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 - Ignition puzzle

On starting my 1980 roadster yesterday from cold with the choke, it started normally and settled down at 1500 rpm happily.
As soon as I attempted to rev the engine it stumbled an nearly died. The rev counter fell to zero and was flicking around wildly. If I took my foot off the throttle it settled again without misfire at 1500rpm. Same symptoms several times.
Strangely as the engine warmed, the throttle response improved, until as it reached temperature it was completely normal.
The car was faultless in a drive shortly after.
If this is a fault in the ignition, why would it only occur in a start-up/cold situation? I recently replaced points/ condenser, removing distributor to fit. Could the timing now be disturbed? The body of the dizzy was not rotated.
ph barton

The timing should be checked after replacing the points.

Even altering the points gap will change the timing.
Dave O'Neill 2

The flicking tach definitely shows the ignition LT circuit being interrupted by something else in addition to the points. The fact that it happened when you touched the throttle says to me it is a fractured wire inside the distributor, either the one from the coil to the points or the one from the points plate to the body. When you move the throttle you alter the distributor vacuum signal which moves the points plate thus bending these wires. A fractured copper wire will be temperature dependant in that as it warms up it will expand which will tend to push the broken ends together, so 'repairing' it. Until the next cold start.

Also as Dave says, after doing *anything* with the points or distributor you must recheck timing. Whilst marking the distributor body and block enables you to remove and replace the distributor close enough for it to start and run, it only takes a very small change in rotational position to affect running in various ways (but not normally as severe as this). Having replaced points and condenser you also need to check everything is correctly assembled, tight (where it should be), there are no trapped wires, and the points plate is free to rotate.
Paul Hunt 2

PH,

when the problem only is common if the engine has cooled down for some hours, i would check all the connetors to the coil and to the dizzy for tight fit first, as when the engine is warming up, material expands and can make a better electrical connetion then but Paul has mentioned this allready better than i can. If there is no tight fit when cold, take off each connector and bend it carefully with a pair of pliers first. If this does not cure the problem, new points and a new condenser/capacitor should be on your list too. Paul also explaned that wires in the LT circuit can break, although not that usual, i would also check the wires at the connectors and examine every wire in this circuit that has a non flexible isolation anymore.

Inside the dissy, there is an other thin wire with black fibre isolation also allready explaned by Paul. It is high flexible and connects the terminal of the dissy with the points. Check it for damage and that there is no fibre of the copper core of this wire projecting through it's isolation.

Hope this helps

Ralph
Ralph

The distributor was a new item replaced recently. I did notice when I changed the points that about half the copper strands at one end of the earthing wire inside the dist. were not connected to its spade terminal. I removed it and soldered each end to improve the connection. It has run well for 100 miles or so until this problem arose.
I will replace it with a new one and by pass the wire from the coil to points and see if things improve.Will reset timing also. Thanks for the clues.
ph barton

Double check your solder joints too. Its difficult to get solder to hold to the steel plates - you need a lot of heat, lots of flux, and clean surfaces to make it happen. The distributor may not be right for the application, making the advance wrong, which in a high idle (cold engine) situation can make the car run poorly.
Jeff Schlemmer

That earthing wire consists of strands of cotton and strands of tinsel copper braided together. This has terminals crimped onto either end as it is not possible to solder this type of 'wire', so it is quite possible that the only thing left securely held in the crimp is the cotton, the copper strands largely having come away and only intermittently touching the points plate or distributor body close by.

FWIW I had a Scimitar GTE with Ford Essex engine where the distributor didn't have such a grounding wire anyway but relied on sliding contact pads between the points plate and a base plate. Used to give me heart failure by stuttering just as I pulled out into traffic, being an auto made it worse. I think installing a ground wire was the first tip I got from the owners club, and solved the problem. Conventional insulated copper wire will work as a stop-gap, but won't have the longevity of the pukka stuff.
Paul Hunt 2

I can't find a part number for the distributor earth lead. Does anyone know a source in the uk....Moss is my local supplier.
Phil Barton

It's not listed even in the Leyland Parts Catalogue, and although my local parts supplier is aware of the item they don't stock it. You might have to find a distributor repairer and see if you can beg one, or you could try Chris Betson of Octarine on 01702 526720. As I say as a stop-gap you can use ordinary insulated wire.
Paul Hunt 2

There is a braided copper wire used for removing solder from printed circuit boards that will work well for that application. It is used to wick away solder. I bought it in the states at Radio Shack the last time I bought some. I have used it successfully on several distributors. Might be worth a look.

John in Music City
John Lifsey

Good point, is is very flexible albeit un-insulated, but that shouldn't matter as everything it is adjacent to is at the same potential anyway. Just don't apply too much solder or too much heat to the terminal tags or you will end up with a solid wire!
Paul Hunt 2

I managed to take time to remove the dist. and check the earth wire. It seemed in good shape, with full movement and no breaks. Checked dwell and timing set at 10deg btdc.
Replaced LT wire with substitute from coil to dist. same symptoms.
6v at coil + with points closed. 12v at coil -ve points open.
The car is still hesitating and missing badly..but only as described when warming up.
I noticed when checking coil with ign on that the fuel pump was ticking every 5 seconds or so. soon became evident that fuel was coming from overflow to rear carb. Gave carb gentle tap with rubber mallet. when fuel had evaporated away started car and no overflow. Again running OK once warm.
I know the tach was bouncing around suggesting LT problem ....but could this actually be a fuel problem??
Have bought a new needle valve kit and will take a closer look at carbs.Fuel staining on carb suggests leak has been going on for a while but not noticed.
ph barton

A bouncing tach is *definitely* an intermittent connection in the ignition LT circuit. This must be dealt with first (OK, after dealing with fuel leaks), anything else will give meaningless results.

Hesitating and missing when cold usually means the mixture hasn't been enriched enough i.e. insufficient choke. But paradoxically a faulty float valve will result in a *rich* mixture. So if it was overflowing yes you have fuel problems as well, but while you have a bouncing tach you still have ignition problems.

Your voltage tests do indicate that you have the correct combination of coil and ignition ballast, but you also need to check that the solenoid is boosting the coil voltage during cranking, as failure to do so will result in a weak spark making starting more difficult than it needs to be. With the engine stopped and the ignition on disconnect the wires from the coil -ve and connect a good earth to the spade instead. You should see 6v at the coil +ve. Now crank the engine (it won't start) and the voltage should go *up*. It all depends on how good your battery and its connections are, but when good in both cases you should see about 10v on the coil +ve. This shows the coil voltage is being boosted by the solenoid as it should be. If the coil voltage drops to 5v or lower then the boost circuit is not working, check the white/light-green connection at the solenoid. A voltage between 5v and 10v indicates that either your battery or its connections (or both!)aren't as good as they should be, which will also contribute to difficult starting.
Paul Hunt 2

Thanks Paul. I have been away for a few days and have just started the car from absolute cold.
It fired up after a couple of seconds, but again any attempt to open the throttle and it stumbled with a hint of a backfire in the exhaust. The tach fell to zero for a second on several occasions and bounced around. Felt like it was only managing to fire on one or two cylinders then briefly picking up.
I'm off for a day or two, so will go through the checks you suggest.
Could there be an internal fault in the coil or is this out of the question given the previous tests??
ph barton

Certainly could, and indeed that was a common cause on older MGBs with rivetted spades - they worked loose. Subsequently they changed to threaded studs and nuts which cured the problem. I'd expect yours to have studs and nuts, but it is worth a look, and check the nuts and spades are tight.

Another check you can make is to connect a voltmeter to the coil +ve while the engine is running. For a points distributor I'd expect to see something between 7v and 14v. This is based on the dwell, which is 51 degrees for a 45D4 distributor, or about 57% (51 degrees of the 90 degrees per ignition cycle for a 4-cylinder), which if my calculations are correct should result in an average as displayed on an analogue meter of about 10v. But the exact voltage isn't as important as what it does when the tach flicks and you get the missfire, though. If the voltage flicks *up*, then you know the circuit is intermittent through the coil and points etc. to the distributor body. But if it flicks *down*, you know the break is between the coil +ve and the voltage supply through the ignition switch.

If it is the former i.e. flicking up, then with the meter connected to the coil -ve this time, again you should get a steady voltage but this time between 0v and 14v, and based on the same calculations as before should be about 6v. But again the important thing is what happens when the tach flicks and the missfire occurs. If it flicks up then again the break is through the points to the distributor body i.e. you have proved the coil (the LT circuit at any rate) is not the cause. But if the voltage flicks down this time (having flicked up when the meter was on the +ve) then the coil *is* the culprit.
Paul Hunt 2

Thanks for that. I think I will read that through a few times and have a go!
I managed to get another LT earth lead from an old distributor. Good continuity with no breaks. Fitted that with new condenser (on the off chance )but starting from cold the symptoms were similar but perhaps the tacho wasn't quite so wild....didn't drop to zero anyway. My tacho seems a bit strange anyway...loves to do nothing for 10 seconds or so on starting and then flick into action.
If all is well in the distributor I suppose the coil is the next suspect.
What does worry me though is if, once warm, the engine behaves faultlessly AND the LT wiring in the disributor is OK what is left??
Phil Barton

It sounds to me like you have 2 issues, or more! The first and most important is that your replacement distributor may not have the right ignition curve - so you're getting the wrong advance rate, which makes your fuel mixture improper for the state of tune of your engine. The advance rate has a direct correlation to the fuel mixture and fuel sounds like your bigger issue right now. Where is the dizzy from - is it new and if so which one did you install? A high idle speed will put you mid-way into the ignition curve.

The second issue is carb tuning. Any time you change the distributor and the advance rate, you need to change the fuel mixture to match. One way you can help compensate for that can be to advance the timing to 14 BTDC at slow idle with the engine warm and the vac line disconnected from the distributor and plugged. You could probably still use a good carb balancing and adjustment of the choke setting.

Please let us know how you've progressed!
Jeff Schlemmer

The tests - if you can get it to misbehave and keep doing so while you are testing! - should prove the coil one way or the other and indicate whereabouts any break in the ignition LT circuit is.

However the tach itself may be cause of its own misbehavior now you mention it takes a while to start responding. Often this can just be a sticky pointer and rapping the glass with a knuckle will unstick it. But the same rap could shake any internal bad connections, and it can also experience the 'when it warms up it works better scenario'. If the tach erraticality is in synch with the misfire then the tach probably *is* a symptom rather than its own cause. But if the two are independant, even only some of the time, then it is a case of almost certainly having more than one fault as Jeff says.
Paul Hunt 2

I changed the coil for a new one (at £9.95 it would always do as a spare),It made no difference at all.
So the whole LT system has now been checked from coil to Dist.
I will now check the wiring from the battery/ solonoid/ ignition switch/ relay, clean it all up and check for continuity.
I have a couple of days off later in the week, so I shall fix the rear carb. overflow, give them a good clean( given they were recon exchange only 300 miles ago, and use Pauls spanners guide and my original factory workshop manual which covers the tuning process well.
Jeff, I have set the timing to 10 deg @ 1000 rpm as per the manual for 18V 847F engine, would advancing to 14 deg create any pinking?
Paul, the tacho doesn't seem to drop to zero now and may be dropping in time with the stumbling of the engine.
Will keep you posted on any progress, and thanks for the support.
ph barton

Phil. Much of how far the initial timing can be advanced will depend on the advance curve of the distributor. That is where you need to start when determining what the actual timing should be. As to pinging, only testing will determine when that happens as it is a function of the state of tune of the engine, the distributor, the fuel mixture and the fuel being used.

It might be interesting to take the tachometer out of the circuit and see what happens. I believe, on your model, a tach with a problem can create a low tension circuit problem. Paul will know more about this than I.

Les
Les Bengtson

Les, I did run a new lead from the coil to the dist. with no tach as a test with no effect.
The new dist. came from either the MGOC or MGBHive a couple of years ago, I will have to check the records. The car has been stored until the summer.
The dist. has no tag or ID on it, but is a 45D4. I would have given the car year when purchased, so presumably they would have matched the adv curve....I believe a 41610?? Or am I being too trusting??
ph barton

A tach dropping in time with the misfire definitely indicates an intermittent connection in the LT circuit. The results of the voltage tests should indicate which side of the coil this is occuring. Personally I don't like digging in willy-nilly, as you may disturb the fault without actually finding and fixing it properly, and although initial results may be good it will come back on some dark and stormy night. Any problem connections round the battery and battery cable connection on the solenoid are more likely to reveal themselves as a failure to crank. Likewise bad connections on the brown wires up from the solenoid are more likely to reveal themselves in problems with the headlights and other circuits.

Is this a UK car? If so the coil is not fed from the ignition relay but direct from the ignition switch. If the ignition supply from that fails then I'd expect the ignition relay to be chattering, at least some of the time it misfires.

Les - yes if the wire to the tach or the circuit inside the tach is intermittently grounding then it will short out the points and stop the engine. This is actually a nifty anti-theft measure for 73 and later cars. But I note Phil bypassed the tach and the problem still happened.

If the distributor doesn't have a reference number on it then you are *definitely* being to trusting! Even with a reference number in my experience rebuilt distributors often bear no relation to that number. OTOH the 41610 was used for the whole of rubber production in the UK (unlike North America which got dozens of different types) so it should stand a better chance of being correct. A wrong curve won't cause this though, at worst it would result in pinking when set to the correct timing, but that is likely to happen with today's low octane fuels anyway, and even more so if you up the advance from 10 to 14.
Paul Hunt 2

Phil,

if you want to know the exact curve of your dizzy, take it to a service garage that can run it on a test bench. There the parametres of the curve can be read out to give clearance upon the setting, but this will not help to cure the problem you are talking about.

I agree with Paul to check all the connections starting at the solenoid, then check their way to the fuse box and from there to the ignition switch, tach, coil. If there are capacitors fitted at the coil, put them away. Check the body of the coil to have proper contact with the body of the car.
For the wiring diagram, advanceautowire.com is the best source you can find your car's specs there and then it would be the easiest way to use a jumper cable to locate the fault.

Hope this helps

Ralph


Ralph

FWIW the case of the coil is electrically isolated from the windings and contacts. The 'other' end of the HT winding is connected to the CB terminal on positive ground cars, for example, the HT current flowing through goodness knows what (like the LT winding, ignition switch and battery) before it gets to the body and engine ground and hence the 'other' side of the plugs to complete the circuit.

I don't know about Germany but I seriously doubt Phil would find a distributor testing machine in the UK anymore, short of sending it off to a rebuilder to be rebuilt, and then I have my doubts that short of actually visiting and seeing it in use, any of them actually have one.
Paul Hunt 2

Why not consider putting a jumper cable direct from the battery. With all other wires removed excepting dist to coil this will isolate your problem. Try it with and without tachometer.
Sandy
Sandy Sanders

Paul,

we have service sations in here, that care for Bosch car eletrics and they are all equiped with a test bench for dizzys, generators and alternators. Many of them also offer rolling road tests too. Is there no adequate Lucas service in the UK? Even workshops that are specialized on French or Swedish cars or VW often have a test bench for dizzys if they are in buisiness for 20 or more years and there are a lot of 'modern' cars still driving around with this old technology ignition systems.

Ralph
Ralph

Thanks for the input guys...I shall be working through your suggestions over the next day or two.
My brother in law reminded me of a fix done on my 1959 mini when it was rather ancient. The dashpot oil was replaced with gear oil to stop the worn SU piston rising too fast on acceleration,causing the engine to stumble.Is it possible that this could be part of my present problem?? The carbs have been reconditioned, but I imagine I could be unlucky.
When I took the car to a MG "specialist" recently they could not find a thing wrong..but did say they had to fill the dashpots.
ph barton

If it cranks then there is nothing wrong with the battery or its connections. If the lights work properly then there is nothing wrong with the browns coming up from the solenoid either. Disconnecting the tach made no difference. You really need to do the voltage tests on the coil +ve and -ve for further diagnosis.

Stumble on opening the throttle can indeed be due to insufficient damping, but this is probably due to mal-adjustment or wear as you shouldn't need thicker oil in winter, and in any case 80 or 90 wt gear oil has almost the same viscosity as 20W/50 engine oil. Stumbling when cold on a steady throttle is just insufficient mixture i.e. choke, but neither of those things will cause the tach to flicker any more than a change in rpm would warrant.

'Fill' the dashpots is not correct. If you unscrew the damper cap, lift it up, press it down and feel the resistance of the oil before the cap reaches the piston cover, you have enough. How much before shows you have much of a 'reserve' you have. Some diagrams show the oil level above the reservoir and some below. If you try to keep it above it will get sucked dowm the small clearance between the outside of the piston reservoir and the inside of the cover tube and burnt.

Mosts garages here haven't dealt with distributors for 15 years or more. They are no longer mechanics but fitters, as all they can do it look at a computer screen and swap parts. And even mechanics knew precious little about anything electrical 15 years ago. It is all part of the general malaise affecting Britain where engineering skills are generally no longer valued, practiced or taught. There are rolling road services around, but again they are unlikely to offer a recurving service as well. I spent some time researching distributors and centrifugal advance springs and didn't come across a single reference to a distributor tester in the UK, only one in commercial use in the US with John Twist, and another in private hands.
Paul Hunt 2

I have at last got to the bottom of the problem....or should I say problems.
As there was a flooding problem on the rear HIF carb, I bought a needle valve/seat kit and yesterday took the carbs off for a good look.I removed the float then the valve and seat.
I was astonished to find a piece of plastic material lodged next to the valve. It measured 6mm long x 3mm...irregular in shape with a consistancy similar to insulation tape.Clearly the needle had been unable to move effectively for some time. Remember these are recently reconditioned carbs...the material must have been left in the float chamber as it certainly could not have entered through the hole in the needle valve seat.
Once that was done I flushed out each orifice with carb cleaner and was amazed when a substantial amount of black ,sand like debris flew out.This must have been blasting material from when the body was cleaned.I was very careful to make sure that all was spotless.
I noticed that the jet in this carb was a lot higher than the front one.
With that in mind I started the carb setting from scratch.
Started her up and it was like a different car. Perfect throttle response and no hint of a misfire!I was delighted.
However, the tacho was still misbehaving! Not moving for ages and bouncing around at times. So the tacho has misled us into believing it was LT problems.
Thanks for all your supportive suggestions...if only the carb had been reconned more diligently!
ph barton

Phil. Glad you solved the problem. You also learned why most of us do our own reconditioning whenever possible. There are a couple of people, here in the US, whom I would trust to do a good job. But, they are a long distance away from where I live. Hence, I do carb recons myself as do most of the people I know.

As to the tach. If it is possible to obtain a second unit, one which can be tested to ensure it works properly before installation, it might be worth installing it to see what happens. Right now, we have either a problem with the wiring or a problem internal to the tach. A dwell/tach meter, wired into the existing system, would allow you to see what is happening and is easy to connect.

Les
Les Bengtson

I recently bought a 1972 MGB and cant get it to fire. I am not big on the points distributor deal and am not sure how to go about making it fire. any tips would help this is my first MG and I really love the car.
eric

Eric. Please go to my website, www.custompistols.com/ and click on the MG section. I have some articles on the ignition system, including one of trouble shooting it to find where a fault may lie. Dave DuBois has some articles on the fuel system, including how to trouble shoot it. Paul Hunt's website, "The Pages of Bee and Vee" will have some articles which will be of use.

The phrase, "can't get it to fire" can mean many things. It can mean, "I cannot get a spark at the plugs", and ignition system problem, or it can mean, "The engine does not start when turning over", an ignition or fuel system problem.

An engine takes three things to start and run--fuel, spark, and compression. The two websites will give you information on how to check out all three.

Les
Les Bengtson

A timing-light clipped on to the coil lead and each plug lead in turn should be the first diagnostic step for a non-starter, just watch for the flashes which should be regular and consistent, just four-times more often on the coil lead of course. Whilst on leads 1 and 4 point it at the timing marks as well and check it is about 10 degrees BTDC (disconnect the manifold vacuum pipe first).

If you get flashing on all leads then take a couple of plugs out after a few seconds cranking and sniff them. You should get a strong smell of petrol but they should not be soaked.

If soaked they are flooded, which will happen with too much cranking with the choke out if there is some other problem, cranking with choke in and throttle wide open should clear it and fire if that was all that was wrong, get ready to release the throttle and half-pull the choke.

If no smell of petrol there is no fuel getting through, did the fuel pump chatter when first turning on the ignition?

If you get flashing on the coil lead but not the plug leads the cap or rotor are probably breaking down.

No flashing on the coil lead means check the LT circuit with a voltmeter. You should have 12v on the coil +ve (white) with the ignition on, and not much less than 10v when cranking. On the coil -ve and with an analogue meter (a digital meter may not give a useable reading) you should be able to see the effect of the points opening and closing during cranking, averaging about 5v.

But if this engine hasn't run in your ownership then any number of things could be wrong, including plug leads on the wrong plugs, timing way out, faulty fuel pump, or just generally in a very poor state of tune.
Paul Hunt 2

This thread was discussed between 13/11/2007 and 12/12/2007

MG MGB Technical index

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