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 Midget and Sprite Technical - Reluctant starting

I have held off from consulting you folks for about a year, but I am not getting closer to solving this. Firstly the engine spec:

998cc, Kent 276 cam, big valve Metro head, HIF44 on MG Metro manifold, K&N cone filter with ram pipe, A+ distributor with Powerspark kit, Maniflow medium bore LCB, Lucas fuel pump with Burlen electronic kit. Silicone HT leads, Lucas sports coil - I have tested this on a meter and it is within spec. New distributor cap. The plugs are NGK BPR6ES, changed in the last 1000 miles.

When cold the engine always fires once but doesn't catch. I can hear the starter gear spinning down, as if the engine has kicked back. After running the starter for 2-3 seconds it then starts normally. Occasionally it won't catch until the third attempt. When warm it always starts instantly. This isn't a major problem, but is quite annoying and it wasn't always like this. I have run most of this set-up for about 12 years and this issue is relatively new.

I recently replaced the failed pertronix ignition with the Powerspark kit, and set the timing at idle with a timing light. This turned out to be about 20 BTDC at fast idle, about 1100 rpm. I was not getting any detonation, but a plug reading suggested too much timing, so I backed it off to 7 BTDC, which is where I had static timing with the Pertronix (it's impossible to set timing statically with the Powerspark). This change made no difference to starting - if anything it's worse as it took 3 tries this morning.

Previously the engine didn't start within the first revolution, it turned over a couple of times and then ran fine. Now it coughs instantly, won't catch, and then needs a few seconds churning before starting. I can hear the pump ticking and it clearly is not filling up an empty float chamber, as it only ticks a few times when I switch on. This behaviour predates the change to the Powerspark - it did just the same on the Pertronix. I had the head off recently (failed gasket, head machined) and there was no serious carbon build-up. I would love to fix this.

Les
L B Rose

The very little I think I know on a postage stamp.

I might be wrong but here goes - having the engine set best for performance running might not be best for start and idle so the starting system needs to be on top of its game. Check the obvious like starter motor and its fit and connections, electric leads and earths from battery and body. Solenoid and connections. Plug gaps. Carb/choke settings. Damper oil.

Adjusting the timing a little might make starting easier.
Nigel Atkins

Mechanical or electric fuel pump?
If the former,then my guess is that when cold - i.e. its been standing a while, the fuel level has dropped a little in the carb. Still enough fuel to give a kick, but not then run properly until you have churned the engine over a bit with the starter, all just to operate the fuel pump and bring things back up to level again.

When its warmed up, implying it has run recently, then the carbs are already fully charged and raring to go.

If this is new, then maybe a slight fuel leak from the carbs?
Similar situation can occur with an electric fuel pump, only now its not dependant on you chruning the engine over, but is dependant on a slight pause whilst the electric thing does its bit. In this case then maybe try turning the ignition on, counting to 10 and then activating the starter?

Or, it could be something else.
GuyW

Thanks chaps

Fuel pump is Lucas electric. I've tried waiting until it stops ticking, but it only does a few ticks anyway. I've checked everything that Nigel suggests and more. There is no solenoid, just a switch on Mark 1s. Just been out for a run, and with this ignition timing the engine is much less responsive - and starting no better. More fiddling required.

Les
L B Rose

Well, with an electric pump and you waiting until it stops ticking - that debunks my theorising!
I may have got that last suggestion right though!
GuyW

I forgot about the starter but remember your Powerspark kit work so perhaps try a wider spark plug gap to take advantage of the electronic ignition.

Is the starter motor in top condition and not getting tired or in need of a bit of attention?

ETA: Guy, perhaps in the singular. :)
Nigel Atkins

Has it always behaved this way Les?

Some might say that an HIF44 is a litle on the large size for a 998cc engine. Its what I have on my 1275 (+60 rebore). I am no expert at all, but I wonder if the larger throat on the carb may not be producing such a fast air stream when its just on the starter and the fuel is not atomising properly from the jet perhaps? Then after a few spins the squish and squeeze effect in the cylinders it raising the temperature enough for the fuel to atomise/ vapourise.
GuyW

If it starts from warm and runs OK then I'd suspect the choke mechanism. What arrangement does the HIF have and how do you use it?
AdrianR

I was about to go on to choke when the computer was taken over from me.

Have you experimented with pulling the choke knob out a bit more or probably a bit less.

I find with my car I need to remember to start the car with the choke knob pushed in a bit then need to pull it out a bit more to hold fire.

Somewhere along the way something changed on the car and/or I lost the knack of instinctive cold starts and I have to think about what I'm doing.

Previously for me when the starter sometimes missed it was simply the starter at fault. Sounded the same as you describe, "can hear the starter gear spinning down, as if the engine has kicked back" but we might be talking about two different sounds.
Nigel Atkins

No Guy it was not always like this. Just the last year or so.

I stripped the starter motor a couple of years ago. Bearings good, Bendix cleaned (and not oiled), new brushes. Spins like a trooper.

I've tried different choke settings - no difference.

The HIF44 is what David Vizard recommended in his book for a a road 998 with hot cam and big valve head. It has worked OK for 10 years or so. I recently renewed the throttle spindle and seals, the needle, and the piston spring. Using the correct SU damper oil.

One other point that might be relevant. I still can't get a stable idle. This was all over the place a while back and I found that the vacuum advance can was leaking. Renewing that seemed to improve the situation. I have also ditched the 1098-style PCV valve and plumbed the breathers into the carb upstream of the throttle. But the idle still goes up by about 300 rpm when hot. Sometimes if I switch off for a second or so and restart it goes back to 900 which is where I set it. I wondered if the ram pipe was causing this but it's just the same without it. I like the ram pipe because it pulls up the torque a bit. Here's a pic of it.

Les



L B Rose

Combination of erratic idle, going fast plus reluctance to catch does sound like it's drawing air in where it shouldn't doesn't it?

Looks like you have the correct breather oil filler cap. How about the glands on the choke spindle, did you renew those with the others?
GuyW

The carb has had a full set of seals. I wondered if the choke spindle seals were softening with heat, but all I can do is renew them.

Another odd thing. Last week I had just loaded up the car with purchases at B&Q and it would not start. There was a spark, but all it would do is cough. So I took the carb off and stripped out the needle valve (in the B&Q car park!). Thought there might be dirt in it as there was fuel getting to the carb but none to the engine. However the float chamber was full, and the needle valve was clean. The pick-up pipe to the jet was also clear. Put it all back together and she started. Ran very rough below 2000 rpm, but by the time I got home (10 mins) it had all cleared and it has not done that since. But I think I will have another look at the feed to the jet. I do have an inline fuel filter and there is never any dirt in it.

As it was a warm day I wondered about fuel vaporisation, but I've had that before (in the Piazzale Roma, Venice, with the H1 carbs!) and that was because the mechanical pump would not suck. Different situation with the electric pump at the back, and anyway the fuel was pumping fine.

Les
L B Rose

Probably not anything to do with your problem but the other week I took apart my dashpots and cleaned and lubricated them and cleaned the carb bodies (engine oil there perhaps from sealing the dipstick too well(?)) and cleaned the jet needles and this seemed to improve performance.

A point I missed and might(?) be of relevance here was that in the SU book it has that dashpot and body are both marked for putting the dashpot back in the correct orientation on the body (marked as 5 below, throttle discs also show as having an 'x').

I've never looked for this mark or noticed it before, perhaps it was only on the carbs for and of a certain period.

I don't know but perhaps the fine clearance between piston and dashpot might be affected if the dashpot is put on the other way round(?), cheap and easy to try so doesn't matter if it doesn't work.



Nigel Atkins

Nigel, l don't know if the HIF 44 carbs are all the same, but on mine the dashpot will only go on one way.

Les, the O ring seals inside on the choke mechanism aren't easy to find, and l am not sure they are always supplied with the rebuild kits, which is why l was checking.

Other thing - did you set the float level as per book?
GuyW

Fair enough Guy.

I know mine can go either way because I've thought about this when putting them back on - but as I don't think of this when I take them off I don't mark them so wouldn't know which way anyway which is why I don't think I've ever noticed any factory marks on mine.
Nigel Atkins

SU on their site have for the HIF disassembly -
"Mark the bottom cover-plate and body to ensure correct reassembly (9), unscrew the retaining screws and remove the cover complete with sealing ring."

Was this done?

I stopped reading after, that too small print for me and too many pages.

http://sucarb.co.uk/technical-hif-type-carburetter-dismantling
Nigel Atkins

Nigel, that's the cover plate on the base that they are referring to, not the dashpot cover. But point taken, certain assemblies do need to be matched up!
GuyW

I didn't read on to see what difference it made. A rushed post but matched assembly and things being different after to before was the point as well as possible small mistakes leading to small variances that can upset the smooth running.

Once, in a rush to fit the new stub stacks inside the air filters (as the rain ran down my back) I put one of the gaskets back on upside down without noticing that it blanked off the air hole. It was only after leaving the A14 that I noticed the car wasn't running as it should.
Nigel Atkins

I'll have another look at the choke seals but I'm sure I replaced them. Also worth a look at the float level. A job for tomorrow, which is 59 years to the day since the car came out of the factory.

Les
L B Rose

Les, apologies if I keep firing off random thoughts on this. But another thing which may be worth checking is that your inlet manifold is properly sealing at the head gasket. If the thickness differs between the exhaust manifold and the inlet manifold flanges, then the bolts may not be nippng the inlet up tight enough to seal. They can leak on the underside where its difficult to check under the manifolds.
GuyW

Oh, I've been there Guy and got the T-shirt. I had Maniflow weld spacers onto the LCB to ensure the same thickness as the inlet. But don't apologise, all comments welcome!

Les
L B Rose

OK, a long shot now. I see you have crankcase extraction from both the timing chain case AND a second one - from the tappet covers? They are both quite large diameter pipes. I wonder that maybe you have overdone this and it is drawing air from the crankcase too readily and this is weakening the mixture too much on start up?

You coud temporarily disconnect the secondary one and close off the pipe and the joining Y piece with corks or bolts secured in the end of the pipes with jubilee clips. Just as a test to see what effect this has.
GuyW

Les
Just out of interest, do you have the vac. advance from the dizzy connected directly to the inlet manifold--or to the carb
If it's direct to the manifold it 'could' be the cause as the base idle throttle position is less using this method and can cause unreliable idle speed and stumbly starting
willy
William Revit

It's on the carb. I have tried both, but worth trying again. Thanks William.

Les
L B Rose

I had somewhat similar issue with my 79-B. I tightened the nuts that hold the intake/exhasut manifolds to the head which solved a vacuum leak. Also check the tightness of the carburetor to the intake as well. After that, the car ran fine.

Cheers

Gary
79 MGB
gary hansen

Les,

You mention silicon HT leads and resistor plugs. Are the HT leads also resistor? Some people say you should have one or the other but not both. I have not experimented myself.

Charley

C R Huff

Les,

My experience with silicon resistor leads and resistor caps, screw on at least, was they didn't last more than a couple of years before they started to misfire. I mentioned this to someone after awhile and they said the 2 weren't compatible as the end where the screw in cap fitted burns out causing the problem. I changed to copper leads and resistor caps and no further problems.
David Billington

Les
I wasn't suggesting to put the vac. hose onto the manifold---On the progression port on the carb is best

just checking that you didn't have it on the manifold
I'd be leaving it on the carb progression port, which should have no vac. at idle speed

willy
William Revit

Apparently HIFs also have a fuel temperature compensation device. I wonder what that would do if it played up?


AdrianR

No I don't have resistor caps, just the silicone leads. They are about 5 years old. But the problem is not misfiring.

Ah, misunderstood you William. Not that it made any difference.

The fuel temp compensation device is just a bimetallic strip, and would only malfunction if broken. It isn't.

Les
L B Rose

"and would only malfunction if broken" - is that 100% correct I wonder (not that I know) could its fixing become loose, metal fatigue, alter with age and use(?).
Nigel Atkins

Hm...you have got me thinking Nigel. Worth another look.

Les
L B Rose

Les
Just noticed while reading back through this that you have silicone plug leads
Are these resistor cored or solid wire--Why I ask is that you also have resistor spark plugs
'If' they are resistor leads, the combination of resistor leads and resistor leads might be a bit much for the coil to feed
If they are resistor leads, maybe a set of BP6ES (non resistor) plugs might be worth a try
willy
William Revit

I think it could be your advance curve. It sounds a bit over advanced for the starter but not ideal for idle either. Might be you need more initial advance so different springs/bodyweights, or you have worn springs so the initial advance is not properly controlled. That plus a marginal battery which is OK once the engine is warm but is struggling first thing. You might try pulling the starter with the ignition off for 3 or 4 churns - that will get over the initial drag and then switch on the ignition. That was the trick I used on a reluctant to start FJ engine. A bigger battery plus a refreshed advance curve may do it. However that super long inlet tract can't be easy on the manifold gaskets. Might be worth a squirt of WD40 to check there are no air leaks on the induction side.
f pollock

The ignition leads are resistor type. Maybe you are right Willy about a change of plugs.

Yes I also wonder about the advance curve. Fiddling with that is a bit beyond my skill set. But as I say this is quite a new problem and the engine ran fine for 10 years with this dissy. I regularly check it for wear and can't detect any.

I am not worried about the battery which is almost new - 48AH and 430A. It spins the engine over lustily.

I have had manifold gasket leaks and know what to look for - happy there are none right now. The ram pipe and filter are very light compared to the carb.

Les
L B Rose

I can't remember if this has been discussed before but if your changing from non-resistor type plugs then what about trying a change of heat range.
Nigel Atkins

Les, Our 1380 has similar starting characteristics as your 998 from cold.

Spins over for ~2 seconds, and then fires up very reliably, no matter what time of year (it is in a dry garage that never freezes). Even if left out overnight still takes the same ~2 seconds of turning over before it starts.

Often, if started the following day, it will either fire immediately and throw the starter bendix out, or sometimes catch first time and run normally.

This has been the case since first buying the 1380 engine 20 years ago, and during that time it has been through a number iterations of cam and head, but always the same HIF44 carb, inlet manifold and large bore LCB exhaust.

For ignition, it started out with points, then quite quickly an Aldon/Pertronix Ignitor and now a 3D mapped Nodiz system using a 4-cyl coil pack - the starting characteristics are exactly the same with all three.

Because it is so consistent, you could count the turns the engine makes, and put good money on it starting on exactly the same number each time!

The only thing I have thought of is the brake servo - could it be that a 'full cold start' draws in more air than normal into the inlet manifold, and so is too weak to start, until the servo vacuum has been created after ~2 seconds?

A 'next day' cold start could have some of the servo vacuum still in place?

Just tried an quick experiment - the car was used yesterday, so before starting it a few minutes ago 'just to see', I pumped the brake pedal a number of times to remove any residual vacuum. Sure enough ~2 seconds of turning, and fired up as usual.

If you don't have a servo, all this 'may be/could be' stuff is totally irrelevant!

Richard
Richard Wale

Willy, I think you may be right. I just fitted a new set of NGK BP6ES and this morning it started instantly. It has not done that for ages. I can't recall exactly when this problem arose, and I have had the resistor leads for longer than that. Maybe the coil has dropped a bit of voltage with age but is still within normal limits. Well today's start is just once but we'll see what happens tomorrow and thereafter.

Les
L B Rose

This morning it again started on the button, so many thanks to Willy. Winter may be a different matter but a long time to wait.

Les
L B Rose

Well done for (hopefully) resolving Les' problem starting. It's been a very interesting thread to follow and I've learned a lot from everyone's helpful and well reasoned comments.
I do like this BBS - always interesting and well informed.
Peter Blockley

All good then Les--
To be fair it was Charley's comment that reminded me of the problems of too much resistance being an issue when using both resistor plugs and resistor leads together
Cheers
willy
William Revit

P.S.

I have silicone leads and NGK BP7ES (non-resistor) plugs, so not much hope for an improvement then??!!

Richard
Richard Wale

Willy,

Thanks for the credit. Quite gentlemanly of you!

Richard,

Even though you have non-resistor plugs, you might check the resistance of your plug leads to see if it is too high. Sorry, I don't remember what the value should be. I think maybe less than 1000 ohms per wire and better around 500. But, keep in mind that you should see if someone else has more confidence in a number.

Charley
C R Huff

Richard
I've been thinking about your problem
Would it be worth blocking the manifold port off for a try, just to confirm your booster theory
If it does turn out to be the problem you could then try a manifold fitting with a much smaller hole, like 1/8" or even smaller
As you know, the booster has enough in reserve for a few pumps on the brake so the hole in the manifold fitting only has to be big enough to keep the reserve topped up
willy
William Revit

Charley,

Thanks, these are the 3rd set (2nd magnecor) in 20 years, and the starting pattern has been exactly the same for the whole time!

The current leads were bought 2 years ago, and are Magnecor 80 Electrosport silicone leads - the spec is 98 ohms/cm, so ~3k ohms for the lead length of ~30cm (12").


Willy,

Yes, I will try blocking the servo port on the manifold. Somehow though, ~8 turns on the starter, without fail (so far) for 20 years, is rather comforting!

Richard
Richard Wale

This thread was discussed between 27/04/2018 and 15/05/2018

MG Midget and Sprite Technical index

This thread is from the archive. The Live MG Midget and Sprite Technical BBS is active now.