 12v (with respect to an earth) on the white wire at the coil +ve. If not the problem lies back towards the ignition switch.

If you have 12v there then test for 12v on the other coil terminal. Open and close the points by hand, and when the points are closed you should see zero volts, and with them open you should see 12v.

If you see zero volts all the time remove the black/white wire from that terminal, and test the terminal again, which could give two results:

1. Now you should have 12v on both coil terminals with the ignition on. If only on one and not the other the coil has failed.

2. If you now have 12v on both coil terminals with the black/white removed, then the points are shorting-out somehow.

If you have 12v on the black/white terminal all the time, then the points are open-circuit somehow. Test the points themselves, and if you have 12v on both contacts the earth wire between the points plate and the distributor body has failed.
Paul Hunt

Made some progress yesterday. Followed Paul's troubleshooting steps with a good voltmeter. Got all the way to the last step, and had 12v all the time. Checked the wire from the points plate to the distributor body - it looked fine, but when I moved it, it fell apart.
So, today I am ordering a new wire. My question in that regard is that pictures of the new wires have a ring fitting at each end, where the original one in my distributor had what looked like a crimp-on attachment at the plate. I'm sure I can snip one end of a new wire, but am wondering if that is a good idea - the plate looks like it has a Philips head or slotted screw holding the crimp point but that may also be some sort of rivet (hard to tell without removing the plate). Any recommendations welcome.
I won't be able to tell if that is the only problem I am facing, but I can imagine that as that wire gradually broke, it was causing some of the intermittent issues that got me into this mess.
I will post again as soon as I have a new earth wire installed.
Thanks for the great help so far.

Pat
Patrick Callan

Good progress.

The wire spot-welded to a plate rivetted (with the cross pattern - top in the attached) to the points plate was used on the 23/25D4, the 43/45D4 used a screw at either end (below in the attached). Hopefully the new wire will be long enough to go from the condenser mounting screw (above the condenser, under the screw head) to one or other of the screws that secure the points plate to the distributor body.


Paul Hunt

Paul,
Mine looks just like the top image (23/25D4). The "crimp" point that I mentioned is the small fold-over that holds the wire to the plate. Hopefully I will be able to make that end connect properly with the new wire.
More when I get to that point.
Thanks again.
Patrick Callan

just a small point (pun unintended) in Paul's photo at least the damaged screw head is a Posi(drive) not a Phillips as it has the additional line marks

although many people don't bother about it there are differences between the Phillips and Pozi screw heads and screwdriver points

if possible use the correct type of screwdriver to save damaging the screw heads or at least the right sort of size as usually sooner or later a damaged screw head will cause problems
Nigel Atkins

Hopefully final follow-up:
Obtained a new breaker plate with the ground wire attached, a new condenser, and new points set.
Replaced the entire assembly. The car started up, and idled well. Waited for the weather to clear and took it for a drive. For the first time in a long while, had no "spitting/coughing" and the idle was correct when coming to a stop. what a change.
So, it appeared to be the breaker plate ground wire all along. Not sure if I should post a separate thread to cover my "lessons learned."
Thanks for all the recommendations and comments, and especially to Paul for the trouble-shooting procedure that isolated the problem. Step by step and a good volt/ohm meter found the problem. I might even order another breaker plate assembly as a spare, since it seems to be easier to replace the entire assembly than to try and change points and condenser with the dizzy in place.
Thanks again,
Pat
Patrick Callan

ground wire problems are known about but can be forgotten so post away

great that things are better and great that you've bought a condenser and CB points that work (many modern made ones have been crap)

not to rain on your parade but do also bear in mind that the springs and mechanism in the lower part of the dissy can wear after only a few years use and can be very variable in use if many years old (decades)
Nigel Atkins

Nigel,
While I had the old breaker plate out, I took the opportunity to clean the inside of the dizzy - the springs looked very good (although after this past experience, appearance can be deceiving). We will see how everything holds up.

Pat
Patrick Callan

Pat,
I've bought cars that looked good but now I know better :)
Nigel Atkins

This thread was discussed between 21/04/2014 and 19/05/2014

MG MGB Technical index

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