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 - 35DM8 & CEI

I'm currently setting up my GTV8 with a 35DM8 which I have re-built. But not having the AB14 box with it's GM HEI (CEI) amplifier module I bought a module and mounted it on a 100mm square of Aluminium heat sink.

My, slight, problem are the connections to the module.

+ and - are easy and well documented.
The issue is with the inductive pick-up wiring.
There are plenty of references to purple and orange wires from the PU coil, which would be fine and dandy, except my coil just has 2 black wires.

Does anyone know whether the purple wire was the one nearest the base plate or nearer the cap?
Allan Reeling

It probably does matter from a polarity point of view, but I'm wondering if you can just suck it and see, plugs out (or no fuel in the carbs) and looking at the timing when cranking.
paulh4

Purple on top and goes to the male pin of the connector
Orange closest to baseplate and goes to the female pin(socket)of the connector
William Revit

Thanks Willy, just what I needed to know.......almost!!! No "selective' plug so just using a waterproof plug on the side of the Dissy.
So the purple to "G" on the module?
Although I've seen it shown the other way, but the consensus is purple to "G".
Allan Reeling

Just an update.

I bought, off fleaby, a used A14 amplifier box for the princely sum of £20. It tests perfect and with a minimal bit of cleaning up and a coat of WRINKLE finish is ready to be installed in my Roadster conversion.

After Willy's help, and bit more delving, as well as looking at the "ossified" wiring which came with the AB14, it seems the purple (top) wire goes to the "W" 6.3mm spade on the HEI.
Allan Reeling

Even another update.
With the module wired as described above, no chance...........bag of nails!!
Swapped the pick-up wires around and it runs smoothly.
But my dial back timing light refuses to show mechanical advance!!
Stripped the dizzy and found no issues.

I'm thinking either modules (tried 2) or timing light. Maybe the increased voltage is too much for the inductive pick-up?
Could there be a cross-firing issue'?

Will have to stew on this some more, but the wiring conundrum is a touch baffling too, considering I have a unit and dizzy with the original wire connections which I copied.

Maybe the pick-up coil wires are reversed!! What was top is now bottom?

????????????
Allan Reeling

To cut short what could be a long list, my V8 has an accumulation of autojumble and ebay aquired HEI parts;

But in effect,
Orange to G ( 3/16" spade )
Purple to W ( 1/4" spade )

Fitted about 4-1/2 years ago, timing tweaked at Peter Burgess', has been running well since, but only about 14500 miles in that time.
J N Gibson

Allan
Interesting you had to swap the wiring over, maybe as you say the windings have been reversed-----the coloured pickups may be different to the allblacks, a bit strange, haven't struck that--but if it work round the other way, so be it--
--
When you say your timing light refuses to show advance, do you mean---

When you twist the knob ,the timing doesn't change
---light issue
or
When you increase engine revs the timing doesn't advance
---distributor issue

I'm into the process of wiring up a 3.9 inj. into a B for a friend at the moment, it was a real headache yesterday--The engine came from a car that had an engine bay fire and half the loom has melted and the car was driving at the time-so ignition was on-even when I do get it hooked up i'm sort of expecting there to be damaged electrical components so it's going to be a long day today--------------but we'll get there
William Revit

Is vacuum advance connected and if so where does it come from? Manifold advance unlike carb advance needs to be disconnected otherwise the vacuum advance reducing as the revs increases cancels out the increase in centrifugal advance. That caused me some puzzlement on an Austin 3-litre where the timing didn't seem to change until the owner said it had a mode to move the vacuum from carb to manifold.

Does twisting the rotor turn it against spring pressure, springing back when you release it? Depending on the curve it may be a two-stage process, and significant effort from fingers should be required otherwise it will go to max centrifugal advance at very low revs.

Does the timing change if you twist the distributor in the engine? If so but not when you rev the engine then it's almost certainly a distributor problem.

The only difference in this area between points and these systems is that these systems use an electronic pickup in place of points. After that the process is basically the same i.e. the pickup controls one or more transistors to switch the current in the coil on and off to generate the spark.

Where the HEI systems do vary is that they are variable dwell - low at low revs and high at high revs, which basically comes from the current pulse to the coil being the same length throughout the rev range. The advantage of this is that a very low resistance coil can be used to give a higher energy HT output without overheating - 0.8 ohms on the 45DM4 system used on late North American spec models for example. Fixed dwell systems such as points give a much longer current pulse at low revs than is needed to charge the coil, which generates wasted heat.

It's possible the variable dwell part of the circuitry of the amplifier (two?) is faulty, but it seems to be a huge coincidence that the increase in rpm is causing the electronics to exactly offset the mechanical advance.
paulh4

Willy,
It's all strange.
The dial back light works fine on my other V8. (magnecor leads) and will show "all-in" timing, but on the GT it remains on the idle timing. It's almost like it's not detecting the voltage pulses, hence my suspicion it's a module thing, compounded by the need to backwards wire them!!
The distributor............I re-built it and had it "spun" by a local firm to set a suitable advance curve. A fault with it was my first thought, hence why I stripped it..........advance weights perfectly free and both springs still in place, inductor star wheel properly located with correct air gap, drew a blank on that. I've re-built quite a few DM8's so I'm pretty confident it's not the issue.
I've got a Lucas AB14 box with it's, working, original GM module. Will wire it in and see. Also on a bit of a wing and a prayer swap No 1 lead for a Magnecor.
Best of luck with inferno survivor
Thans Mr Gibson, that just confirms what I found out and Willy suggested.

Another thought, maybe the issue is the pick-up coil in the dizzy!!!

By the way, the dizzy to module is also a screened lead.
Hey Ho!
Thanks for your suggestions.
Al

Allan Reeling

That's really weird
You'd think the mechanical advance, advancing movement would show on the pulley regardless of where the timing light was set
Really puzzling
My gut feeling is that it's something wrong with the springs/weights, but what--??
It 'might' be possible to get the top shaft of the dizzy sitting on the wrong side of the pins of the bottom half and have the springs pulling the wrong way but if you've had it set up on a dizzy machine, playing with curves,you'd have to believe it was ok
If you grab hold of the rotor button,and try and rotate it anticlock, does it move(spring loaded)and return back ok
But that doesn't explain why you can't move the timing flash around with the wheel on the t/light
weird

This thing I've been swearing at today has the two black wires coming out of the dist. as well, like yours---I don't want to dismantle the dist. at this stage to check which wire is which but, looking at the outside where the wires come out the side through the housing, the top wire goes to the blue wire on the harness connector--
William Revit

Update.
Wired in the AB14/GM unit...........no change!
Replaced the pick-up coil...............ditto!
Replaced leads............................ditto!
Replaced dizzy cap.....................ditto!

I have another 35 DM8, so that's next?? Later today when I've calmed down!!
It could be the springs are so weak they are allowing the weights to spin out at idle speed, but can't see it................. I've got some more springs to try if all else fails...........I have to say it does sound/feel advanced when cranking! That's beginning to sound logical, but could it explain the need to wire the module backwards? I might wire the bob weights just to see if my idle advance changes.Or just really retard it then see if it advances any at higher rpm!
If the pulley had lost it's rubber bond I wouldn't expect the idle timing to be so constant even at 3000rpm!!!

Allan Reeling

Forgive me, but the V8 distributor does rotate in the opposite direction to the 4-cylinder.

If the engine is running then the pick-up must be doing it's thing at approximately the right time.

Centrifugal advance causes the top half of the distributor shaft to move forwards of the bottom half, i.e. further clockwise, to advance the timing.

I can't see cap or leads making any causing this, or rotor.

What vacuum advance are you using?

Does turning the distributor body advance/retard the timing?
paulh4

Ah yes, confused myself there clockwise it is, thanks Paul--you're forgiven--i had a dumb moment there

So Allan--
If you grab hold of the rotor button,and try and rotate it 'clockwise', does it move(spring loaded)and return back ok or does it feel sloppy or seized compared to your other dist.
and
did they give you a readout of the advance curve that's supposedly in it now, if it's been spun up and checked you'd think it'd be ok but if it feels advanced while cranking maybe as you say, weak springs or 2 slotted springs if it feels sloppy,
Strange again though if it's been checked earlier
William Revit

Paul I re-interpreted Willy's cw acw, water goes down the plug hole the wrong way down there.
Vacuum advance is not an issue when timing!
The cap and the leads where just something to do with no expectation of a change.
I've stripped the dizzy again and maybe the springs are a bit feeble, but non of my selection of springs are the right length. So one of the other distributors is going in!!
No read out!!
Allan Reeling

Vacuum advance is definitely an issue when timing - if it comes from the manifold. In that position it applies near maximum advance at idle, which is why it has to be disconnected. As the throttle is opened the vacuum decreases, hence the amount of vacuum advance. But at the same time the centrifugal mechanism is advancing the timing so the one is tending to cancel out the other, and depending on the particular characteristics can cause the timing to vary very little at all when revving the engine.

When vacuum comes from the carb there is no advance at idle, so dynamic timing can be done without disconnecting. However if the rpm quoted for dynamic is higher than idle vacuum advance can start to come in and confuse the issue again, this time showing more advance than is coming purely from centrifugal. Which means it really needs to be disconnected in both cases so you know you are seeing purely centrifugal.
paulh4

Paul, Vacuum is disconnected and plugged to time at idle and when checking maximum mechanical advance!

Anyway, alternate distributor fitted (much beefier advance weight springs) and the inability to see mechanical advance has ceased. Now seeing 30 degrees at 3500rpm. Whether it gets there with correct progression, as yet I don't know. But pretty sure now the springs in the other DM8 were too flimsy by half.
Maybe a question for the curve setters?
Allan Reeling

You probably know all this - and I'm digging deep in memory - but ...

Those of us without the luxury of an oscilloscope might be able to identify which lead is +ve by a simple test on the bench.
Connect a basic moving coil meter to the leads, spin the distributor by hand ( in the correct direction of course ! ) and you should see which way the needle kicks,
the sensor output is +ve first then a lower -ve peak. I don't remember if I ever did that check - and I don't propose to take my distributor out to try it.

Yes, the leads from the distributor to the module must be screened, screen grounded at only one end - usually convenient to use one of the module mounting screws.
Some part numbers PRC 4503, DRC 5799, but I doubt anybody has one, easy enough to make.

I'm pretty sure that suppression HT leads are required for some of the electronics in the module to function properly;
I'm using a BOSCH set for some Range Rover - Bosch part no forgotten I'm afraid.
I had to shorten a couple of them to fit the BV8 neatly.

That reminds me; there were several advance curves for the DM8. I think differences between the Carburettor and EFI engines were significant,
but for each fuelling type the variations weren't great. They were possibly just to pass slightly different emission tests without much overall performance difference.
But, don't quote me as any sort of authority on that.

Jim

( I must see if I can change my user name to something less formal )



J N Gibson

Jim,
The reluctor unit in the dissy puts out an AC current. Is that as per your description?
The beast is now running and behaving, but a test drive is to come.
I can attempt an approximation of an advance curve using my dial back light, my meter set to rpm..............and a suitable scribe!!



Allan Reeling

Well, the current ( or Voltage to be pedantic ) does alternate but it is not quite a symmetrical waveform like yout image. Apparantly it is deliberately distorted by the module.

My electronics knowledge is very rusty, but ..

If this link works then there is a good description of the HEI - and other - systems here ...

http://www.jagweb.com/aj6eng/ignition.php

For detailed info on the internals of the module search for articles by a Dr Holden, I can't find my copies at present.

For most of us we probably only need to know that once installed it works very well, go drive.
J N Gibson

Jim,
I have Dr Holden's "thesis" on customising reluctor style distributor and CEI ignition, permanently installed on my computer.
I have found it really interesting but can only extract and partially digest some of it. That's where the waveform graphic came from by the way!
I must have read it a dozen times, picking up a bit more every time, but the schematics of the circuits are completely over my head, I do now know what a Darlington is and what the zener diode does..............that's new knowledge, and ~I do understand the waveform distortion and it's retarding effect plus the clever control of the, low resistance coil imparted by the module, and that the reluctor is more than just an on/off switch.
It's just fascinating and impressive!
I like knowing, if you hadn't already guessed!!!!
Allan Reeling

I can't help thinking this is getting way too involved.

If the engine runs smoothly - as it seems to do with the wires connected a certain way round, there can't be much wrong with the trigger, the module, the coil, or anything else in the HT section.

Centrifugal timing advance comes from the upper half of the distributor shaft advancing relative to the lower half and hence engine rotation as revs increase, to trigger the spark earlier relative to a given point in engine rotation i.e. TDC.

If there is no centrifugal advance showing on a timing light then on the face of it that shaft movement is not occurring. Other than that it would seem that somehow the timing light is delaying its flash when the HT pulses come in quicker, or the module is somehow delaying HT pulses when the trigger signal comes in quicker. It should just be a matter of determining which of those it is by substitution.

Waveform distortion and its retarding effect is mentioned, but from the waveforms given not only does the amplitude increase with rpm but so does the slope, so the amplitude reaches a certain threshold sooner, which should cause the timing to be advanced. That's on Allan's waveform, but the waveform in Jim's link shows a very similar slope turning the coil off, to generate the spark, at both low speed and high speed.

But as I say to me that is getting bogged down in an area that you have no control over other than by substitution.
paulh4

I think Allan probably thinks similar to me Paul
Just getting it going isn't enough
i have to know exactly how things operate and from what Allan has said he won't be happy till he knows the system inside out
I believe the height of the peak in the wave has zero to do with timing the firing
All the switching is done at the 1v point of the slope------coil on, on the way up through 1v and coil off on the way back down through 1v
This 1v point is recognised by the module which creates a square wave signal (on-off)to operate the coil

Your dizzy guy sounds like he has some explaining to do, he couldn't have run it up to check
Although, he might have done a full race curve that would flick up to 20 or so on startup then stay there till 2500 or so then increase to full advance by 3000-3500----------but that'd be useless for the road

Allan
A zenner diode caps voltage so would be in the module to limit the signal for the coil to a reliable constant instead of increasing with revs
as the dist, signal does

One little issue I have is dial up timing lights
I had to send my machine away twice before it read correctly--sort of kills any trust in it, but i've checked it a few times since and it seems to be behaving ok now
I reckon if you had ten of them they'd all read different, i'd check it, just mark say 30 deg on your pulley and see what the light says--It'd be interesting to know the result
William Revit

Will.
On the early AB14 boxes the GM module output transistor (Darlington) was a bit iffy if the voltage on it's base got anywhere near 400v. The separate Lucas 350v zener "clamped the max voltage to 350v.
I gather the later modules were safe at 500v and that zener wasn't necessary
The picture shows the AB14 used on a Rover 3.5 Vitesse.
The Lucas Zener is top right.
The white plastic "Y" is a double resister which triggered the Injection and the tachometer.
The module is a pretty standard GM HEI.


Allan Reeling

All good
cheers
willy
William Revit

Of course, Willy, I'm as interested in the intellectual exercise of finding out how things work as the next person, it can make the difference between fixing a problem and not fixing it. But in this case where the engine is reported us running well enough but the timing doesn't advance when it is revved, I don't see how it is helping. All the electronic ignition is doing is sensing the rotational position of the distributor shaft and generating an HT pulse, which is exactly the same as points. It has no knowledge of when the last pulse was generated, or when the next one will be.

I'd be taking a step back and thinking I'm missing something simple.
paulh4

You're right Paul, up to a point. But if you think the discussion is tedious or unnecessary, just ignore it.
As you intimate, knowing how something works, is most of the way to knowing why it doesn't, sometimes!!
As posted previously, the problem was, the supposed re-curved, distributor.
But now I have a good handle on how this, new to me, electronic system functions which has provided interesting discussions and reading and welcome diversion from Covid issues!!
Thanks to all who contributed, including Dr Holden!!!
Allan Reeling

My mistake - didn't spot the centrifugal advance resolution a couple of days ago.

No complaint of tedious or unnecessary from me, as I said I'm as interested in what makes things tick as the next person.
paulh4

At the risk of boring Paul and putting him off his Curry, I have some final comments.
1. I'm an idiot!!!
What I failed to comprehend was the degree (pun intended) of timing change imparted by the HEI module. I'd accurately set the timing before installing the module, and the fact that it wouldn't run got me on the "over-thinking" route.
2. With it wired the wrong way, once you get an idle timing set, the car will run. As I found out. But what it won't do is accelerate past about 1800rpm with any enthusiasm. because there was no mechanical advance being "allowed" as my timing light indicated!
3. The reluctor and module will switch, but for reasons unknown to me, (too electronic for my understanding), will not do mechanical advance.
Once wired as Will and Jim described, i.e. purple or black top wire to 1/4" spade and orange to 3/16". Quite a bit of re-timing is necessary, then it's incredible!!
So there you have it, a toUch wiser...............as Homer says.............Doh!!!
As I read somewhere once, "we don't learn until we make mistakes"!

Unfortunately that doesn't seem to apply to politicians! Discussing that one would be even more frustrating!
Allan Reeling

'Conventional' points replacement modules seem to move the trigger point or 'phasing', hence an adjustment of the distributor to get back to the original timing. In one case it was so far out it was firing between two cap contacts so hit and miss as to which plug fired if either.
paulh4

Allan
Thanks for getting back with that--
Interesting feature that it won't advance with the wiring reversed--must remember that--thankyou

Something that might interest you-
I had a Mustang in that wouldn't perform and on checking, the spark wasn't advancing, I tried for ages to work it out and had almost given up until a guy came in just at the right time with his F truck with the same issue and said it only went off song since he fitted a remote starter button because his ign. switch had died---he had fitted a toggle for ign. and a press button wired direct to the starter
I fitted up a new ign. switch for him and it went ok all of a sudden
Got onto my mate at Rousch and found that On the mid to late 80's F250 Ford trucks and Mustangs, the ecu gets a crank signal at startup to activate parts of the ecu and if the vehicle is push started then ign. advance is deactivated----WHY--only Henry Ford would know-
Big flashback moment 'That' Mustang has a push button start, Rang the guy up and yep, he'd fitted it to make it look a bit blingy and wired it direct to the starter---I rewired it using the original wiring on the back of the ign. switch and removed his effort and away it went, spark advance to die for
Lesson learnt by this little black duck
William Revit

This thread was discussed between 20/08/2020 and 10/09/2020

MG MGB Technical index

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