MG-Cars.info

Welcome to our Site for MG, Triumph and Austin-Healey Car Information.

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 GT V8 Factory Originals Technical - Engine cuts out

I have a rover 3.5 with the 35DLM8 distributor, For some time I have had the engine cutting out when you get to about 4500RPM, the tacho then drops to zero and within a couple of seconds the engine restarts all on it's own, cluth disengaged and not using the starter! which is most peculiar. I am going to fit a new coil this weekend but am wondering if this is tacho related, i believe the low tension circuit to the coil goes through the tacho, the car is a 75 rubber bumper model. I've also got a spare amp for the dizzy but will change one thing at a time.

If it is possibly the tacho can it be bypassed to see if thats the problem. I have had this problem for some time but for normal driving with the V8 there is no real need to use all the revs, but like to get it fixed.

Kevin.
Kevin Jackson

Kevin, I suppose the tach could cause that type of a problem, but I think you're on the right track with the coil. It depends on what type of tach you have in the car, the current loop type or the impulse type. If the former the entire ignition current passes through the loop and an open circuit will cause the engine to quit, but the tach wouldn't be likely to be the cause. If the later impulse type then yes, a shorting capicitor in the tach could short out the coil signal. I'd still go for the coil first. As to bypassing the tach, not easy to do with the current loop type, for the later type, just unplug the sense wire from the coil terminal that runs to the tach.
Bill Young

If the tach drops to zero when the engine first cuts out but is still spinning that definitely tells you it is loss of ignition LT pulses. If the ignition warning light comes on at the same time i.e. while the engine is still spinning that tells you the 12v supply through the ignition switch has gone missing. If the warning light only comes on when the engine stops spinning then the ignition supply through the switch is OK and the problem is closer to the distributor. A short to ground on the tach pulse wire or inside the tach will cause the cutting-out part of the problem. To *isolate* the tach (you can't bypass it as such the tach on a 75, that style ended in 72) you would need to cut the black/white wire feeding the tach where it and the distributor wire go to the coil -ve.

It's actually easier to bypass the tach on the earlier system as one simply disconnects the white from the coil +ve and provides another from the fusebox. It is the later arrangement that is more difficult as both wires are in the same spade and there is a 50/50 chance you will cut the wrong one unless you can identify the distributor wire by unwrapping the harness tape.

But if the engine *cranks* itself and starts itself, together with the cutting-out that implies problems round the ignition switch, i.e. nothing to do with coil, tach or distributor. Take care with that as I have known cars start cranking while unattended, disconnect the battery while parked until you have sorted it. In one case the owner always left it in gear and it cranked itself across two neighbours front gardens before he stopped it. That was bad enough, but continual cranking could cause a starter fire.
Paul Hunt 2

Hi Paul,

I was hoping you would log on an give me the benefit of your knowledge of 'B' electrics.

I think I understand what your saying and will test one thing at a time, and check whether I am getting an ignition light as suggested which should narrow the search.

The part about the engine restarting, after hard acceleration to about 4500RPM the tacho goes dead and engine cuts out, about 3 seconds later the tacho shows tickover RPM and the engine restarts without the starter motor engaging as far as I can tell, but I would have thought I would have heard it engage will listen carefully on the next test, I will print out your suggestions and take it with me when I do a test on Saturday. I do have a battery isolator so no risk of it going for a drive on it's own I hope!

I'll put the new coil on first as that won't do any harm. the existing coil is off the SD1 donor so 20 years old at least.

Many thanks

Kevin



Kevin Jackson

Kevin,

the normal SD1 coil must allways be used with a ballast resitor!
You should check out wheter there is enough current to the coil at high rpm.
You are talking about a restart with clutch disengaged... is your engine still cranking when it restarts?

Ralph
Ralph

Like Ralph I'm also confused about just how the engine is restarting.

If you are saying that the gearbox is still in gear, and the clutch pedal isn't pressed to the floor, i.e. the forward motion of the car is still spinning the engine, then I can understand how the engine is 'restarting', which will simply be the ignition fault 'fixing' itself i.e. the broken connection reconnecting.

But if you are saying the clutch pedal is pressed down, or the gearbox is out of gear, and the engine is no longer spinning, then the only way it can restart is if the starter is energised. And if you are not doing that yourself, that would be very unusual and indicates major electrics problems.

A 75 i.e. rubber bumper should have a ballasted ignition feed, and so needs a 6v coil. But you need to confirm exactly what you have in both cases as the ballast can be bypassed, and coils can have very misleading information on them. With coils both terminals must have the wiring disconnected and the resistance between them measured. A ballasted system needs a coil measuring about 1.5 ohms, or slightly lower if a sport coil i.e. about 1.2 ohms. If you measure 3 ohms that is a 12v coil for a chrome bumper MGB with unballasted system. With anything other than points it is a bit tricky determining whether you have a ballasted ignition feed or not. Generally if you have white/light-green or white/light-blue wires on the coil +ve that indicates it was ballasted originally, but could still be bypassed. If you have white it was originally unballasted i.e. needs a 12v coil, but a ballast resistance could have been added. If any other colour it could be either. The only real way to check is to measure the resistance between the wire removed from the +ve terminal of the coil and the white or white/brown at the fuse box. If that shows about 1.5 ohms then it is ballasted and you need a 6v or 1.5 ohm coil. If zero ohms or very near it then it is not ballasted, and in this case you need a 12v or 3 ohm coil.

In theory you can remove the wiring from the coil -ve terminal and connect a known good earth to it instead, then with the ignition switched on measure the voltage at the coil +ve, which should be about 6v. However this only shows problems if only one component is incorrect. I have known cases where a specialist (who should know better) fitted a 12v coil and an *additional* ballast resistance. The result was the correct 6v at the coil, but only 1/4 of the current through the coil and hence only 1/4 of the spark.
Paul Hunt 2

When I did the conversion we bypassed the ballast resistor with a new feed from ignition to the coil as I had been told that the SD1 or Range Rover with the 35DLM ran off a 12v coil.

The engine is re starting with the clutch pedal down, and as far as I can hear, no assistance from the starter motor, I assumed there was just enough flywheel effect to get it going, but just before it starts the tacho indicates tickover speed and then the engine cuts back in!

Some bright spark! suggested an exorcism.

Kevin.
Kevin Jackson

Kevin,

the SD1 setup normaly made use of the 35DE8 and a Lucas 22 C 12 coil with ballast resitor BBR. Some later models used another electronic distributer but
kept the coil with ballast resitor.
The coil shold be stamped at the bottom.

If your mechanical distributer has not had the points changed for some time, fit new ones and check the advance (static 8° with vac disconnected 10° at 1200, 17° at 1800 and 22°at 2600rpm), may be the spring has become weak so the points might not meet at higher rpm and if you do not run the ballast resitor, you should try a good new 12 Volt coil first.

Hope this helps

Ralph
Ralph

Kevin,

i forgot to mention that your problems at 4500 rpm+ might also be due to fuel presure.
As you did not mention the carb you use, all Carter, Edelbrock and Holley must be fed with a pump capable of ~ 6,5 Psi presure that might drop down to 4 Psi under heavy excelleration. Mikuni, Facet or SU pumps do not suit such carb installations for this reason and you will have the symptoms of ignition cut off at high rpm although it is fuel starvation.
Just check fuel presure and you will know.

Ralph
Ralph

Ralph,

It's definitely an electrical problem because the behaviour of the tacho but your also right about fuel pressure. I'm using an Edelbrock/Weber 500 and the book states a fuel pressure requirement of 5.5 psi.

I have a new facet red top pump to fit and a fuel pressure regulator and guage so can set up the fuel supply properly.

I am going to work on the car today and hope to get the electric sorted, taking large hammer just in case!

Kevin
Kevin Jackson

If the cut-out is only brief then it's quite possible that the engine is still spinning, even with the clutch pedal down, so when the problem clears the engine simply fires up again without ever having actually stopped turning. I'm going to stick my neck out here and say that is the only way it can possibly restart if the clutch pedal is down and the starter motor isn't engaged.

This what I mean about coil markings being confusing. For the 4-cylinder coils were HA12 for the chrome bumper i.e. unballasted i.e. 12v. For the rubber bumper and V8 which were ballasted i.e. 6v coil required it was 15C6, 16C6 or 32C5. It is tempting to assume that the suffix number indicates the nominal running voltage, the 32C5 giving a bit more spark energy at a nominal 6v for later North American applications which needed the extra oomph for the lean mixture. If so, that would make the 22C12 also a 12v coil and hence should be run unballasted. But http://www.roversd1.co.uk/tech_data1.htm lists variously a 22C12 with 1.3 to 1.5 ohm ballast for the first application as Ralph says, the same coil with no ballast mentioned for the next application, then a 32C5 with no ballast listed, then a 32C12 with no ballast listed albeit as 'ignition module' rather than 'coil' as previously. I've also seen coils labelled '12v' but also a comment that it needed a ballast resistor. And don't even mention internal ballast (which is irrelevant, by the way). The only way you can be sure you have the correct combination is to measure the primary resistance of the coil you are using, and the resistance of the ignition feed to it.

The bottom line is that if you bypassed the ballast resistance in the car, and fitted a standard V8 coil, you are almost certainly running it at double voltage and current which will quite likely overheat it which can cause intermittent ignition problems.
Paul Hunt 2

Thanks Paul.

I was going to work on the car today but it is to damn cold, and as it's not urgent did some work round my flat instead.

The coil I bought from a Lucas agent is from memory a DB198 which is the correct coil for either the late SD1 or Range Rover from I think 1986-1993 so should be ok.

The old coil is getting fairly hot and I'm guessing it's breaking down under load when asked to provide a lot of sparks at higher RPM.

I'll try and get over to the car tomorrow and get it fitted and let you know how I get on.

Kevin.
Kevin Jackson

Kevin I've struck this problem with a friend's V8 converted MGB It ended up being a dud condenser in the ditributor. His car had the same symptoms as yours with the tacho needle dropping down and the engine cutting till the revs dropped back a bit and then off it would go again It might just pay to check the polarity of your ign. coil as well and make sure it's around the right way
Goodluck Gary
V8 Gary

Probably DLB198. I haven't ben able to determine whether that is a 6v coil or a 12v, although it is described as being for use with the original Rover V8 electronic ignition, and so is probably a 6v i.e. it needs the harness ballast resistance in circuit. I don't have a diagram for V8 SD1s but those for the 4 and 6 cylinder cars from 77 to 79 had ballasted ignition, as did the factory MGB V8 of course. I rather suspect that your problems ar being caused by the ballast having been bypassed.

Coils do get pretty warm anyway. Points ignition system generally maintain that temperature through the rev range as the points system results in the proportion of on time to off time remaining constant. By contrast DM and DLM systems use constant dwell, which means they will get hotter as the revs increase. However they should actually run *cooler* at loer revs than points coils, meaning the heat increase is only relative.

A short-circuit condenser will indeed cause these symptoms. And while the coil should be connected the right way round the difference is marginal and won't be causing this.
Paul Hunt 2

Does the car have an ignition relay ? If so, I would replace that as a matter of course. Intermittant faults are always the hardest to cure. Barrie E
Barrie Egerton

A 75 wouldn't originally had an ignition relay. However problems with the relay are indicated if all the other ignition circuits fail as well like wipers, washers, heater fan, indicators etc.
Paul Hunt 2

Still not fitted the coil because I wanted to make sure that I had firstly the correct coil and was connecting it to the proper voltage.

After talking to Lucas who of course are no longer Lucas, and getting more conflicting information it was recommended that I speak to H & H Ignition solutions.

They state that the DLB198 is the correct coil for the 35DLM8 distributor and requires a full 12v supply with no ballast resistance.

They also said that the symptoms I had previously described could be related to a failing amp on the distributor so will change the coil first, then if no improvement I will swap the amp for a spare one I have.

With the weather warming up this weekend I'll get it done and at long last report back hopefully with positive results.

Kevin.
Kevin Jackson

Problem solved, it was the coil.

I fitted a new Lucas DLB198 which is the correct coil for a late SD1 or Range Rover 1986 - 1993 and has a full 12v supply with no ballast resistor.

The car now revs past 5000 RPM with no problem and is running a lot smoother than before,and also will tickover at 800RPM or even a bit lower, minimum before was a 1000RPM any lower and it got a bit lumpy.

So very pleased it's running properly,but will have to resist the temptation to drive it like a hooligan!

Thanks for all the advice

Kevin Jackson.
Kevin Jackson

Forgot to mention that the old coil was a Bosch unit and was not marked with the conventional + and - on the LT terminals, it's actually marked with a 1 which I originally took for a minus and it has 15 on the other terminal, any one make any sense of that just out of curiosity.

Kevin.
Kevin Jackson

The 1 and 15 are the DIN numbers (Deutsches Institut für Normung e.V. i.e. the German Institute for Standardization), not surprising since it was a Bosch, see http://en.wikipedia.org/wiki/DIN_72552.

Are you absolutely sure you now have the right combination? Did you measure the resistance of the coil and confirm whether ballast resistance is in the ignition supply or not? If over-voltage were the problem damaging the previous coil, then fitting a new coil of the same type *will* fix it, but the new coil will fail prematurely as well. I only ask because in my experience most systems of that type changed to ballasted feed and 6v coil in the mid 70s because of its better starting.
Paul Hunt 2

Thanks for this thread Kevin. I've been having a similar issue with my SD1 with EFI.

I've been chasing the EFI but will now look at the coil.

Edd Weninger
Edd Weninger

Ed, Your welcome.

Paul,

The Lucas agent I bought the coil from looked it up in the book and said it was the right coil for the 35DLM8 distributor and was 12v so no ballast required.

This was independently confirmed by H & H ignition solutions, so I'm hoping all will now be well.
Incidentally the new coil does not get anywhere near as hot as the old one.

I looked up the link to the Din. standard and it states that 15 is positive from an unballasted supply so I did accidently have it correctly wired as I assumed the number 1 was a minus sign!!

Kevin.
Kevin Jackson

In addition to above if the Bosch coil has 15A positive connection it indicates a ballasted supply is required.

Kevin.
Kevin Jackson

If you are still struggling with a misfire it may be a mis-match between the ignition system and the coil. the DLB198 is a 12v coil so needs no ballast, however it is also designed for use on constant enery ignition systems where the dwell in terms of degrees varies with rpm which ensures that the coil receives a constant charge time measured in milliseconds regardless of engine speed. This allows the coil to be wound with a low primary resistance which allows a faster soak time giving a better spark at higher RPM. However if you use one of these coils with a points type ignition system or an electronic ignition system which is not of the constant energy type at moderate revs the coil will be charged for too long and will overheat and fail.
I'm pretty sure that the 35DE8 with the integral ignition module whilst being an electronic coil was not of the constant energy type whilst the 35DM8 used an external ignition control box and was a constant energy therefore for a 35DE8 one should be using a DLB105 coil (still no ballast required).
Bruce Chambers

Bruce,

Thanks for the additional information, the 35DE8 I think was the original Opus distributor for the early SD1. The second was the 35DM8 with the remote amp.

The distributor I have is the 35DLM8 with the attached amp and was the last type of distributor used prior to going to a distributorless ignition.

From what you have said it confirms the DLB198 is the correct coil.

Kevin.

Kevin Jackson

I don't know how 35DLM8 compares with either 35DE8 or 35DM8. But 35DE8 is the same as points i.e. it isn't constant energy but constant duty cycle/dwell, so needs a ballast if the coil measures 1.5 ohms. 35DM8 is constant energy i.e. has a constant charge time, but on the applications I have seen also used a 1.5 ohm coil on a ballasted systems. Very low primary resistance coils are used with coils that are energised with a very short pulse. From what I've seen the 35DLM8 is a modification of the 35DM8 with basically the only difference being the electronics module attached to the distributor body rather than separate. As such the implication is that it needs the same coil and ballast resistor as the 35DM8. I'm afraid I've lost any confidence I had in professionals since I saw a certain V8 specialist send a car out with a 12v coil and *two* ballast resistances all in series, and the coil consequently running on 1/3 power.

Personally I'd still like to know just what the primary resistance of the DLB198 is.
Paul Hunt 2

Paul,

I'll be happy to measure it next time I get the car out. I was hoping to do some small jobs on it this weekend but the forecast is so bad I think I'll give it a miss.

Refresh my memory, what do I do to measure the resistance. I do have a meter but which setting do I need and how do I connect it to the coil,

Kevin.
Kevin Jackson

Take the wiring off the spades (you can leave the HT connected), and with the meter switched to a 'low' range just connect the probes to the coil spade terminals. Digital meters are more likely to have a low range, analogue less so, especially if they are intended for general use rather than specifically automotive. If an analogue calibrate it with the adjuster screw over the pivot of the meter needle to show infinity with the probes not connectee to anything (not even your hands), then connect the probes together and use the calibration knob to show zero. Then connect to the coil. If either type of meter shows zero again then try the other ranges. If the meter has a suitable range it should clearly indicate either around 1.5 ohms for a 6v coil (needs ballast) or 3 ohms for a 12v (no ballast needed).
Paul Hunt 2

Paul,

For the sake of clarity I will do the suggested test, hopefully next weekend if the weather is better and I can get over to the car, which is kept in a nice warm garage but unfortunately not where I live and just popping out to do the odd little job is not feasable.

Kevin.
Kevin Jackson

Same as me and the roadster. I keep the V8 in my garage at home and the roadster at one nearby.
Paul Hunt 2

This thread was discussed between 13/02/2008 and 10/03/2008

MG MGB GT V8 Factory Originals Technical index

This thread is from the archives. Join the live MG MGB GT V8 Factory Originals Technical BBS now