Welcome to our resource for MG Car Information.



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 - Poor idling with 276 cam

Now that I have a tacho I can believe, I am trying to get a better idle. To recap, the engine is 998cc with big valve Metro head and HIF44. I know, that's a big carb for a 998, but I slavishly followed Vizard's recipe as far as I could. I find that with the engine hot I can get the idle down to about 1100, and if I lower it more, after a few seconds the revs drop right off and it gets increasingly lumpy until the engine stops. That's after setting the mixture at a fast idle. When the revs drop off I can hear the carb piston chattering, which is odd. The damper is full of oil of course, and I have the correct spring.

I had a long ram pipe on the carb for a while and the idle is worse with that - but I does give a bit more mid-range torque.

I previously had a Piper 270 cam, and it idled OK then. That cam was ruined by over-strong valve springs (fitted by a well known tuner - no names no pack drill). I could not get another Piper 270 for decent money so settled for the Kent 276. Maybe it's a cam too far? Long shot, but has anyone else relevant experience?
L B Rose

I will chuck a couple of easy starter questions into the ring before the real brains come along...

Air leaks? Is the carb in good order, good, non-leaking throttle spindle?

Is the dizzy and ignition system in good order and set correctly?

My 1275 uses a 276 cam and it idles quite happily and steady at 700rpm. CR is 10.5:1 which is quite high, if that makes a difference as well?

HIF 's are prone to cracked floats which make the carb flood up at idle------a maybe
Adjusting the idle mixture well above the normal idle speed is a boo boo, that's a job for different needles if the mixture is out there, for a starting point your needle shoulder 'has' to be level with the bottom face of the piston as normal and adjust the jet tube so that it's .070" down from the bridge of the carb body---it should idle fine at this but might need a fine adjustment running to suit your spec/engine/air cleaner etc
also are all the breather holes clear on the face of the carb where the aircleaner mounts-important that they are open
also although you have oil in your damper , is it working properly ,is there a resistance when you lift the piston there could be a problem with the little damper piston not operating properly-
BUT--as usual , basics first as Malcolm has suggested , ign timing , air leaks, plugs etc have to be right before adjusting carburettors
and as far as a 44 on 1000cc goes it'll be fine-not too big at all-
William Revit

Thanks chaps. I can't find any air leaks. The carb has a new spindle and there is no bearing play. I service it regularly and there is nothing wrong that I can see. The damper has resistance. The needle and jet are correctly set. All carb breather holes clear. K&N cone filter which was recently cleaned and oiled with correct materials. Ignition is via Powerspark and set with timing light. Distributor is from 998 Metro, bearings perfect and advance mechanism checked and lubricated. Plugs are almost new Champion N7YC. Compressions are equal and 170psi hot. CR is 10.5:1.

I am wondering about the PCV valve. I tested it yesterday and at about 1200 rpm taking the filler cap off caused a very slight rise in rpm, at anything below 1200 the revs collapse. The beer mat test showed barely detectable suction. The valve has a new plunger and diaphragm, and works when sucked by mouth (yuk). The filler cap gauze has been taken out and cleaned.

I just took a plug out. Black and oily, but that was after a long session at idle rather than WOT. I think the next step is to remove the PCV valve, get the engine hot, and try again.

L B Rose

William Revit

Agree with Willie. Could be high fuel level. Or the HIF44 has that odd choke arrangement which AFAIR relies on a couple of O rings which are prone to wearing out.

I have rebuilt this carb countless times, always with new seals, but will have another go when my cracked fingertips are better. Typing is agony!

L B Rose

The HIF also has the bi metal blade holding the main jet which adjusts the mixture slightly with changes in temperature in the carb. It’s important to make sure you have a decent heat shield, especially with an LCB where the centre branch can run very close to the float chamber.

It’s probably not that if you had it running ok with the 270. I’ve got a 1275 with a 286 cam and HIF44 and there’s definitely no smooth idle with that!

Personally I’d have thought the 44 is a bit big for a 1000 unless running at high revs all the time, ie racing but I won’t try and challenge Willy or Vizzard!
John Payne

no need to type a reply, at the risk of exasperating you but do bearing in mind Colin's recent bitsas carbs.

Does your HIF44 carb have all the correct bits for it and its use to the car as it is - the damper spring colour/weight, correct damper, dustproof or not dashpot, fixed or biased needle, ect.. Do the part/cast numbers of each part match to what should be on the carb. I know I was amazed at the various HS6 carbs applications and variances when looking for a neighbour and it was only by checking the part numbers of each bit on each carb we could be sure the dampers were correct as it wasn't certain what the donor vehicle was.

Colin's missmatched carb bits and set was sorted a bit more by checking cast/part numbers.

Nigel Atkins

If the float's ok maybe one of the o rings on the choke have died or are wrong, some have one some have two on the choke body, if the furthest in one is damaged/missing /incorrect it can cause a rich mixture -dark plugs
William Revit

Clean the carb out thoroughly - especially all jets, orifices and all the wee holes in all the jet holders, carb body, pilot jets etc etc. Must admit, I do use some thin wire to poke them clean and plenty of carb cleaner.

Yes check the float height - as per the manual. It is crucial to get this right otherwise it will flood. Check float needle is seating correctly and not leaking petrol when float is up.

IIRC HIF 44 has some sort of breather hole around the inlet flange somewhere. Check the gasket is not blocking it.

Plugs should not be particularly black even if mainly running at idle. Check pilot jet settings, jet and orifices within the carb. Use 1 and 1/2 turns out and take it from there.


I have had the carb apart and all I could find was that the screws holding the cold start device were loose. They have loctite now. The official SU manual shows only one O-ring for the cold start. Existing one is fine, and I have fitted a new gasket. All other seals are almost new and look perfect. I would change them if I had any in stock. The carb came from an MG Metro (scrapyard). I have had it for nearly 20 years and it worked perfectly until after the cam change. I've checked everything that you guys have mentioned, except part numbers, and all major parts are as I bought it except for the throttle spindle. Well I found some while ago that the spring was wrong and I fitted the correct one.

The float is fine and I checked the height with the vernier. Needle valve was replaced recently and is also fine.

I just went for a run and tried setting it up again. I still can't get the idle below 1100 (the tacho is the new electronic one). I found that when the electric rad fan kicks in the revs drop a lot, and the engine goes very lumpy. I realise that the fan puts a lot of load on the system - about 10A - so the alternator will take more effort. But I'm surprised it has this effect.

The plugs are still black, so it's clearly too rich. I don't have a CO meter so I am limited to tuning by watching engine speed. The exhaust is perfectly clean. I will have to do repeated road tests as I weaken the mixture, and see if that helps. I have a weaker needle and will try that.

L B Rose


Dirty plugs themselves may cause lumpy idle. Root cause of dirty plugs is more difficult! Clean them before starting to fiddle again.

For sure check the mix - and idle mix is the screw behind and a bit below the fuel-in pipe. Clockwise for rich, anti-clock for lean. This screw is for adjusting the idle mix only. Once it is set correctly, the rest of the fuelling is determined by the needle profile and main jet size so be careful in changing needles.

Also check the throttle spindle. If it is worn/too much play it can allow air in at idle

Is choke OK? I assume it is not restricting air when seemingly off?


apart from my incompetence, lack of aptitude and bad attitude, I found with my carbs, that had been functioning fine before, that one issue with them was that I thought I'd sorted the issue already but hadn't (it was intermittent) and another issue that wasn't the carbs.

I'm sure you've been through the basics a few times like air leaks, connections, other related parts and components (ignition side, exhaust). But sometimes it pays to recheck with the full tick-list.

Spare plugs are so cheap I found it easier to buy a new set than yet again clean the not very old set and start with a known clean and working level.

I note you have Champion (colder) plugs, the only time I've ever had a problem plug has been a Champion, so I'd suggest trying a set of NGK for next new set.

As you got your carb off a scrap car I'd still check the various cast/part numbers on the carb as one bit has already proved to be wrong, the spring, and whilst all worked before with previous cam it doesn't now necessarily with the new cam. Did you try a blue 2 1/4 ounce instead of the red.

I've got a high powered cooling fan and it's rated at about 8.5a against the previous standard at about 7a but both seemed to pull the tickover down by the same amount when running at about 100-150rpm on my car.

I find with the very intermittent use over the last year that the idle isn't really settled until the car's had a good blow out bit of the drive, then the tickover seems highish and I'm sure will seem higher when it gets warmer in summer but without more frequent and regular runs it's difficult to tell.

Trying stringing together a series of more regular and frequent (test) runs before making any adjustments rather at each time to get things well running and settled. I wish I had the opportunities as mine is fine when I park up after the run but then it sits for weeks getting square tyres and losing the fitness it's just gained for another set of weeks. I seem to be constantly thwarted at excuses to use mine.
Nigel Atkins

Nice one Nigel, I'd missed the cold plugs
They are a bit on the cold side for a roady and being Champion as well, could be an issue, I don't think it's causing this rough idle on it's own but it could explain the dirty plugs
NGK BP6ES is the go-

Also- Les, if you havn't got a co meter
I realise you know your way around your car and don't want to tread on your toes, but the easiest way to find out if your idle mixture is too rich or lean is to get it up to temp, poke your finger in the front of the carb while it's idling and lift the piston a bit
If you slowly lift it up the revs should very very slightly increase at about 1/8" rise then as you lift it further they should drop off and possibly stall as you get up 1/2" or so
If it's too rich at idle as you lift the piston the idle will smoothen out as you slowly lift the piston and you will be able to lift the piston up probably 1/2" without effecting things much depending on how rich it is
If it's too lean depending on how lean, the revs won't increase as you lift the piston that 1/8" and will drop off with any further lifting, if it's way lean the idle will drop off with any piston lifting at all

I'd be getting rid of them plugs
William Revit

Thanks again chaps. I think it's better now. Plug colour is correct, and I managed to get it down to 800rpm albeit lumpy. Next time I go for a run it will be different of course! Throttle response and power are fine.

Yes I have a blue spring. I chose those plugs on a recommendation from another thread, can't remember where now. I had the NGK ones before.

The only fault I could find was the loose cold start screws. I take your point about checking the carb has all the right parts, but it ran perfectly well for many years after I rebuilt the car and this idling problem only arose after the cam change. Well after actually, so the two might not be related.

L B Rose

Well done Les.

If 800 remains too lumpy for you and all else is fine sometime later why not try 900.

The red spring was the standard but blue might be better for bigger carb as I remember it from the SU Tuning book.

Another check and clean of the electric ignition side might turn up something.

Nigel Atkins

I found a set of NGK BP6E and put those in (after cleaning). Starting better - the engine often coughed and started on second go. Maybe idling is a shade better. I got it to a steady 900, blipped the throttle and it settled to 1100. Yes I have checked the cable and and throttle stop. I think I'll leave it at that for a bit as this is getting boring.

L B Rose

when you feel like it (and I know how you feel as I've been farting about on my car today)as you've changed the plugs rating then you may also perhaps need to adjust your timing and carb to get the match.

I'd also have a god look at the rotor arm, dissy cap and HT lead set and clean or replace as required. I've found dissy caps and HT leads particularly can look OK when they're not, the leads might even meter OK but play up on the car.

A dark night and a water mist spray bottle used to sometimes give a good light display, I suppose you have to be a bit more careful now with electronic ignitions but if the leak is that bad it'd probably show up with just the dark.
Nigel Atkins

I did find I could lean out the mixture slightly after the plug change. I replaced the dizzy cap about 3 years ago. Rotor arm also new when I installed the Powerspark a short while before that. I haven't fiddled with timing - I'll see how it goes on the road for a bit.

I'm puzzled about this "dark night and a water mist spray bottle" idea. What is it supposed to show? Do you mean spark tracking?

L B Rose

Les, had a similar issue with my 79-B. Turned out to be a vacuum leak on the manifold mating with the cylinder head. Tightened up the manifold nuts and the problem corrected itself.


79 MGB
Gary Hansen

Gary, this has been such a persistent issue that I have had the manifold on and off scores of times, with numerous new gaskets. I'm totally sure it isn't a manifold leak. But thanks for the suggestion!

BTW I revisited the camshaft spec and the inlet period is actually 270º not 276, so not as hot a cam as I thought. 276 is just the name - which makes no sense!

L B Rose

Resurrecting this thread in the light of today's developments. Years ago I made a longish ram pipe for the carb, inspired by Vizard's book. See photo. I took it off last year to see if I could get the idling better. It was slightly, but still not quite right. Anyway I put it back the other day and tried the car today. It would not idle without the choke, so I warmed it up and tried to reset the carb. I found that I had to enrich the mixture a lot before it would idle at all. There followed an hour of fiddling with mixture and throttle stop, and I could not get it below 1200rpm. It would idle lumpily at that speed for about 10 seconds, and then suddenly drop right down and stall. Anyway I took it out for another run and did a plug test, which seemed to be spot on for colour and markings (maybe slightly advanced ignition). I have given up for today. I'm thinking of taking the carb across town to Burlen. I have rebuilt it countless times, but it's clearly not right. The throttle stop screws right up until I can feel the plate sticking, and still the idle is 1200. However this is much worse with the ram pipe - I have not changed anything else.

I like the ram pipe because it does pull up the torque a bit. I really should get a rolling road session but there's no point until I'm sure the carb is correct.

L B Rose

Les, sounds like a trip to Peter Burgess' rolling road may help resolve your issues with the ram pipe. He does a good job.
Jeremy MkIII

LB rose. Where -(if you are), are you connnecting the vacuum advance?

Are you aware that you need ported vacuum, and that unless you drill another vacuum take off on the HIF44, that you are using manifold vacuum at idle?

I can't see this mentioned in the thread discussion.

I had exactly your problem when I put an hif44 on my 1275. I couldn't get a slow even idle with the distributor vacuum connected unless I ran retarded ignition timing. At the time I didn't understand the significance of ported vs manifold vacuum.

See kieth Calver for full explanation.

Of course your tacho may be reading high, but piston chattering is often because the diametrical clearance of piston to suction chamber is high at the point of idle. What vacuum readings do you see?
f pollock

Good point with the vac advance--
Where have you got that hooked into-(pic)
William Revit

If the vacuum takeoff is in the none optimum place wouldn't disconnecting and plugging the takeoff improve the idle?
I've got a 286 in mine and a HIF44 from a late Maestro so my takeoff is in the none optimum place as well. Maybe that is one cause of my varying idle speed which I set to about 1000 rpm but varies a lot and can go up to 1250 rpm under certain circumstances.
MG Moneypit

I have a 276 with an HIF44. I am not sure what vehicle the carb initially came from but it has a vac adv port built in. I wouldn't say my tick over is particularly smooth but it isn't too lumpy to idle ok at around 850rpm

If the vac. advance is hooked to the manifold vac, depending on which vac pot is on the distributor the idle timing could be as much as 25 deg and unhooking and plugging the port should(will) reduce the idle speed significantly and will probably need the throttle stop adjusting up
This is the problem with having full vac onto the dist. at idle, the idle strength is weak and prone to vary caused by very little throttle opening and idle speed relying on manifold vac---As idle speed varies the vac level varies and the advance varies as a result causing unstable idle speeds
Ported vac. off the carb is the go with zero vac advance at idle allowing a slightly more open throttle plate and a stronger more stable idle
The other issue with it hooked direct to the manifold is that at idle you have full vac advance operating but as you drive off with a bit of throttle vac drops and timing drops with it making the car feel as if it bogs down a little as you move off, The reverse happens with ported vac. nothing at idle but vac advance operates as you move off giving a timing boost to make it all feel stronger
William Revit

Just getting back to this thread, and I appreciate the interesting discussion. I just realised that the relevant photo didn't upload, so here it is. You can't quite see it, but the vac advance is connected to the carb just downstream of the dashpot. What do you mean by ported vacuum?

Sadly I am a long way from Peter Burgess.


L B Rose

Slight hijack warning!

Les, I see you have a canister type chaincase cover on your 998cc engine. But where does the other breather pipe go/ come from?

I'd hazard a guess that it comes from an adapter plate bolted where the mechanical fuel pump should be.
MG Moneypit

Willy. My setup seems to have one of these for the vacuum to distributor although the one on mine is from the Maestro I took the carb off.

MG Moneypit

That version would give manifold vacuum then.

Yep--that's full manifold vac--I think that port was used originally to connect to the carb float chamber breather for some sort of emissions thing

Ported vac. is basically a port(connection)on the carb. dedicated to dist. vac advance that has no vac at idle but gets vac at slight/part throttle and very low or 0 vac at full throttle
I suspect that you have your hose onto the port that's for a crankcase breather. I'm not sure what sort of vac supply that port has-it may or may not have vac at idle-?
Had a hunt and found a little explanation writeup here-specially the last bit down below the pic

Don't jump in drilling just yet though ,we'll try and find some better plan where to drill if you go that way
William Revit

Yes that's the Keith Calver explanation about too much vacuum at idle, that I drew attention to earlier.

The easiest way to test it LB Rose, is remove the distributor vacuum connection, plug the carb take off, and see how the idle improves.

You can easily drive it without vac advance, but I guess the best solution is to drill the hif imo. I've been meaning to do that on mine, ( it's well described by Calver) but I put twins back, on and haven't found a pressing need.

"The solution is easy - remove the vac take-off from the manifold and fit it back to just in front of the butterfly where it should be. If you need to drill a port to accept the vac pipe take-off it needs to be sited 9/16" back from the carb to manifold mounting face, offset slightly to one side. I.e. NOT at the dead centre, or 'peak' of the butterfly bore. And usually set off away from the crankcase breather port. Usually the vac pipe take-offs fit a 1/8" diameter hole - but measure whatever you're going to use before drilling. You're looking for an interference/air tight fit."

Or get an mg metro distributor, as I'm pretty certain the vacuuum can on that, is designed to take manifold vacuum. It retards under vacuum as opposed to advances I believe.

Or put your twins back on and solve it that way.


Yep on a 45D distributor, the can works as a retard unit, with manifold vacuum being used.

I think, but may be wrong, later mgb's use this as well as mg metros, and possibly later minis too.

Cheaper to drill the hif than swap the dustributor.

This is where you have your distributor vacuum connection I assume LB Rose.

You can just see the small pipe to the rear of the dashpot and adjacent to the mounting flange.

It's effectively connected to the manifold.


The vacuum generated at the carb is actually very low. I measured it once with a gauge and its barely 2psi at 3500. And as far as I recall it doesn't vary much with revs when they are held steady, but does increase briefly when the throttle is snapped open as on acceleration. Its what you would expect as the engine attempts to gulp air. So the vac increases and the dizzy advances to aid the accelerating engine.

The photos show what happens with the carb breather port which is close to the dizzy vacuum port (as in Anam's photo)and I think will be much the same vacuum. You can see the gauge connected to my HIF breather port and the glove is demonstrating positive crankcase pressure as the breather pipe was blanked off. The gauge was showing 2psi when at a steady 3500 rpm. The second photo is with the breather hose reconnected (via illy catch tank, bottom left) also at 3500rpm and now the glove is collapsed as the carb suction pulls the crankcase pressure down.

I know this was the breather port, but the two are close together in the same part of the carb. Suction is much less there than at a manifold connection, and more critically manifold depression is at its greatest when the engine is spinning fast (gulping air) but the carb butterfly is closed (foot off throttle, on the over-run) That is why some dizzys are designed to retard at high vacuum pull, so as to reduce pinking when the mixture is weakened when the carb is shut down at speed.


And the reason for that low and almost constant vacuum, is because SU's are constant depression, variable throat carburettors. The piston rises and falls, keeping the depression 'constant', in the vicinity of the distributor and breather take off points. Simple but brilliant.

You have to love Guy's pictures of that rubber glove. Oh for an oil sucking thread. Lol.

Here's the gauge reading - observant will note its a diffferent gauge, but the same experiment. The other gauge was "borrowed" from my welding bottles and it wasn't really accurate enough for low pressures.

Second photo is just for you Anam as you seem to like alternative uses for rubber gloves. This was to stop residual oil dripping on me. I think it was as I undid big ends to remove the pistons for fitting new rings.


Very versatile tool that rubber glove. 😄. I guess you turned it inside out and put on again afterwards? 😉.

I assume the green is the optimum reading when cruising in the zone. I used to have a redex vacuum gauge. Remember those? I think it had red green and a yellow or orange zones. Each gave an indication of potential engine condition/problem. Or maybe it was only a green section on the redex gauge, and another maker that did the multi coloured gauge dial? I forget, it was back in the 1970's.

wandering off the issue a bit but--the blown up glove only proves that the crankcase pressure was greater than the ability of the breather port to suck it away
Guys experience shows exactly why the vac advance shouldn't be on that breather port and if drilling the carb it needs to be on the opposite side of the carb away from that port
With it drilled in the correct position, only just clear of the butterfly, -as the butterfly opens a tiddle and clears the drilled port the airflow in that are creates a greater vac. for the dist.
There's also a possible problem drilling the carb--
Some HIFs have the butterfly turning the other way and need the dist. port on the bottom (MGB)
Also the hole doesn't want to be large, about 2mm is it and then drill the outer part of the body to take the approx 1/8"(3mm) tube
William Revit

Willie, you misunderstand the experiment/ demonstration. The inflated glove was with the vacuum extraction pipe disconnected and blocked off whilst I measured the carb port vacuum with the guage. That shows the pressure created in the crankcase that forces oil against the effect of the rotation of the rear crank scroll plus how little vacuum is available at that port. The second is at exactly the same revs with the breather reconnected to the extraction port on the HIF44, showing how the system then copes with the crankcase pressure up to around 3500rpm on that engine (YMMV)

There was a discussion going on about PCV systems as against opening up as many breathers as possible to keep the internal pressure close to ambient. Different principles and different applications. I believe open breathers (with catch tanks) are favoured by racers as they spend most of their time at higher revs and the PCV systems run out of adequate suck at around 4000rpm

The vac for the dizzy isn't taken from the breather port. That carb is already fitted with the correct vac take off for the dizzy. You can see the black rubber connector, central and just downstream of the dashpot/ upstream of the butterfly.

It's also running an A+ electronic dizzy from an MG metro, so I will have to think about Anam's comments about this dizzy needing manifold vacuum. I am not convinced by that.

Got it-sorry Guy i jumped in boots and all there-
On breathers--on our old racer chev, because it was on full throttle more often than not, all the breathers from the crankcase and valve covers went to a sealed catch can up on the firewall then a vac hose went to a collector on the headers -it worked quite well after a bit of fiddling with pipe sizes into the collector--got the idea off a friends supercharged speedboat that has zero vac in the manifold
William Revit

Guy. From the tech archives.

Chris at Octarine Services.

"The manifold vacuum and 45D work as a vacuum retard system, so the advance responds to meet the same engine demands as the carb / 25D vac advance system but runs a higher advance at idle allowing a leaner mixture hence saving fuel and lowering emissions."

Is Chris at Octarine Services reading this?

I've read the same thing in a few places since. You have an mg metro electronic distributor I thought, so you must already have the best solution, the right vacuum (retard) for best possible idle.

Next best is 25D with drilled hif to get back to ported vacuum, or go back to original twins, or I think an HS6 still used ported vacuum for the distributor.

My point was that my HIF already had the dizzy vac take off fitted, ex factory, and in the position that Willie describes. So I have an MG Metro dizzy mated to a ported vacuum point. But it works well, pulls cleanly to 7000 rpm (though rarely used to that level) and has survived like this for 15+ years so I don't think there is anything wrong with it.

Guy, we may be talking at cross purpose, if that's the correct phrase.

If your distributor vacuum is connected in the place shown in this picture, -- the small vertical pipe behind the dashpot--, then you have manifold vacuum;--- unless you have connected to larger bore brass pipe on the side which is the crankcase breather, which as you say stays pretty constant as per the design/operation of constant depression carb's.

I thought all hif su's have manifold distributor vacuum take off, from the factory.

Did your hif come from the metro you got the distributor from too?

Edit:. Just looked at your picture Guy. That HIF looks exactly the same as mine. Mine came of an MG Metro too. It pulls manifold vacuum in that position.


That will be why it works as you describe then, and I haven't (yet) blown holes in my pistons!

I had assumed that the high vacuum on the manifold on overrun was because the butterfly closes, cutting off the air supply to the engine, hence the vacuum as the spinning engine is still sucking air. But if the 'barrier' is predominantly the piston dropping rather than the butterfly closing, then the vacuum take off upstream of the butterfly would indeed be at manifold pressure. But I suspect its more complex and dynamic than that as the butterfly will close slightly ahead of the piston dropping.

I'm kicking myself for not understanding ported vs manifold vacuum when I went to the breaker's to get the hif from the mg metro. The distributor was on the car too, I could have had it for a fiver. You made a good move getting one. It didn't occur to me that it would be better.

They come up on ebay sometimes for as little as 50 quid though, and even new for about 100 quid.

Not sure that I understood the relevance either, but I was collecting useful parts back in the day when you could wander around a yard and strip what you wanted. I also knew the owner, which helped. I paid £5 each for the carb and dizzy, and went back the next day for the MG Metro head and rockers for £15. Happy days!

Sorry Les, we really have wandered off on a tangent on this thread!

OK I’m confused !
Calver notes 9/16” back from the butterfly to manifold joint - is this not where the take off on Guys carb is positioned ?
And this to gain Ported vac ?

As the butterfly does not fully close at idle, the gap being equal above and below the throttle disc would not the air speed/ pressure be similar at both locations being so close to the butterfly centre line.

I run a HIF6 with vac position as Guys, and a Aldon yellow top 45D. When I bought the Aldon dizzi I was running 2x1.25’s. It has been on Aldon R/R with both set ups and no comments about incorrect advance - hence confused again as to what is correct.

Need Peter B’s words of wisdom !

richard b

My hif is sitting on my garage shelf. Before I took it off and put twins back on, I measured the vacuum from that pipe at idle. It's manifold.

That's why I raised the subject in this thread.

I had EXACTLY tbe same problems that LB Rose is having now, and I have a standard 1275 cam. As soon as I disconnected the dis' vacuum advance pipe from the carb, and blocked the carb pipe, I was able to obtain a stable idle below 1000rpm. Prior to that, I had to keep the revs up.

I looked for an explanation at the time, and found it via minimania and Keith Calver. Hence the link I posted.

I'll try and take some pictures later. But I remember at the time, looking closely and realising, that at idle, that hif carb pipe is open to the manifold depression.

I have read that the change was made after HS6 carbs (1978 ish?), which were still 'ported' vacuum. If I hadn't put my twins back on, my other plan was to fit an HS6. But I was so satisfied by going back to twins, even my old pretty worn ones, that I've never bothered. And I never got around to making measurements to drill a ported vac take off on the HIF. I'll measure it today.

I now think the only advantage I gained from a single hif, was not having to balance twins. But even that's a bit spurious, given that once balanced, twins don't need rebalancing. Just a fashionable swap I think. Mpg didn't change much that I noticed.

My first midget back in the day - a 67 1275 fairly early model had a 23D dizzy i.e. no vac advance - same as the Cooper S set up.

My current set up gives a pretty good idle at about 800 revs (266 cam) so Ill leave well alone !

richard b

That was one option. Run without vac advance.

A bit of research suggested I might lose a few mpg at worst, and add a bit more pollution to the atmosphere, which I admit these days is a good reason for emissions control.

My understanding of it is this:
Considering the manifold tract as a chamber with an extraction pump at one end (the engine) and a controlled entry valve at the other (the carb). Easiest for me to understand is when the engine is running fast (wanting to suck lots of air) but the entry through the carb is suddenly closed down as you take your foot off the throttle. (approaching that busy roundabout at 70, at the end of a dual carriageway). Manifold vacuum skyrockets as the air flow is cut off and any connection to the manifold (brake servo, crankcase breather, vacuum advance connection ??) would suddenly peak. This is good for the servo, and less good for the breather - so this has a PCV valve fitted!

But where, exactly does the carb cut off this air supply? It is surely the butterfly valve as the piston won't begin to drop so long as the manifold vacuum is high. It's only when the butterfly closes and the engine suction is isolated from the carb throat that the piston then drops, so there must be a momentary time lag. Essentially, it is the butterfly that seals off the manifold "vacuum chamber" so the high vacuum only occurs downstream of that. Upstream will pull a vacuum too, but much less depending on how free-flowing the carb, air filters etc are relative to the suction of the engine.

I am not going to take my HIF off just to check, but poking a wire through the dizzy vacuum port on my twin SUs the orifice is right on the lip of the closed butterfly, but on the carb, not the manifold edge of it. It is visible through the carb throat, but not when I look at it from the manifold flange side.

When the disc is closed the vacuum pipe IS isolated from the manifold. But not quite! How much depends on how tight shut the butterfly is at the tickover setting. I have not fiddled with those carbs since I took them off to fit the HIF but on mine the disc is more or less fully sealed at the top, with a sliver of daylight showing at the bottom. I'm not sure if that is by design though!

My guess is that in the overrun scenario the throttle doesn't fully close down to its tickover position and with the engine suddenly strangled for air (throttled) it will be pulling in air from wherever it can (worn throttle shaft and dizzy pipe connection included). I can visualise a linear graph of the vacuum being pulled at the dizzy pipe connection with a short term blip increase as the butterfly closes, followed followed by a drop as the engine slows.

I am not really familiar with dizzy advance curves and what might be needed. That is Peter B's speciality! But in simple terms I think when you want an engine to increase its speed you provide more atomised fuel and advance the spark progressively to encourage an earlier / faster burn. The vac advance does this before the engine speeds up and the bob weights begin to take over. But with the bob weights fully out does this then literally over take the effect of the vacuum? Or is it in addition to the vac advance? If the former, then on that type of dizzy a slight blip in vacuum pull at the start of the overrun when bob weights are still out, isn't going to make any difference to the timing is it?

All this theorising! I need to go and get oily hands!!

Just spent an hour revisiting Mr Vizard and found little of substance on the subject, which I was rather surprised at as it seems now to be quite important !

richard b

Here you go. HIF44. Manifold vacuum at distributor vacuum advance/retard pipe connection, as from factory.

Note the wire down the vacuum pipe.

The answers are, get an mg metro distributor as Guy has, get an HS6 with ported vacuum, keep twins and get pirted vacuum, or drill the Hif. I'll measure later for, as Calver suggests fot where to drill.


Note the wire. Throttle closed. FULL Manifold vacuum.

2nd pic throttle open, vacuum reduced by air intake as Guy describes.


Throttle closed again, back to full manifold vacuum. This is why you have a problem JB Rose. Not your camshaft.


Great pic’s
Have you measured the distance from the port to the carb flange ? Assume it’s more than the specified 9/16 but not a huge amount more.

richard b

Blimey, what have I started? Seriously, it's all very interesting and I have learned a thing or two. But I have to put my embarrassed face on, because I found the vac connection had fallen off, hence the fast idle. The pic herewith shows where I have it at present. There is also a connection on the manifold and I have used that in the past. From what you guys say I have it in the correct place now.

I have an A Plus dizzy with pertronix kit, so that should work with the ported vacuum. One breather comes from the front engine side cover, the other from the redundant fuel pump position. The PCV valve fits to the usual place on the rear branch of the manifold. The big tube on the carb is blanked off, just like your pic Anam. The idle is still not quite right, even with the vac tube correctly connected. It tends to drop off a cliff after a minute at around 800, and when hot sometimes jumps to 1200 or so. Tomorrow I'll try it without the vac advance and see what we get. I was quite staggered to learn that the A Plus dizzy's vacuum unit will retard the spark at some speeds.


L B Rose

🤣. So you have the right distributor for hif with manifold vac after all,-- and the pipe fell off? 🤣🤣🤣🤣. Excellent.

Still something to chinwag over.😁

Just getting some mrasurements now.

Edit. Ah reading further down, you still have an idle problem. I thought an A+ distributor would match the hif. Did they come from the same car LB?

Interesting that the little pipe connection on the twin carbs definately orifices (?) on the inside on the upstream side of the butterfly, whilst on the HIF its on the downstream side. Who would have guessed! Glad I used an A+ dizzy as I wouldn't have realised. What do people using aftermarket devices like the 123 dizzy do? Are they aware they need a different curve?

My question went unanswered though. The vac advance activates before the engine speed increases, so its effect comes in before the bob weights. But does their advance then overtake the vacuum advance or is the effect of the two processes cumulative?

With a dizzy connection to manifold vac, the suction must be greater on tickover when available air is strangled by a closed butterfly, and drop when you put your foot on the pedal so the arrangement at the dizzy must be completely the other way around such that it's a DROP in vacuum that cause it to advance. Is that right?

I investigated that for my Capri Guy. I **think** it's cumulative. I'll check.

And your point about the vacuum effect at the dis' being opposite in some way, is what I surmised too. Chris at octogan services confurmed it by saying it has a retard can instead of an advance can. I've found pictures where the canister has both advance and retard I seem to recall.

Edit. Just found this Guy.

"Vacuum retard was not used by Jaguar until emission controls mandated its use. Even then, Jaguar did not use the retard unit until 1970. The 68 & 69 models used a distributor without either unit. The mechanical advance on this unit was 20 degrees which gave 40 at the crank. Add initial of 5 and you get 45 degrees total advance."

"Vacuum retard came on the XKE in 1970 when Jaguar changed the air filter design to a flat style mounted directly on the face of the carburetors. The retard was necessary to control Nitrides of Oxygen or NOx. Peak combustion chamber temperatures caused excessive NOx so retarding the timing kept the combustion from completing and getting to hot. Timing on these cars was usually set with the vacuum line disconnected and plugged. You would notice when you hooked it back up that the engine would slow down."


Cumulative as I understand it Guy. The mechanical advances the rotor via the cam. The vacuum advances ( or retards) the plate via the canister.

From Wiki.
"Some vacuum advance units have two vacuum connections, one at each side of the actuator membrane, connected to both manifold vacuum and ported vacuum. These units will both advance and retard the ignition timing."


Yes, as you point out, the bob weights act to turn the rotor (cam) but the vacuum can is hooked to the points base plate so they are bound to operate cumulatively.

Crude measurement but good enough to illustrate.

16.5mm 21/32in, so greater than 9/16ths, from manifold flange to back of butterfly.


Measurement to vacuum hole.

Mark looks too low, but that's just the angle of the camera. It is gnat's cock level with the flange.

About 12.2mm. About 1/2inch, so less than 9/16ths.

So any new hole needs drilling back a little more than the 9/16ths Calver quotes by the looks of it. Or maybe that's because I measured the larger distance to the top of the butterfly, and at the offset 9/16ths will suffice.


Obviously not been doing much car work lately Anam. Your hands are way too clean! Or maybe you have been taking the Covid hand washing thing too seriously.

Ha ha. I wore orange latex gloves to take it apart. And then, "Now hands that do dishes can be ----" , complete for a well know brand that I used to wash off a bit of grime I picked up anyway. 😉.

It occurs to me that if the vac pipe fell off LB, you would have advanced the timing, as your dis' should retard it at idle. OR, maybe your retard can', has a leak. Same effect, not retarding the timing at idle. Counterintuitive, but could be correct.

I wonder if there's anybody over on the Metro/Montego boards who can add to the knowledge on this.

Perhaps not. From technical.

"If no one has ever posted here why does it still exist.
I think some of the forum locations should be deleted.
I wonder if anybody will read this. If you do please respond.

Good night."


Thanks Anon for the pics
I think my HIF6 was off a 1978/80’s Austin of some sort - so quite late.
My drilling /spigot is vertical and by eye is centred on the butterfly spindle.
The spigot measures 11/16 from the flange.

So if / when I take it off again I’ll do some more technical measuring !

Aldon don’t note any A series dizzies with retard and Vizard is silent.

Definitely need Peter B to put us right !

richard b

Found this discussion in the metro general archive from 1998.


Thread: '72 Distributor Advance Mechanism
Ross Smith, Austin Texas

I discovered awhile back (quite by accident) that Lucas made
both vacuum advance and vacuum retard mechanisms. One has
the vacuum port coming out the back of the unit, the other
out the side.

With the motor off, take off the distributor cap and suck on
the vacuum hose to see which way yours moves the distrib.
plate. You may have to search the junk yards to find the
opposite kind from what you have now.

Also, when you test the vacuum unit with the motor running,
make sure to plug the manifold tap. You mentioned it revved
up while you were testing - this extra speed would cause
the fly weights to retard the timing and maybe give a
false impression of what effect the vacuum unit is having.


Add your comment
Posted 13 March 1998 at 00:09:19 UK time Up arrow Down arrow

A. Big A-hole,
< I have seen in the info that came with my Webber kit >

Duh Jeff, It's "WEBER" not "WEBBER", ooo... I hate it when
people get it wrong.

Just kidding... your one of the most helpful people on this
Bulletin Board, and we all appreciate it.

Saul in Cincy, Ohio

Posted 11 March 1998 at 02:52:02 UK time Up arrow Down arrow

Jeff - Ohio - USA -
I have seen in the info that came with my Webber kit that the vacuum on the distributer is noted as vacuum retard. I have never seen this but it could make sense to retard the timing with vacuum. ( I left it off and plugged it as recommended in the instructions) The way to check it is to move the lever conected to the diaphram by hand and see which way it moves the points in relation to engine timing.
You can also check the "fly weights" for the mechanical timing by moving the plate and feeling the springs pull it back. Sometimes the springs and plate get rusty or dirty and don't respond.
I hope this generic info helps

Posted 10 March 1998 at 23:16:53 UK time Up arrow Down arrow

ChuckC Durham NC
I've been spending a lot of time lately tuning
and retuning the beast. Last night I pulled the
vacuum advance tube from the manafold to test the
advance mechanism.

When I pulled the tube off the manifold, I could hear the
vacuum, and the revs increased.

When I sucked thru the vacuum tube, while it was
connected to the distributor, nothing happened - no
increase in revs.

Seems reversed to me. What should I be looking for. Neither
Haynes nor Bentley are helpful about this.

Chuck ( OR

Posted 10 March 1998 at 12:57:14 UK time

That's odd. Ambiguous but there are a couple of passes there that suggest that the "fly weights" would retard the ignition. Surely this is wrong or am I just misreading it?

Whoops, actually I found it in the Spridget general archive. Looks like somebody fitted an A+ Dis' and made those discoveries.

I agree, I think he may have been confused about the weights retarding it.

Not sure I fancy drilling my hif. It seems to me it needs to be pretty accurately placed.

This is a company who understands. Unfortunately it's in the US. They also supply and repair vacuum retard units.

Ported Vacuum
The correct vacuum advance source is important
Understanding the operation of the vacuum advance unit
​with the PORTED VACUUM unit connection.
​Not applied “to manifold type” vacuum units.
Location is critical for proper operation.

They also describe vacuum retard.


Does this look anything like your vacuum unit LB Rose, or Guy?

This is a Spitfire mk4 1500 retard unit, to fit a 45DM4 distributor.


No Anam, this is mine.

I just warmed up the engine with the vac tube disconnected and blanked off at the carb, then set it to a rather lumpy idle at about 800rpm. I then reconnected the vac and the idle went up at least 500rpm. Reset back down to around 900 - still lumpy but acceptable. I find that with or without the vac connected, the engine is very slow to settle back to idle. I am used to engines falling back to idle immediately the throttle is closed, but this one takes several seconds. It will stay at about 1200 for maybe 3 seconds, and then seems to remember what it is supposed to be doing and slowly drop to 900 or so.

Anyway, I have learned quite a bit from this thread, of which one lesson is that if you stray from standard set-up you must expect to do a lot of fiddling.


L B Rose

Yep, I'm pretty certain that too is a RETARD unit. The vacuum isn't fed directly to the centre of the diaphragm. I believe it takes a route to the reverse side in order to pull the diaphragm in to retard the timing. Clearly must be working as the revs change when you connect it.

You could strobe the timing with the pipe connected and removed to prove it is retarding when connected.

The time taken to drop the revs may be due to the types of butterflies in your HIF.
"Around 1975 --- an over-run valve was fitted to each butterfly, allowing extra air into the inlet manifold when the throttle was closed." Found that info on an MGB posting.

Did you get your Dis and HIF from the same car? What was it?

Here's another retard unit similar to yours LB, but for a TR6. Same location for the vacuum pipe connection.

It would be worth reading the model number of that unit on yours in advance(no pun intended )of it failing, as it may take a bit of searching to find a replacement --- if it does fail.


Unstable tickover is sometimes due to air getting in along a worn butterfly spindle.
If it is worn, often it is much improved by just fitting a new spindle and doesn't need the carb to be re-bushed and reamed, which is a PITA.

Another thing worth trying is to seal the stub end of the butterfly spindle with a blob of silicone, after carefully painting the end of the spindle with grease so that the silicone only sticks to the carb body. At the other end, a squirt of spray grease will also help seal a worn spindle, at least for a while and long enough to see if it makes tickover adjustments any easier.

I cant see any reason to fit vac retard to an engine that has been modified for performance other than to improve emissions especially at idle - sort of German car makers type of logic !!

So I would either use vac advance or just use a non vac dissie.

Must check my HIF to check if it’s ported - another job on the list !

richard b

Here's another retard unit similar to yours LB, but for a TR6. Same location for the vacuum pipe connection.


Reference my earlier puctures showing the butterly and vacuum hole and measurenents. It doesn't matter where that hole is for it to be MANIFOLD VACUUM, if it's drilled in the top.

To make it PORTED vacuum, the hole MUST be in the bottom not the top due to the direction the throttle opens.

Obvious really, but I didn't notice it yesterday, despite Willy making a recerence to it, and that comment being acknowledged in my subconcious. 😉

A worn butterfly spindle stub can also be sealed by pushing on the later valve guide stem seal. The spring grips the spindle and still allows axial rotation.
f pollock

"To make it PORTED vacuum, the hole MUST be in the bottom not the top"

That may be true for the HIF carb, though I don't see why. For the twin SU it iS in the top centre, AND it is ported. Its simply the other side of the closed butterfly. As soon as the butterfly opens, its all one chamber with the manifold anyway. There will be a venturi effect on the orifice, as well as simple vacuum.

There must be something more about the positioning being very close to the butterfly edge in relation to the airflow or turbulance just as it begins to open and expose it to the outside, ambient air pressure. If it was just a matter of needing manifold pressure then an existing tapping to the manifold, with adapter maybe, would suffice. There would be no need for a connection to the carb, or a need to accurately drill a carb body if you have an HIV without a connection point.

The article Anon posted with the sketches notes ‘early carbs ported vac may be just under the throttle butterfly.
Later carbs ported vac is over the throttle butterfly’ - as per the sketches.
Do it seems more complicated than it first appears and does not seem to be quite as the Calver notes above either.

richard b

Some caution---not all HIFs turn the throttle shaft in the same direction----------
The pic. that Les put up of his carby clearly shows that it opens up turning clockwise being viewed from the front of the car--In this case a drilling for ported vac. advance would need to be in the top half of the carby body---It needs to be off centre to work properly because if it is dead centre, that is where the most airflow is at idle being the only part of the butterfly with an open gap and could cause a vac. in the vac advance hose, if in this central position
When drilling the carb for normal vac. advance operation the hole (drilling)should be hidden by the butterfly in the normal idle position so that there's vac registered in the hose---then as the throttle opens the hole is exposed and vac. is applied to the dist.
pic#1---showing hidden hole at idle, viewed from the manifold
pic#2---part throttle showing how far the butterfly moves to open the vac. port

William Revit

Yes, thanks for unconfusing me Guy. LOL. I was just looking at my HS2's and they open the same way.

Trouble with too much reading.
I've found numerous posts now on people who have already done this drill modification, to make ported vacuum. And they all talk about drilling the bottom of the carb body, and they are talking about the same HIF, with the same opening direction. Easy to get confused it seems, -- in my own defence. :)

But I've also just found a discussion on MiniMania with pictures of a modified HIF, complete with vacuum readings. This shows the drilled HIF. Brass ported pipe, next to the original, which he caps off, but reveals in this picture for illustration.

And a pic of an HS4 at just over idle, showing the ported hole.

Have a read. Very useful.

Edit. Just saw your post too Willy.


These are the vacuum readings. One taken at idle and one taken at 2000rpm.

Note the 'Blue' one (Manifold Vac) is showing around 17" and the 'Red' one (Vac Advance) is showing about 0.5".

Manifold Vac is around 19" and the Vac Advance is showing about 7" at 2000rpm

Looks good to me.


Willie, that pretty well exactly where the hole is on my twin SUs, 11 O'clock and upstream of the butterfly. Ported vac for use with a normal advancing dizzy.

I think for the manifold vacuum type we are ignoring the venturi effect. With butterfly closed it will have a high vacuum, and this will drop as soon as the manifold is connected to outside air (butterfly open). But in the early stage of opening, and possibly if the butterfly is slightly ajar with a fast tickover setting, the air speed through that small crescent of a gap will be high and the venturi effect will create its own vacuum in the pipe. I suspect this is by design to soften the sudden drop off in vacuum, smoothing out the change over at the dizzy. Probably also relates to Willie's comment about the hole not wanting to be at tdc


"On many cars it is found desirable to use vacuum operated ignition advance to obtain optimum part throttle consumption figures. The take-off point for this vacuum is arranged slightly to the air intake side of the butterfly, and in such a position that opening the butterfly allows the throttle disc to pass over the vacuum take-off point so that it then communicates with the manifold depression. By this means the vacuum is small at the distributor during idling and full throttle conditions, and is large at part throttle, being at a maximum when the throttle is open a few degrees."

Taken from the horses mouth. SU.
Scroll down a little way to the section titled "The H type carburetter"

Interesting readings Anam. I am surprised that such a relatively small change in manifold vac (17 to 19") has such a big effect on the dizzy vac (0.5 up to 7"). Surely that must be on a ported, not a manifold vac systm?

Ah, just read your later post (crossed) So that IS drilled to revert to the standard vac advance dizzy, as per original twin SUs.

Yep. That's from a drilled HIF (to obtain ported), that was previously manifold the same as mine. See the minimania link to the discussion and where the pictures came from.

Nothing new under the sun as they say. Someone's always already been there, done that, and got the T shirt it seems. lol.

Which is good news, for people like me, as I don't have to risk screwing up a good HIF :)

The other thing to consider, is the actual depression needed to activate the advance.

Looking at the 1275 distributor vacuum advance stat's, vacuum advance doesn't START until 5inHg, and finishes at 8inHg.

So I'm **guessing** that it may not be *too* critical, if the newly drilled hole exposes the vacuum pipe to vacuum a little sooner than ideal.

What say you?

Willy, do your racing mates use small chisels instead of screwdrivers, or is that a tuning mod. 😊

Nigel Atkins

Nigel's chisel beat me to the edit. lol.

So I'll repeat.

The other thing to consider, is the actual depression needed to activate the advance.

Looking at the 1275 distributor vacuum advance stat's, vacuum advance doesn't START until 5inHg, and finishes at 8inHg.

So I'm **guessing** that it may not be *too* critical, if the newly drilled hole exposes the vacuum pipe to vacuum a little sooner than ideal.

The MiniMania chap appears to have drilled at tdc, and obtained a good result with those readings.

What say you?
Willy? You think those readings look ok?

On closer inspection of the HIF, I think it's possible to drill the new hole, at the same offset as the original hole. --- Assuming that **is** the best offset, given that the original hole was drilled for manifold and not ported vacuum.

The original is drilled at an angle towards the engine, to ensure manifold vacuum.

I wonder if it were cut off(and blocked), to allow a vertical drilling immediately behind it, if that drilling would then come out in just the correct position over the closed butterfly.


That one looks like it was purposefully angled so as to come out on the downstream side of the butterfly. If that is yours Anam, it would be easy to plug that with a bit of brass rod and drill a new hole from the same point but either straight or angled back the other way.

Yep that is mine Guy. But all the pics I've seen show it at that angle.

Yes a bit of brass rod was my thought, but I figured the drill might drift if I tried to drill at the same point through a mix of ally and brass.

Maybe I'll ask the local garage if they'd do it on a pillar drill for me, since I don't posses one. A cheap one could cost the same, but are cheap ones precise enough?

Well it seems not all HIF's are the same !

Had a good look at mine which is an HIF6 which I think is off an Austin of about 1980 - so emissions time.

The spigot is vertical and drilled at about 11.00 o'clock - so off centre from max opening.

I pushed a strand of wire down it and it came out on the aircleaner side of the butterfly - its a thin bit of wire so has deflected forward a bit from the actual drilling.

Its interesting that there appears to be a sort of raised featuring around the drilling etc which may be on other carbs ? this would assist location if there.

So mine is ported ! perhaps SU got it right !

If you holiday thro' Zumerset/Bris'ol you can drill it yourself on my 'sensitive drilling machine' very good for tiny drill bits etc.


richard b

Wire down the hole !

richard b

Well that's really good useful knowledge for anyone contemplating an hif conversion in the future Richard. Wish I'd known that when I looked for an an Hif. Lol.

Pity I'm so far from your neck of the woods. I've got a neighbour with a pillar drill. I think it's a cheap one but I can experiment with a bit of 44mm plastic waste tube to see if it drills straight.

Looking back at LB's picture, you can clearly see the angle of the vacuum pipe. So now Richard has shown that the ported hif was drilled vertically, that definitely makes LB's manifold vacuum too.

Nice easy way to spot the difference.


I would not attempt it with a large pillar drill - total overkill with little control. It is light alloy so drill with hand drill - new sharp drill bits - lube with wd40 and use minimal pressure.

Centre punch and start with say a BS1 centre drill (small stiff drill just to get the centre started) then drill through for Port and drill tapping size for upper spigot connection.

If you want a spigot to screw in let me know and I’ll machine something up - probably BA thread.

richard b

Anam, no my carb does not have the overrun valve. I removed it and filled up the hole (as per Vizard's advice). The carb and manifold came from an MG Metro, and the dizzy is 998 A+ Mini. Maybe this is a mismatch? I had the car set up on Brian Slark's rolling road and they didn't remark on that.

I am puzzled that reconnecting the vac tube causes a big increase in idle revs. I don't think it's retarding.

Am I to assume, after all this fascinating chat, that my carb's vac port is in the correct position?

L B Rose


You can easily do the check I did - as per the pics. Also the pics show the location of the port.

Just remove the dashpot and try and insert a single thin piece of wire down the port tube and see where it comes out.

Aircleaner side is port vac.

richard b

If your idle speed increases when you fit the hose then it is increasing the advance
from that,your vac pot on the dizzy must be an advance type not a retard type and connecting it up and getting advance means it's on manifold vac . and not onto a ported vac fitting as it should be
This is the whole reason why your idle is unstable
You'd be better off with the hose left off and plugged rather than run it like it is

IF-You do go to drill the carb, make sure you do it on the opposite side to where that crankcase vent port is to get away from any other drillings AND the hole where it comes through near the butterfly only needs to be 2mm MAX and stay away from dead centre up the top, it needs to be down the side a bit
Them last 3 pics of Richard's are the go , as you can see it's off centre and drilled straight down so the exit point is probably around 30deg or so from the centre/top

William Revit

BMC or Lucas, didn't make it easy did they? Lol There are so many different designs of vacuum can, retard and advance, and some look externally the same.

LB, definitely your HIF is drilled for MANIFOLD vacuum, but clearly then your distributor is fitted with an ADVANCE vacuum can, when you need a RETARD can. The carb port is correct for a RETARD vacuum can, but not an Advance vacuum can which you have now.

As you already have a 45D distributor (assuming it IS a 45D), albeit from a mini and not an mg metro, instead of drilling the Hif, you can buy the correct RETARD vacuum can and bolt it on. Much easier to correct it that way.

Perhaps Guy could give you the correct part number, if it's easily visible on his Sprite?

It would be MUCH easier with the part number, as so many descriptions call them advance/retard units, which a bit misleading.

Richard, I don't know if I could drill it precisely enough to come out where needed, just above the butterfly edge. That's why I was thinking pillar drill. I'll take a gander. I'm not in a rush as I have twins on the car.

We have thrashed to the death where a ported or a manifold tapping shuld be. But what is taking me a while to comprehend is this change. After nearly 60 years of messing with A series cars, the words vac and advance sort of go together. Much more so than vac and retard!

The old understanding was that you have a pre-set tickover, or static timing, and as you put your foot down the vac advances the timing a little to assit in acceleration. Then as speed increases the bob weights come into play. All seems logical. The ported vacuum starts off at a low level on tickover and then increases steadily with speed, though never getting to the high vacuum of the manifold as when the engine is slowing from speed.

I can also understand that when a more lean burn anti-pollution set up was needed they may have wanted a system that retarded the ignition at the top end, and especially as the engine was running lean on the over-run. But surely you still need a system that advances the spark at lower engine speeds and on acceleration? A manifold ported retarding ignition seems wrong to me for when you want to accelerate!

I did find an explanation for the change somewhere Guy, I'll see if I can find it again. But some cars, some Jags I have read, have a vacuum unit that has TWO inputs. One that advances and one that retards. It's no doubt a lot to do with better emmisions control rather than perhaps just better engine performance.

Any chance you can see the number on your vacuum cannister Guy? -- Since you have a matched setup, it would be the easist way for LB to correct his I think.

I had forgotten, and perhaps we all had, but we have had this discussion before, covering pretty well the exact same issues! So you CAN live your life twice over after all!!

My MG Metro dizzy. Part number for vac unit is 54403883. I cannot see a makers mark on the unit, but it looks more like the Bosch variety used on VWs than a Lucas one.

Googling on the part number brings up a medical treatment for urinary tract infection.


I had forgotten, but I do have a spare as well. Same part number. Definitely a Lucas unit. Not sure what this one was off, though it looks the same.


You only live twice eh Guy?😉.

Shhh, you know who. Lol

Sounds like part numbers live twice too then, if it also describes something to do with UTI. 😄

Still that number could well help LB if he can source one cheap enough to avoid/make drilling not worthwhile.

I might see if I can find a bargain basement mg metro dis' complete with can. Used stuff comes up badly described, and cheap, on ebay all the time, so worth a look.


Wow, this is vastly more complicated than I imagined. I have the carb and dizzy on the bench now. Let me summarise:

* The carb port is on the top, just downstream of the throttle plate. I have checked it and it's clear. Does this still count as manifold vac? There is no port on the upstream side.

* The numbers on the dizzy are 41907A and 3786. There is no number on the vac can (which is almost new).

* The idle speed increased with the vac connected to the carb port not the manifold.

* There is no overrun valve.

* Throttle spindle wear keeps coming up, but I have replaced that and all the seals.

* Sucking on the vac tube causes the dizzy base plate to rotate clockwise, which I think means it is advancing. Of course, when fitting the new can I did this test and thought it was correct. The idea of a vacuum retard seems utterly bizarre to me. It's not mentioned in Vizard's book.

But if I have effectively manifold vac (port downstream of throttle) and vac advance, isn't that a correct combination?

L B Rose

Les, to add to your woes and decision making, the MG Metro dizzy is for an A+ design block and needs adaptation to fit the clamp for a 1275 Spridget. Plus the CI collar piece that the dizzy mounts in needs reducing otherwise the dizzy shaft won't reach down far enough to connect with the lower part of the drive off the camshaft.

It's not a difficult modification, but neither is it just a straight swap! But maybe one can just swap the Anti-vacuum can over?

"But if I have effectively manifold vac (port downstream of throttle) and vac advance, isn't that a correct combination?"

No it's the wrong combination. You are advancing the ignition timing by at least 8 degrees, at idle, and probably more than that, depending on the distributor vacuum cannister you have, because you are pulling full manifold vacuum at idle.

Yes, rotor anti clockwise, cam plate pulled clockwise by vacum can, is advancing the timing.

Guy, I would not even be talking about the A+ dizzy if I hadn't already modified it to fit the early block :). I removed the collar in the lathe and made a brass collar that fits the original clamp, fixed to the body with grub screws. Obviously I also changed the drive shaft.

L B Rose

LB there should be 3 numbers, apart from the model number, stamped on the can somewhere.

See this picture. I think it's off my Sprite. But the principle is the same. The 3 numbers in my case are: 5 8 3. This means, Starts to advance at 5insHg(of vaccum). Finishes at 8insHg. For 3 degress Distributor, which is 6 degrees Crankshaft.

You must have some numbers somewhere on the can, otherwise nobody could have selected it for that distributor.

Agreed, it is complicated, which is why (in part) I went back to my original twin SU's, when I discovered this problem. lol.


The distributor rotates anti clockwise
If the baseplte rotates clockwise with vac. applied then that is a vac.'advance' can which i repeat again is the best option for your car but 'not' compatible with the vac port that your carby has-
As your port is down stream of the butterfly it has full manifold vac, and will result in giving you full vac. advance at idle--THIS is why you have unstable idle, too much advance and an idle speed being controlled by your vac advance-
This causes the car to loose advance and stumble as you drive off instead of gaining advance and will cause it to feel a bit boggy off from a start instead of generating urge as it does hooked in properly
To run the correct vac can which you have you need ported vac.--as in Richard's earlier pics
Today I measured up a carb.
I think there are a couple of different throttle shaft depths from the gasket face ,this one measures 14mm from the gasket face to the centre of the shaft
The centre of the vac port is 21mm from the gasket face
At the point where the drilling comes through into the barrel is also 21mm so drilled straight down
The butterfly at this drilling point is 19mm from the gasket face
The hole as it enters the barrel--A 1mm drill fits through ok but 1.5mm won't--So being Brittish I'd say it would be 3/64"
BUT if your carb doesn't measure 14mm to the centre of the shaft these measurements won't work for you and you'll have to + - to suit
William Revit

Yep we all agree, that's all been established quite a few posts ago, not least by me. :) smiley face :). It's advance at idle when it apparently shouldn't be, and that's causing bad idle.

Moving on, apart from drilling, changing the vacuum can may be a simpler way to solve the issue.

Being simple minded my thought was just to change the vac unit, though be careful where you buy these from to get a proper made one, but I'd no idea even that was so involved.

If you can get old NOS or there's a bloke in USA that people rely on for these new(?) or Dissy Doc has stock of IIRC old ones.
Nigel Atkins

Yep Nigel, I agree, that was the conclusion I came to, too. Change the can.

HOWEVER ---- read on.

To quote you LB,
"Anyway, I have learned quite a bit from this thread, of which one lesson is that if you stray from standard set-up you must expect to do a lot of fiddling."

Well here is potentially more woe. And defintely more confusion. I'm confused anyway.

I found a Haynes MG Metro W/S manual. See picture.

According to the specs, the vacuum can IS an ADVANCE can, on the MG Metro.

Vacuum advance (maximum):
1275cc 14° to 18° at 8 in Hg.

Note: it says ADVANCE not retard.
Note: advance is at least 14 degrees at 8inHg. Full manifold vacuum is way above that.

And, these are the strobe timing figures, vacuum disconnected, at 1500rpm (See 2nd picture).

So, there must be something else going on, that allows the MG Metro to cope with manifold vacuum at idle, if it ISN'T a vacuum retard canister.

AND btw, the distributor is a 59DM4, or a 65DM4, not a 45D4, according to Haynes. It may be a Lucas or a Ducellier.

Clearly, there is a SIMPLE explanation for all this. What is it? lol.

Perhaps from reading those timing figures, the timing is retarded STATIC. So at idle, manifold vacuum establishes the correct 'static' timing?

Which, as it happens, IS, what I used to do to get a half decent idle when I had my HIF fitted to my Sprite. But what about the higher RPM's, AND when vacuum drops at open throttle?


Note that the post 1985 MG Metros have two very distinct static timings quoted for different engine numbers. Maybe this suggests a switch from advance to retard vacuum systems?
I suggested a while back when I posted the photos that maybe one could just switch the can. It will depend on what your dizzy looks like there. Mine is not the same as the normal Sprite one.

And the can doesn't have the 3 numbers stamped on it that you mention, Anam.

I am not sure one can rely on Haynes either. I also have the Haynes manual, but looks like it's a different edition to yours Anam. It's number 718. It lists the MG versions in a supplement at the back but only refers to electronic dizzy for the Turbo version and I am fairly sure that mine both came off standard MG Metros.

For the Turbo it has a line drawing of the dizzy parts with the air connection dead centre (advancing type) but also has photos in which one can clearly see a side entry pipe (retarding type)
Clear as mud!

You could be right Guy.

This is the version of Haynes that I downloaded.


I tried to upload the pdf but it's too big.

Here's the link to it so you can download it.

On the spine:718
Inside: 718-10J4

But there are errors and contradictions, as is often the case with Haynes, though on balance I think they are good and useful to have!

Far out guys ,this is bullsh*t
Les has the correct can (advance) and needs to drill the carby to make it work properly------
All these vac retard setups are for pollution requirements and all they do is knock the timing back at idle so the butterfly has to be more open to create more air flow to meet the requirements
The newer the car the more this happened,specially if the car has a cat. fitted, then the exhaust has to be hotter to keep the cat. lit up which they do by retarding the ignition and opening the throttle to maintain a suitable idle speed but the motor itself is under load at idle

Has the mechanical advance in the dist. been measured as in what is base timing set at and what is it at full advance--with the vac hose-OFF
If this dist originally had a vac RETARD can then it would have had very little mechanical advance in it and will need to be modified to suit vac advance setup anyway---This will be the deciding factor as to which way to go--

#--drill the carb in the correct position for vac advance , check the amount of mechanical advance and make it work correctly
#2--get a vac retard can and make it into a mixture of a good car with a crap distributor


William Revit

I don't think any of this discussion, or any of the suggestions, have been 'bs' at all.

If it were strictly limited to advice on what Les should do to fix his specific problem, then some of the discussion has ranged further than that since last week when Anam raised the question of where the dizzy vacuum pipe connected. I have learnt a lot from this but some of it it might seem irrelevant b-s.

I do still struggle with this retarding ignition as you start to accelerate though!

With a "retarding vacuum pot dizzy" - which still seems inherently wrong to me - perhaps the explanation is this:

1. Static timing is set with lots of initial advance.
2. On tickover, connection is from manifold side of butterfly which is closed so high vacuum which retards ignition from its excessive static setting to a more moderate level BTDC timing.
3. When you accelerate, butterfly opens, manifold vacuum drops as it opens to inrush of air and the lower vacuum then REDUCES the amount that the can retards by. Reduced retardation = advance!! So viewed this way the retarding ignition actually advances as you accelerate. Just as my last 60 years of working with these engines would expect it to! Perhaps this was obvious to everyone else but penny has only just dropped for me 😁😮
4. Finally, take your foot off the throttle (the fast approach to roundabout over-run scenario) and the vac skyrockets, adding more retardation to go with the now weak mixture and you have a happy engine that doesn't blow holes in its pistons!


Yep I like that explanation Guy. It seems logical and workable.

I've been trying to put that kind of thinking into words, for the opposite scenario too. -- Manifold vacuum on the Hif port, **advance** can on the distributor.

Ignition initially set retarded.
As soon as engine starts, ignition advances to where it should be on idle.
Open throttle, vacuum falls, revs rise and mechanical advance keeps timing where it should be. At high revs on overrun throttle closing(ed), vacuum rises, timing advances a bit more. -- Or something like that etc.--- And it could tie in with Haynes settings of strobe timing done at 1500rpm vacuum disconnected.

Your task is to complete or ammend. Lol

So both can be made to work, but it would be very nice to see an official/technical (bmc or lucas) explanation.

I agree Guy it's been interesting but we have to keep in mind that Les asked a question and has had quite a few confusing maybes thrown at him--

I do still struggle with this retarding ignition as you start to accelerate though!"
I guess this came from what I said about having a vac.advance pot hooked to full manifold vac.
I'll try to explain--
We'll take it that the vac pot gives 15deg advance
We set base timing at say 12 degrees with the hose off and set the idle speed, say at 800rpm, fit the hose(manifold vac onto the advance pot) timing goes to 27degrees and idle speed goes to probably 1200-1500 or thereabouts--it will definately gain at least 500 rpSo now the idle is too high so we adjust the carb down to get to 800 again, so now we're idling at 800rpm @27btdc --It'll be idling fidgety because of the advanced timing--same as if the dist had been manually screwed up tp 27 without vac being involved
With it set like this if the vac hose is pulled off at idle the engine will stall because of the extra revs that were being applied by the vac pot being taken away
This is why they feel soggy moving off from stand still set up like this, as soon as you go to drive off, the more throttle you give it the less vac to the pot and less ign advance until the revs build and the mechanical advance comes into play
I shouldn't quote figures so for example only--
We're at a standstill and creep away with high vac and 27 deg advance, then give it some light throttle and it'll loose some vac resulting in loss of vac advance timing so back to 12 all of a sudden and the loss of the extra engine speed that came with that which gives a soggy loss of instant throttle feel--not major but you can feel it

William Revit

You are right Willy, I have always thought the car was a bit soggy from, standing start. I am going to drill the carb for ported vac. I think I have absorbed all the excellent intelligence from here and elsewhere. Except that Calver is wrong about the distance from the manifold face - it is more than 9/16". That is where the existing manifold take-off is. I will either drill near to the dashpot screw, as in one of the photos earlier, or plug the existing hole and drill next to it. It will be good to get this right. Exciting times!

L B Rose

Thanks Willie. Actually when I said I was struggling to understand retarding vac when you accelerate, it wasnt with the mis-match components that you describe, very clearly. It was for the correct combination of manifold vac and retarding can dizzy. What I hadn't cottonned on to is that although it retards, it retards LESS as soon as the butterfly opens and the manifold vac reduces. As it is relative, the timing advances between tickover (manifold only vac) and initial acceleration, which is exactky as one would expect.

LB, being fair to Keith Calver, I think he was correct if you measure from the outside, not the inside where the hole emerges. The pipe hole is drilled on a slant towards the manifold.
I've just measured mine from the outside, and the pipe base looks pretty damn near 9/16" to the centre of the pipe. But as it's drilled angled towards the manifold, the hole emerges engine side of the butterfly by some distance, hence manifold vacuum. Drilled vertically at that point, I'm guessing as I haven't measured it precisely, I think the hole would emerge just where it would need to be for ported vacuum.

I look forward to seeing your result. I'll be happy to copy your successful outcome. 😄.

Yes Willy, what happens with manifold vaccum applied to an advance vacuum unit on a distributor, has been understood since it was raised by me, and exlained fully by Calver, in the link I included in my post.

What hasn't been fully explained (or may be it has been by my crude explanation), is how it might work **properly** with manifold vacuum at idle, applied to an advance vacuum unit in the distributor. Guy I reckon has explained how it could work with a retard vacuum unit in the distributor.

Guy, Haynes suggests --if accurate-- that there were advance cans in the distributors, and manifold derived vacuum source in the Hif. AND, further to that, I've been told now, by a Metro owners group member, that they were all manifold vacuum at the Hif, and advance cans in the distributors, confirming Haynes. But, another member suggests they were ported at the Hif. It begins to look as if different methods were applied in Metros, and all must have functioned as intended. Yours obviously works, so getting matched kit as you did is important. I'll post more as I find out.

Edit. I also asked a question on the metro section here. But no answer.

They clearly didn't switch from ported to manifold systems just for fun. If you think about the age of these changes it is before smart systems or on board computers. They were looking for something physical that could be used to trigger the timing change needed to match the new anti-pollution requirements, especially with the lean burn situations of high sustained speed and even more so on overrun. The obvious was to use the high manifold vacuum available at these times, and use that to reduce the amount of advance of the ignition. Ported vacuum doesn't provide a change in vacuum for those specific conditions. Hence use of manifold and a retarding can.

Manifold vacuum at tickover is higher than once the butterfly opens, but no where near as high as on overrun when the fast running engine is very greedy for air. But even so, that smaller vacuum drop as you start off can be used to reduce the amount of retardation and thus give the initial advance ahead of the bob weights coming into play.

I think I have caught up! 😀

I think Calver should have noted where the throttle spindle was in relation to his 9/16 dim.

Mine (measured in situ so a little difficult) is as I noted above 11/16” in from the flange.

I think Willy’s measuring also gave different dims, but were related to where the spindle was centred on the carb he was measuring - hence easy to interpolate if differing spindle/ flange mounts are encountered.

Les be careful not to move towards the breather drilling if you ‘go right’ on the barrel or the centre of the butterfly . I would check and if ok and clear drill behind the existing and fit a brass plug into the existing.

richard b

Richard, I have done exactly that - drilled behind the original one. New bit of brass tube inserted, with resin to be on the safe side, and I am just waiting for that to set before trying the engine. Calver doesn't actually say anything about the angle of the hole, but that it should be just upstream of the butterfly, and position not critical.

L B Rose

Pictures would be good for others if you took any LB.

Apart from me, you won't be the first or the last to have this problem. Worth pictorial documentation for the archives. 🙂

Had it confirmed, if it's accurate, that all Metros had an ADVANCE can on the distributor. And at least some had manifold vac take off on the HIF. As the mg metro was pretty nippy, it must have worked ok, so there must be something in the sekection of springs and weights for mechanical centrifugal advance, plus the base timings, that prevented problems at closed throttle low revs. But I've also been told that the mg metro didn't have a particularly smooth idle anyway. IF that is, that's all accurate.

The photo of the dizzy in my Metro Haynes manual clearly shows the can with side entry connection pipe, which would suggest a retarding one.

I wonder if people confuse the terms as although it sounds like a contradiction I think that a retarding can does advance the timing as the manifold suction is reduced, relative to its initial tickover position, that is.

Yep that's right. I've seen quite a few described as advance/retard units. As you say, it actually does both, as the vacuum increases and then decreases, the timing advances and retards accordingly. I've been looking for a BL official w/s manual online, as that would probably have the details, but can't find one. Triumph TR6's are listed as having retard vacuum in the distributors too. There might be a description in a w/s manual for one of those that gives more detail. Just out of interest I'll see if I can find a manual for that.

That was easy. Tr6 manual pdf on this google search link.


Triumph bikes, topped by Triumph lingerie for women ? Interesting, but I don't see a tr6 manual though.

Bingo. A description by timing, of how a vacuum retard systen works. See pictures. TRIUMPH TR6.

Static timing is set Btdc.
At idle timing is Atdc, courtesy of the VACUUM RETARD.

See page 86 35 00 sheet1.

1st column indicates those engines that did not have a retard unit.

Also, there appears to be a thermostat switch, that cuts out the vacuum retard, when the engine reaches a certain temperature. Complicated stuff.

2nd picture shows retard unit timing.

Not a Metro or a Spridget, but interesting and informative all the same I think, given all the previous discussion.

Edit. Lol. It is there Guy. Honest.


The link needs to be copied and pasted at least on my system as everything after Triumph gets lost when clicking on the link above and after doing so the TR6 manual PDF is there.
David Billington

I'll try again. Lol. Triumph womens underwear. 🤣

This should work.

I got the lingerie too. Triumph lingerie? Who knew?

Whoops not that either. I'll try a tiny url. Or just google triumph tr6 pdf workshop manual. That's how I found it.

See picture. Search for this.


Links probably worked first time, but I am easily distracted by luscious triumph pictures.
More interesting than figures in tables.

Haha. Nice figures ON tables would be more interesting. 😉

Now that Les is done-
Peter asked if there was a car with it's vac pot that ran on manifold vac successfully ----------

Here's a pic of a can you can get here for a Bosch dizzy
It's built for turbos, spring loaded in both directions ,it has to be plumbed direct to the manifold and gives vac advance and also retards with boost pressure--the amount of boost retard can be adjusted on the end of the pot

The only other cars I can think of that had full manifold vac. advance were mid 70's American V8 Ford V8s which were set up like that to try and get some economy driving around town---If you plumbed a manual gearbox equipped similar car up to have manifold vac on the pot, they'd be horrible to drive specially at around 30-35 mph in top gear, they'd jiggle and jerk at that speed and the tailshaft sounded like it was going to rattle it's way out of the car----The autos didn't suffer from it, I guess the sloppy old torque convertors sucked it up

On about 80 model 6cylinder Fords here they had both ported and manifold feeds to the pot through a ported vac switch in to the cooling system
While the car was cold they would have full vac to increase idle speed etc and then as the car warmed up the switch would flick over to ported and run as normal---a bit oposite to modern cars where the ecu retards timing when cold to speed up the warmup process

William Revit

Test run this morning. Once warmed up, engine settled to a steady but lumpy 800rpm. Drops back to that reliably. I have tweaked it up to 900 and it is better there. Photo herewith showing the new vac take-off just in front of the old one, which is sealed with silicone. I should explain that the old connection is shorter because it was bent, and broke when I tried to straighten it. I drilled out the remains, and then had second thoughts and put the shorter piece back in, just in case I needed it again. I will probably remove it again and plug the hole properly at some stage.

Many thanks guys, I owe you a round of beers. For the first time in years I have an engine that idles properly. I never knew this was such a complicated subject. It looks like it was simple until manufacturers tried one fiddle after another to meet emission targets. All in the past now engines are computerised.


L B Rose

Good news Les--happy it worked out for you

William Revit

Good news LB.

All the credit is due to Keith Calver really. That's where I first learned about the solution to this problem 7 years ago, and I think he wrote that long before that. But I doubt he was the first to know either.

Original is drilled in at an angle. I'm going to try and bend it vertical, by putting a thin copper rod through it to act as a former as I bend it; assumming it's steel. -- Is it? Must check.

It's only really been a problem, because we like to move away from standard, and didn't know, in this case, about the Hif being manifold vacuum on the Metro.

Obviously the Metros, at least some of them, were happily running manifold vaccum and and distributor vacuum advance. 1987 Honda Civic 1500 (Carb), that has manifold vacuum as the vacuum source for its vacuum advance diaphragm. I'll bet that now I know to look for it, I could find a lot more.

I've also now read, that ----;
a) ported vacuum is an effective method for rapid warm-up BEFORE the engine is fully warmed up ..
b) manifold sourced vacuum its the most EFFICIENT solution for vacuum advance for the engine AFTER the engine is warmed up ...
c) that the best solution for our classic cars is therefore to have BOTH ported vacuum (during warm-up) AND manifold vacuum (at operating temp).

All very interesting.😁


Well done Les.

Great feeling when a problem is finally sorted !

I would scrap the old pipe remains and machine up a bit of brass as a short plug with a slight taper and tap it in with a bit of retainer/epoxy just to be sure.

richard b

We have half a dozen MG Metros in the family, none have idle issues an all idle smoothly.

Rubber bumper Bs also run manifold vac advance at idle. Havent seen any stability issues. Yes, they are more susceptible to throttle plate wear resulting in high idle when the throttle stop is fully off, the solution is a new plate.

While I dont doubt Willy's reported problems with manifold vac and agree with the logic, the above two suggest it may not be universally so. Manifold vac raises idlw speed, allowing the throttle to be closed further, suggesting it may have been introduced as an early emmisions measure along with the timing of introduction, when euro stds came in the mid 70s.

I have successfully used the lower idle throttle opening enabled by manifold vac to reduce run-on in problem Bs in which all the usual other measures have failed. Agaon no idle stability issues. Maybe it is the influence of the primitive siamesed inlet ports....
Paul Walbran

Yes Richard, that old pipe was removed. I put a thread down the hole with a taper tap and inserted a grub screw, sealed with epoxy.

But the problem is not solved, although the symptoms have changed. On Saturday we were off at the crack of dawn to the show at Middle Wallop airfield. All was well until we arrived, when the idle had gone up to 2500rpm. Since then I have tried everything I know and I can't solve it. Let me explain what happens.

As I have said, the idle is pretty lumpy, with the engine hunting. Normally this indicates a rich mixture, but it's as close as I can get it by relying on plug colour, in the absence of a CO meter. I can get it to idle at about 700, which is unpleasantly rough, but if I turn it up the speed slowly increases until at 1000 it takes off of its own accord and settles at about 2500. I try to turn it back down and it won't stabilise until I'm back down to 700, when it threatens to stall. I have tried:

* Removing the ram pipe. No change.

* Refitting mechanical points. No change.

* Removing air cleaner. No change.

* Disconnecting vacuum advance (and plugging the tube on the carb). No change.

* Changing the carb piston spring, back to the one that was in it when I bought it. Not sure what that is, and I put a blue spring in a while ago. No change anyway. When the idle rises spontaneously the piston rises but not a great deal.

This problem only occurs when the engine is fully warmed up. Normal running is at about 190ºF on the gauge, and the cooling fan kicks in at about 200. So I only get this in slow traffic. Oh yes, I have also fitted an extended heat shield for the carb.

There has however been some progress. The engine starts much better. Previously, it usually coughed on first pull and then I had to churn it for a few seconds before it caught properly. Now it starts on first pull. I think the carb was advancing the ignition too much when the engine was being cranked, with the manifold vacuum very high on a closed throttle. When running, the engine is also more responsive.

If I lived nearer to Yorkshire I would take the car to Peter Burgess, so I'll have to get help more locally. Any suggestions for Wiltshire?
Les Rose

Engine hunting is usually a weak mixture I always thought.

Perhaps an air leak that only becomes apparent when hot?

There is also a temp dependant device in the fuel bowl on the hif. That could be acting up when hot.

Found this on a mini forum.

"If it is your bimetallic lever, your problem may rest with how oy set your idle mixture. The basic problem is that as you adjust mixture, the carb is warming up and the lever is ALSO modifying the mixture. Since you are doing it with the car stopped and the bonnet up, the heat from the exhaust manifold is over-warming the carb and lever, messing up your adjustment as you are trying to make it. You then set idle speed for an incorrect mixture setting. When you get out on the road, air flow through the engine bay is better, the lever readjusts the mix and your rpms increase."

"The lever sits with the end having the small slot pointing up. The mixture adjusting screw tip pushes against the slot. The main jet has an elbow at its bottom end, which fits into the keyhole in the 'toe'. The back 'heel' has a round hole through which a screw and spring provide a fulcrum point for the lever. So, as you screw in the mixture screw, it pushes the top leg inward, whcih moves the toe and jet downward, making the mixture richer (and confusing most people!).

The "toe" section is the bimetallic part. As it warms up, it curls upward very slightly, pushing the jet up and making the mixture leaner. Even though submerged in fuel, this is affected by heat off the exhaust manifold as described above."


Actually it's from spridgetmania.

Pics are:--
Original bimetallic lever is on the right. Modified version is on the left - dismantld and modified to remove the bimetallic effect.

Or from same discussion.

"I had virtually the same problem, as time went on, it got even worse (900 rpmm at start, 1500 warm, then went to 1800). Turned out to be the choke area of my HIF44. When I pushed in on the choke assembly, the RPM's would drop back down to normal. The choke assembly on my HIF44 still has a slight amount of play in it (in and out) but the rpm's stay steady now. Not sure if that play in the choke shaft is normal though. It was a simple .50 cent o-ring....."

I've run out of ideas now.😁


That's interesting. I wonder if that relates to the reputation that the early Maestros had with starting problems. They were notorious for playing up if you started and then stopped again for any reason within a couple of miles of running. It was the Achilles heel that just about did for sales on what was basically a very well designed car.

Here's how it's supposed to work.

Fuel Temperature Compensation (Viscosity Compensator)

This device alters the jet position in relation to the metering needle to compensate for changes in fuel viscosity which takes place with changes in fuel temperature.

When the fuel temperature rises, the viscosity is lowered, and in an uncompensated assembly, this would allow more fuel to flow for a given jet/needle relationship.

In the HIF jet assembly the jet head is attached to a bi-metal blade (1). This bi-metal is immersed in the fuel in the float chamber and will move in the vertical plane in response to changes in fuel temperature. The jet will be raised to a weaker position on the jet needle when the fuel temperature rises and will be lowered to a richer position when the temperature falls.

From this it will be seen that once the jet position has been selected by adjusting the mixture screw, alterations of fuel temperature will bring about slight alterations in jet position to compensate for the change in fuel viscosity.

The effect of this device is that drivability is improved over wide ranges of temperature, and the exhaust emissions can be kept within closer limits during cold starting and the warm-up period. Temperature compensation also allows carburetters to have the mixture setting pre-set and sealed before a vehicle is delivered."----

HOWEVER --- Read the whole discussion about problem idle speeds with hifs on mgb's,-- it sounds just like LB's problem.

That's where I found the description of the bi-metal device described above, and the problems it can cause.


Just for clarity Peter's not in Yorkshire but in Derbyshire, close to the border with Nottinghamshire.

M1 - junction 28.

Peter Burgess
Unit 1 Amber Buildings
Meadow Lane
DE55 7EZ

01773 520 021
Nigel Atkins

Forget all about the bimetal bit , it'll adjust itself--if it's assembled correctly
It really sounds like the idle mixture is way out, but 2500 rpm is a lot,
You mentioned plug colour as a guide but that can still be ok with the idle mixture up the creek
Have you set the idle mixture using the piston lifting method
What needle do you have init
William Revit

Yeah I wouldn't bother investigating anything loads of other people have reported as a problem. Might as well just drive it as it is, it'll adjust itself. 😉.

If the bimetal strip is straight and assembled correctly they work----------
to be honest I've never had a problem with one, if it's got the right spring on it and the jet moves freely in the tube it's bulletproof
The big issue with them is having the idle mixture adjusted when the engine isn't fully warmed up, that really throws them out of whack, it all has to be nice and up to temp to adjust the idle mixture properly

Les has had his carb apart and I'm sure he knows his way round it and would have noticed if the strip had been bent somehow

I have had one with no mixture adjustment and when it was pulled apart to check it had had the bimetal fitted upside down with the end cap forced on- but i doubt that's the case here
William Revit

Thanks Nigel, but Derbyshire is still a bit far for me. I have had the cold start apart yet again and found nothing wrong. Back together with a bit of Hylomar on a new gasket. Engine of course behaves exactly the same. There is endfloat on the cold start but pushing it in when running makes no difference.

I'll have another look at the bimetal thingy when it's cool enough to handle.

I am still puzzled by the horribly lumpy idle. I am just going to try removing the PCV valve.
Les Rose

Without the PCV valve the engine is - exactly the same of course. I'm just keeping this thread going to give you all the satisfaction of knowing what the diagnosis was - when I get to it.
Les Rose

I just went across town to Burlen and spent £25 on a new bimetal strip. Guess what - problem worse if anything. If I turn up the idle to 900 the engine suddenly takes off and revs to 3000. Since correcting the vacuum error something else has gone wrong, as whatever is, it's getting worse. I wish I had not sold the other carbs as I have nothing else to try. The chap at Burlen had no advice to give.
Les Rose

Sorry to say the new Burlen part making things worse than the existing part doesn't surprise me and the chap not being able advise doesn't surprise me either.

In the interest of balance and to calm those with reverence to SU - the new part making things worse may not be the part at fault but the new part further highlighting a pre-existing fault, and the chap at Burlen may have insufficient data to make a reasoned assessment.
Nigel Atkins

Well, here we are at bedtime and I think I have worked out what is wrong. All evening I have been replaying in my head the sound the engine was making, and I reckon the manifold gasket has failed. I shall look at it in the morning.

If I am right, I have fallen for a classic cognitive error, very familiar to accident investigators. I was convinced that it was all the same problem, when it was actually a new one. The new SU part didn't make it worse, it was getting worse on its own thanks very much. The gasket hole was just getting bigger. In my experience it's usually the exhaust part of a gasket that fails, so I'm not familiar with what it sounds like when an inlet port leaks that badly. I did check the gasket very recently, but that was before the trip last Saturday when the idling went out of control again. Sadly I don't have a spare in stock, so will have to wait for a delivery.

Goodnight all, and watch this space!
Les Rose

LB. Earlier you said-- "The throttle stop screws right up until I can feel the plate sticking, and still the idle is 1200. However this is much worse with the ram pipe."

With the throttle closed, if it's still revving, there must be another path for the fuel mixture mustn't there?

Edit, you posted before I had my latest thought. Lol.

I often think of more than one fault combining to make things confusing but know so little about all this thread subject. Other than last year with my engine and carbs playing me up it was four things but with one I thought I'd resolved but hadn't that kept kicking me in the balls. But as it was intermittent I do give myself a little leeway however I didn't fully follow my own advice of checking again what I'd doubled checked before.

Here's hoping you're right.
Nigel Atkins

I didn't say it again this time and look what (might) happens. 😁
Nigel Atkins

-Ahh, sucked in Nigella- LOL
William Revit

I was almost right. Nothing wrong with the manifold gasket, and all the nuts were tight. Not too tight, they are brass. I did find the spigot for the PCV valve was not as tight as it could be. Should have checked this before. I put some sealant on it and reassembled everything. Engine started on the button but was rough as hell as the mixture was way out. Another tuning session, and the idle is back to its slightly lumpy 900-1000. Residual variation is maybe due to the alternator kicking in and out. When the cooling fan turns on the revs drop at least 1000. The fan takes about 10 amps and the alternator has to compensate. Anyway, I'm back to where I was after correcting the vacuum error. Lesson: always look at the simple things first.
Les Rose

Well that's good news LB. Let's hope it stays that way.

you found it, well done.
Nigel Atkins

No I don't think it's a "well done" for me, I should have checked that first. But thanks for the kind words Nigel.

...and obviously I meant 100 rpm not 1000.
Les Rose

Les, you found it, well done for that.

You realised you should have rechecked what you'd already checked - some wouldn't, and you admitted and acknowledged your mistake - very many wouldn't, I think that's also well done to you. We all make mistakes.

Nigel Atkins

I'll drink to that Nigela--Well said
Well done Les and congrats for sticking at it
William Revit

Having finally gotten around to repositioning the vacuum advance port on my own Hif44, I bunged it on, turned on the ignition, and had petrol oozing out of the fuel bowl.

In the years in storage, the o'ring fuel bowl lid seal, gave up being a seal. So I ordered a new one. Then I checked my spares and discovered I already had FOUR new ones. I don't remember buying it, but I have an hif4 service kit, dating from 2012. Gawd knows why there were 4 seals in there though.

Anyway, ----

As Les said in his last post. -- "I should have checked that first".

Luckily I relised before it was shipped, and 'Minimine' on ebay are very accomodating, and cancelled/refunded without quibble.


I couldn't get the spout out of the carb body, so I cut it flush and blocked the hole.

Then I drilled a fresh hole, more towards the centre to stay clear of the original hole, just airfilter side of the butterfly, and at angle so that the opening inside would be at the same centre offset as the original.

Then I tapped in the remainder of the original spout, with a touch of loctite for good measure.

Fired it up, and seems fine, no vac advance now at idle.

Useful thread this. It's only taken me several years, and a further year since this thread last closed, to get around to this job.

And I'm only doing it as an experiment, because I already have the Hif44 and manifold bolt on ready from my previous use. I'll run it for a while, but may yet revert back to twin hs2's.

What's an Hif44 with Spridget suitable manifold worth these days I wonder?


One advantage, not a big one, is the extra space having a single hif44 gives, over having twins.

But if there's precious little, or no improvement in performance or economy, why did I (many of us) think this was/is such a great idea?


What're the pros and cons of running the heater hoses through the inlet manifold? Is it worth it?


The official reason for the heated manifold is to prevent carburettor icing. I do run my heater hoses through the manifold. Perhaps I should experiment with bypassing that.

A new HIF 44 is currently sold for around 545 pounds. I have the MG Metro manifold which is not correct as it's a 30 degree one. After much messing about I ended up chamfering the manifold flange on the surface plate to drop the carb by a few degrees.
Les Rose

I have an HIF44 on a Titan manifold (the correct angled version for Spridgets) I definitely found it an improvement on the original twin su when I first fitted it. But without a lot of checking and analysis I couldn't be specific as to why. The twins may have not been optimal condition or maybe not set up right. I think one common advantage of the single HIF carb is its easier to get the settings correct.

The prices being sought for a HIF and manifold are pretty high these days. If I was ever selling the car it would definitely be worth putting the twin carbs back on and selling the HIF separately!
I wonder what else that applies to! 5 speed kit, anybody? LOL!

Same manifold here Les. But I cut the phenolic spacer to drop the angle.

545 quid? Blimey. Think I paid a tenner, albeit not new.

Bit of a wrench going back to a spridget box though Guy. The T9 is a distinctly worthy upgrade imo. But as you say, if ever selling, sold separately, a t9 kit and hif44/manifold kit would be worth a tidy sum; -- with the addition of a frontline front shock conversion to push up the price even more.

But it's a bit like the money in your house. You don't want to be homeless, and you don't want to be Spridgetless, just to cash in on the wedge.

Just did a bit of reading about the heated manifold. Still confused. Getting cool air in through the filter is better for power, but then you heat it up in the manifold?

Engine tuner input required please.


I did say IF I was ever selling the car, Anam. I wouldn't 'downgrade' from the HIF or the t9 otherwise, or any of the many other modifications on the car. Doing all of those almost all DIY, has been for me a major part of the fun of ownership. Anyway the other one, the Frogeye, is more or less standard and a very different driving experience. No point in having two the same!

I agree Guy. Much as I may curse sometimes, and moan about getting old, playing around with the Sprite has been and still is an enjoyable chunk of my life. The diy aspect of it is a major factor, if only for the satisfaction derived.

Just gave it a tune, and had a blast. Too many years ago to compare directly with the last time, but it clearly is going well now with the single hif44. Before on manifold vacuum, my idle timing was shot, and I wasn't getting vacuum advance on part throttle. Now I am getting that advance, and I can feel it. Compared to the twins, perhaps just my imagination, it feels to have more poke for the same amount of lead foot. I'll have to take it to a long incline that I blast up on full throttle, and see how that compares to the twins.

But I reckon I'm happy enough to keep the hif on there until I fancy a change again in another x number of years.

Yep Guy I did note your "IF". 😁

Remember that film?


I was fortunate. My HIF was already one with ported vacuum, not manifold so no modification needed. Not certain what it came off as it was already off its donor vehicle and just in a crate of alloy carbs in the local scrappies. Possibly off a Maestro? Cost £5. And then got the Titan for £15. Must be about 10 or 12 years ago.

Heated manifold re-evaporates fuel which drops out of the airstream under some adverse conditions. These are particularly prevalent when gas speed is low (large carb, small engine, low speed) or when pressure rises suddenly in the manifold (throttle opened suddenly). Then there is inertia: with the single carb the flow is presented with two 90 deg bends, which larger fuel droplets are less inclined to negotiate, understeering themselves on the manifold walls.

Evaporating/better atomising fuel increases combustion efficiency, making for better fuel mileage and/or improved output. Apart from those benefits, tractabilty at low speed with the large carb is improved. Fuel injection achieves exactly all of these benefits for the principal reason that it atomises fuel so well. The last iteration of the Jag D type when fitted with the early Lucas fuel injection would reputably pull away from 10mph in top gear despite its high gearing.

As to temperature effects on the gas stream flowing through the manifold, it would be interesting to measure but I suspect that under full throttle they would be minimal.
If you take an A series at 6000 RPM (speed chosen as it is 100 revs/min to keep the maths simple) the gas will be resident in the inlet tract for 5 milliseconds, if that. I suspect not much heat could be transferred in that time. At part throttle it would of course be a different story: a longer time and also less gas mass to heat. But if you are on part throttle then you have already decided you dont need full power so the heating effect on power is irrelevant.
Paul Walbran

Thanks Paul. Very informative.

Sounds like I should/might as well connect the hoses then.

Will result in a neater installation, but with a small downside of having to drain coolant to remove inlet manifold for any reason.

It's actually the setup on my Capri with a 28/36 weber, and it hasn't been an inconvenience.

One thing I notice, the Hif seems to be smoother over the twins when the engine is cold.

This thread was discussed between 23/03/2021 and 21/10/2022

MG Midget and Sprite Technical index

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