Welcome to our Site for MG, Triumph and Austin-Healey 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 MGB Technical - HS4 Carb Setup

I have a 1979 MGB with a rebuilt engine, mild street cam and a little over bore on the engine. The exhaust is a PECO header, pipes and muffler.

I converted to twin SU HS4s and need some advice on setting the mixture. It has been over 25 years since I played with this type SU and have forgotten some valuable lessons.

1. The mixture nut on the bottom of the carb - which direction do I turn it to enrich the mixture? Clockwise or counter clockwise? ( Is in rich or lean?)

2. If I start with the nut all the way in, how many flats should I turn the nut out to start the process of setting the mixture? I presume that with the nut all the way in the needle is on the "shoulder".

3. Should I start with both carb mixture nuts screwed all the way in and then back out equal amounts to start the engine and tune.
Dick Field

looking down on the carburetors from the top, clockwise = rich and counterclockwise = lean. Start with the jets level with the bridge and lower them 6 flats (counterclockwise). Then you can raise or lower the jet as required for proper mixture.
Glenn Mallory

Here is a great web page on tuning the SU. Dave Braun does a great job of explaining.
Bruce Cunha

Sorry, the link is not working. Go to Dave's home page at and the SU tuning data is on the bottom of the page.
Bruce Cunha

Thank you Bruce. The article is relevant and helpful.
Glenn Mallory

These websites will help you with the set up and tuning of the HS carb

Andy Robinson

Glenn, Andy, & Bruce

Thanks, I'll read all this and work the car tomorrow.

Dick Field

Starting point is jets flush with the bridge then two full turns downs down from there. They will go higher, but not for the starting position.

Adjust for maximum idle, then really you need to use the lifting pins for fine adjustment. Can be tricky to differentiate that momentary rise for correct from staying up which is too rich, or dying which is too weak, but if you note where it just starts to become one, then count the flats until it just starts to become the other, and position between them, it will be about right. Experience and practice helps.

But first you much balance air-flow at idle and just off idle, and before that you must get everything else right e.g. valve clearances, points gap, timing, plug gap etc.



I've adjusted the valves, new plugs, plug gap is seen, air gap in the distributor is good, timing is set and I've done a check on all of the hoses for leaks. The manifolds were recently reinstalled with all new gaskets and torqued.

I'll work on the carb setup tomorrow.

All thanks for the advice.

Dick Field

Had an interesting time trying to tune these two carbs.

I got them balanced at idle using a uni-syc from my old MGA days.

I counted 24 flats up before the jet was flush with the bridge on one carb and about 30 on the second . This doesn't match anything I read in the above post or articles. Looking at the plugs, the car was running rich (dark black) color.

I adjusted to two turns down and didn't see the response of change in idle either going up or down on the adjustment. I'm down to 17 flats now and the car is running lean. The plus are bright after a short trip.

Car doesn't idle consistently. Exhaust has a burble sound.

Has anyone experienced having to go down 24 or more flats on the mixture nuts?

Any ideas?

Dick Field

I have just spent a great 3 hours with Peter Burgess, Keith and his dynamometer, plus my V8 roadster. It was running, not too well when I got there and it took much experimentation and head scratching to get to the bottom, actually a few "bottoms"! One was that small clip/locator which is on the damper rod (not always fitted, but these are new) which was jamming in the piston as it lifted. Another was simply that the "leading" edge of the venturi bridge was rounded off and not a sharp edge, that was remedied and made an surprising (to me) difference to the fuelling, i.e., fuel pick up. The K & N air filter, supposedly good for up to 220 bhp, was starving the engine, so was removed, even the filter housing mounting plate affected the fuelling!! The most inexplicable difference was removing and plugging the vacuum advance. No dodgy diaphragm and in theory it's presence should only have an effect at less than part throttle. But it led to a persistent misfire all through the rev range!! As yet I can't explain that!
So in answer to the question, it could be just one problem or several more subtle issues.
Wrong or badly installed needles, what needles are you using? Blocked float chamber vents, sticking piston(s), insufficient fuel supply. There's a reason why you are having to, in essence, over-fuel using the jets.
Balancing and mixture adjustment is best done around 1200 to 1500 rpm. Don't use the idle screws, but get the speed and do the balancing on the throttle cable. After all, that is the way the carbs spend their fuctional time. Then use the idle stops to balance at idle and give your preferred idle speed.
Allan Reeling

Having to wind it down that much indicates are lack of fuel, or a vacuum leak, about two full turns down from flush is the starting position and it shouldn't vary much from that.

The latter is easy to check using carb cleaner sprayed round all the carb and manifold gaskets. You should do a delivery check at the carbs for the former by directing a feed pipe into a container and switching on the ignition. It should deliver a minimum of half an Imperial pint in 30 secs, and in practice more than double that.

After that it is a case of checking the piston drops back onto the bridge with a clonk when raised by hand on a stopped engine, float height, piston drop time, and stuff like that.

A blocked float chamber vent/overflow will cause flooding out of the jet, as air pressure prevents the fuel level and hence the float from rising.

Air-flow balance does have to be done off idle as Allan says, but it does have to be done at idle first so you can get the interconnecting shaft clamp screws tightened in about the correct place. Then small adjustments to one or other clamp so you have the same balance off idle as at idle. When you finally adjust the idle screws to get the desired idle speed, you must always and only ever adjust both by the same amount in the same direction, or you will mess up the balance again.

Paul, Once the throttle connecting bars are clamped off to give balance on the cable pull, you can't upset the balance as such, you just have to set the idle screws to give dance while on the stops at idle. With inter connectors slacked you have to use the idle screws to get the balance before you tighten them back up, this is a bit of a fiddle until you've done it a few dozen times as the linkages tend to "wind down" a bit once you back off the throttle stops. The secret is to make a slight adjustment to allow for this.
Allan Reeling

Paul and Allan,

Thanks, I'm back at it. Let you know what I find.

Dick Field

Hi Dick,

Go to this forum page, I have posted full instructions, from go to whoa.,3500815,3500979#msg-3500979

Unfortunately I can't post here because it is a PDF file.

H J Adler

I have had that little missfire problem before
The early3.5 dizzy runs a more agresive mechanical advance curve than the bigger engines
I had one in a 4.4 and you could feel it all the time not a real missfire as such but more of a little offbeat minor semi jerky feeling and sounded harsh around 3-4000
No pinging at all
Pulled the dizzy apart and removed a bit off the outside ends of the weights and reassembled and all good
If you have an early 3.5 dizzy and your engine is 3.9 or bigger it might be worth checking out
William Revit

Willy, What you describe is more or less what it was suffering from. But just blanking off the vacuum advance cured it, all through the rev range strangely!!
My dizzy is a touch modified anyway. Restricted max advance with springs to control the curve. My current theory is that for reasons to yet be determined it's getting vacuum at all revs!!!
Allan Reeling

Allan - you can upset the balance by adjusting one idle screw more than the other as there is free play between the pin on the clamp and the slot in the 'lost motion' lever on the spindle. Maybe only at idle, but still.

Hi Paul, well you have to set balance at idle, plus while on the fast idle cams, which I forgot to include in my thesis above..
Get your point on the float chamber breather, makes perfect sense.
As for balancing etc., we all have a method which works for us. I only ever slacken one connector clamp, wind down the idle screw till the carbs balance, add a touch to allow for twist, tighten up, then wind off the idle screw and see where we are in terms of balance.

Hi Willy, the engine is a mildly tuned 3.5. Lightening the bob weights would be similar to changing the springs tensions for stronger ones, i.e., slow down the centrifugal advance as you suggested. But what is the mystery is why blanking off the vacuum advance removed the "miss-beat", upping the output as well. One goes with the other I suppose!
Allan Reeling

Just a thought - could the vacuum signal be pulsing and upsetting the advance mechanism?
Chris at Octarine Services

"I only ever slacken one connector clamp"

Indeed yes. Yet another wrinkle, is that the butterflies must be sitting on the idle screws with the throttle pedal released, with a smidgen of slack in the throttle cable, not hanging on the cable.

Allan's B only missed on part throttle, with resharpening the bridges to return the pressure drop signal on the jets back to that which nature (Mr SU) intended the oart throttle miss was above 3500 rpm only which disappeared when Allan blanked off the vac advance. I never checked what the advance was but I suppose it could have been lots and lots and maybe caused the problem.
We could calibrate fuelling to accommodate the soft edge bridge under wide open throttle but could not sort part throttle fuelling, until we sharpened the bridge when we had to revert to a much weaker needle, BBW. BBC was perfect with the soft bridge edge but too weak cruising. BBW sharp bridge good WOT and cruising fuelling.

Peter Burgess Tuning

Just got back home from Buxton after a sun drenched journey!!. The right hand carb was jamming again (the 6 mile syndrome I suspect!!), that bloody clip were the culptrit so they have both gone for good!! Checked the vacuum advance by lung power....... 8 degrees max, which according to my data is what it should be, but of course have no idea how it rises to that???
Allan Reeling

Lucky you - if you had come south east you would be frozen ....
Chris at Octarine Services

Allen, you say that the clip on the damper rod is "jamming in the piston". Do you mean it is jamming in the bore in the suction chamber ? It clips in the piston tube, so it moves up and down with the piston. There should be plenty of clearance between the clip and the bore in the suction chamber. I guess that a bent damper rod could cause jamming, but would be pretty obvious.

But here is a random, and probably unlikely, thought. What carburettors do you have ( HIF6 I expect on a V8, not the HS this thread started with ) ?
Do they have the ball bearing suction chambers ? If not then maybe the bore in the suction chamber is a smaller diameter and the clip does foul ? I have only a HS4 to hand to compare with a ball bearing HIF6, and the clip would have almost no clearance in the HS4 bore.
Hopefully somebody will have more detailed knowledge, maybe actual measurements etc.
J N Gibson

Is this the clip that retains the damper piston in the oil chamber? My V8 has them - pressed into the oil reservoir, and the ball-bearing suction chamber. The damper rod passes freely through the clip, so the main piston with needle can rise and fall freely. As such I can't really see how the clip on its own could interfere with the main piston.

When unscrewing the damper cap and lifting it up it rises until the damper piston reaches the clip, then if you lift further it will lift the suction chamber ... until you lift too much and the clip pings out.

There seem to be dire warning about not dislodging this clip, but only early ball-bearing suction chambers had them, and one of mine has been dislodged for years with no apparent ill-effects. It seems to me the only purpose is to keep each damper with its correct piston. Maybe the clips did cause a problem and were deleted, or maybe someone just decided to save a groat or two. As the damper cap can only be lifted a short way before it starts to lift the piston, the clip does have the unwanted side-effect of making it more awkward to top up with oil than if the damper is completely removed.

On mine out of the oil reservoir (where it would be slightly wider then when it is fitted) the dimension at 'A' is 0.565". The diameter of the suction chamber is 0.620", so no chance of fouling when correctly fitted. Loose like mine, I suppose there is a chance of the clip rubbing on the inside of the suction chamber, but as I say it doesn't seem to have caused a problem.


If the clip comes loose from the piston tube it can cock over to one side and jam on the damper rod so hampering the rise of the piston.

Best either to spread the clip so it is a firmer fit in the tube or delete entirely.
Chris at Octarine Services

That was exactly what was happening. It may be significant that the dampers were brand new!!! I say no more but they have been deleted, along with their potential to cause a problem!! I couldn't see their function apart from Paul's theories that is. I only mentioned this problem to illustrate that carb problems can have multiple causes.
Allan Reeling

Incidentaly, this shows (section 1, 5a) the spring clip I have and an alternative circlip. However it describes them as 'bearing retainer', which makes it seem as if the purpose is to prevent the bearing assembly between the piston and the cover from riding up above the top of the oil dashpot in the piston.

This thread was discussed between 24/04/2017 and 06/05/2017

MG MGB Technical index

This thread is from the archives. Join the live MG MGB Technical BBS now