MG-Cars.info

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

Recommendations

Parts

MG parts spares and accessories are available for MG T Series (TA, MG TB, MG TC, MG TD, MG TF), Magnette, MGA, Twin cam, MGB, MGBGT, MGC, MGC GT, MG Midget, Sprite and other MG models from British car spares company LBCarCo.

MG MGB Technical - HS 4s Carburettor setting

I have always used thin oil in the dash pots, usually a 20 grade or cycle oil but recently watched a piece on YouTube from John Twist of University motors and he recommended using 90ep oil which surprised me, his thinking was the thin oil was fine on new carburettors but as they wear heavier oil is better. According to him it gave better acceleration .wondered what other’s thought.
Trevor Harvey

Never used anything other than 20W/50 that I had for the engine, for both 4-cylinder and V8.

If the oil is too thin the symptom would be the piston rising too quickly and not enough initial enrichment on acceleration which would cause stumbling until it had 'caught up'.

Without stumbling then I wouldn't see a need for EP90, although if the mixture had been enriched to cure stumbling because of a worn damper/dashpot it will worsen the mpg in steady running and in that case a heavier oil might be a benefit.

However! It's a misconception that 80/90EP is significantly heavier or thicker than 20W/50. The V8 used 80/90 gear oil in the gearbox whereas the 4-cylinder uses engine oil of course, but it doesn't make the gear change any heavier even in very cold weather as some claim. Gear oil has additives needed for the torque involved in the sliding action of gears hence being used in the V8. The 'numbers' for gear oil are in a completely different range to engine oil so the two don't get confused and using engine oil instead of gear oil engine oil would cause damage. The actual viscosity of SAE80 gear oil is 7cSt, which falls between the viscosity of 20W engine oil at 5.6cSt and 25W at 9.3cSt. SAE90 gear oil is 13.5cSt which falls between SAE40 engine oil at 12.5cSt and SAE50 at 16.3cSt.
paulh4

Whilst SWMBO watches the soaps, I've spent a lot of time watching various YouTube videos on SU carb's and their tuning. Mainly because I've always found it difficult to get and maintain a good state of tune and performance.

I've come to the conclusion that any carb(s) fitted to any MG (even new or fully refurbished) is likely to be a compromise. Changing anything without correctly measuring the effects is just one more compromise! John Twist (and Paul) has a point that heavier oil will cause greater enrichment during acceleration and/or compensate for wear but to what effect? Too rich wastes fuel, increases emissions and could potentially cause bore wash.

Therefore, I'm now of the opinion that the only way to get the best out of any particular engine requires a proper rolling road tune up by someone who understands what each part of the carb is there to do e.g. float, jet/needle, dashpot spring, damper, oil weight etc. A "static" tune with an AFR will get you closer to the ideal but it can't replicate the effects operating loads.

Tuning with an AFR meter at various engine speeds and up to full throttle is a must as anything else is a compromise in some way. For example, how can you ensure the needle profile is correct for an engine that's probably been rebuilt or runs on modern unleaded fuel for which the original needle was simply not designed? Setting the mixture at idle only means the AFR at idle is correct but as soon as the carb moves away from idle the AFR (at steady speed) is determined by the needle profile. There are so many profiles available and who's to say the standard one is still the most appropriate for any given engine? It probably isn't as it was likely a compromise by design!

I guess what I'm saying is we shouldn't meddle with the variables unless we can measure the impact. But that won't stop me trying ;O)

best of.....
MGmike





M McAndrew

Changing the breathing can have a huge effect on how well carbs perform, fitting K&Ns and and on the V8 tubular manifolds in place of cast iron is a case in point. My V8 with both of those when I bought it had a flat spot when accelerating from part throttle cruise - unless I floored it. I looked at needle profiles and chose a 'richer' one which cured the problem.

Like mechanical distributors carbs can only ever be an approximation to what is ideal for a particular engine at each combination of throttle opening, engine load and rpm. With engines like ours even when new having been manufactured to less than tight tolerances the settings of both were designed so that they ran without obvious problems and were 'safe' i.e. didn't result in engine damage - something they didn't get quite right with the Twin Cam MGA. It's always been a source of some amazement to me that they can be made as good as they are. Compare the SU with a fixed-jet carb and surely they can't be as good, getting their additional performance at the expense of fuel economy which is of little interest in a high-performance application.

I have known people go to huge lengths with AFR meters and electronic distributors to try and get the best performance and eventually give up. That was without a rolling road, but even with one of those available it would need many hours, and to be brutally frank to what end for most of us on a road car?

If the rolling road is used to put engines under load at high throttle openings you can end up causing damage. Peter Burgess uses the no-load method where he floors the throttle to max rpm and records the rpm profile for each tweak of carbs (which may include different needles) and timing. Unless you have a specific fault such as stumbling and mis-firing that's probably as much as we need. If everything else is done right in the first place (which is why carbs are done last in the setting-up process) then SUs should maintain their settings over many thousands of miles.
paulh4

As Paul suggested, the oil and piston is there to off-set the mixture discrepancy, when the throttle opens quickly, i.e., the fuel is slower to react than the air. On fixed jet carbs that off-set it crudely done with a small pump, to squirt fuel into the Venturi. SU did it more intelligently by slowing down the piston rise to give the fuel time to catch up with the faster moving air.
I've seen countless views on which is an appropriate oil for the piston, and like Paul have stayed with multigrade engine oil, partially because it "adjusts" it's viscosity with engine temperature, so can be more accurate, more of the time. Certainly better than a shirt of fuel and fingers crossed.
Throw in thin cycle, or sewing machine oil and it will be like water when hot, hence a flat spot!!
I don't know how temperature stable gear oil is, but look at the chart for viscosity eqivalents.



Allan Reeling

All good but I believe the SU dashpot oil is a straight 20 grade oil---multigrade oils do change their viscosity differently (slower) than straight oils with increases in temp.
Is it possible that SU uses the straight oil to compensate for temperatures more acurately---As in when the oil/engine is cold there would be more mixture enrichment when needed(cold) then as the temp increases less enrichment is needed the oil thins and lets the dashpot piston rise more quickly--------just a thought.

I use atf in my strombergs but that's purely to protect the diaphrams----atf is roughly the same as 20w engine oil at operating temps so roughly same as SU oil--???
William Revit

My WSM text does say "top up with new engine oil (preferably SAE 20 ..." but the lubricants chart shows same as the engine oil grade for your local climate. SUs 100ml bottle works out a lot more expensive!

I can go years without having to add any, but then I only unscrew the cap, lift it, press it down again, and if I can feel the resistance of the damper before the cap reaches the piston cover I have enough. Information is conflicting in that some indicate the oil should be below the top of the 'hollow rod' i.e. only above the damper whereas others indicate it should be above the top of the hollow rod. As the piston slides up and down on the outside of the hollow rod I'd imagine you would need to keep topping it up if it is above.

There are 'plain' piston covers and 'ball bearing' types with a plastic sleeve containing 12 ball bearings as in the attached (the plastic sleeve is visible from above just by unscrewing the damper cap) which could explain the difference between below and above but not all info for the ball bearing type shows it above.

I've always maintained it below on both my roadster plain covers and V8 ball bearing covers for over 30 years each with no ill-effects.


paulh4

Thanks for all the advice, I have used 20/50 oil in the dash pots and the car definitely picks up better but I have just finished a carburettor and distributor rebuild so it’s difficult to tell if the dashpot oil has made a difference Trev
Trevor Harvey

Nothing has been mentioned about Jet Height!
The jet height should be .060"-.065" below the bridge!
Each flat (on H4) = .0063. 10 flats = .063". This is the best for fuel atomization ( and was where they were set at from the factory.
Once this is set NEVER change the mixture again!
Len
Abingdon Performance Ltd.
Len Fanelli

The WSM instruction is two full turns down from flush with the bridge, or two full turns for an HIF screw, rather than poking something down inside to measure.

But that's only the starting position, then adjust the idle screws for a decent idle with equal suck from both carbs. Once started and fully warmed up the coarse setting is turned one way or the other to get the highest idle. If either one has to be turned a lot or one more than the other then there is a problem somewhere.

Fine tuning on the HS is then with the lifting pins, which needs practice and experience. But probably not needed now as HIFs do away with that and are turned towards the weak side a tiny amount until the idle speed just starts to fall, then turn back just enough to regain that highest idle as per the attached drawing. I can't see why that can't equally be applied to HSs.

You might need to make small changes between winter and summer, but once correctly balanced for air-flow and mixture only ever adjust both by the same amount in the same direction or you will have to start all over again.


paulh4

The US racers, Manley Ford & Derick Chima, have done lots of research on this subject along with dyno testing.
The SU rebuilders usually set the jet height at .075" expecting that the customer will lean it out 2 flats.
IF your jet height is not very close to .060"-.063" below the bridge you WILL notice much more torque at tip in throttle at highway speeds.
Oh, 90W gear oil is the same as 20 weight motor oil!
20W50 synthetic motor oil is recommended by a Few SU experts.
Len Fanelli Abingdon Performance Ltd.
Retired ASE, Ford Motor Company & Land Rover Master Technician.


Len Fanelli

Some set it up as per the factory, then measure it, then use that as a future reference. If nothing else that can show there are differences between carbs that shouldn't be there. These were never close-tolerance machines and especially after all these years no two engines or pairs of carbs will be identical in terms of what they need or how they best operate. Which is why - for me - setting things up dynamically to suit my car is preferable to static measurements, as with timing and dwell. Taking just one component the 'fixed' nylon float has a tolerance of 3.5mm in float height, that will affect the fuel level in the jet and that will result in a difference in jet height for the 'ideal' mixture. It does beg the question as to why rebuilders set it at a specific figure 'expecting' the customer to lean it out two flats instead of setting it to 0.062".

Sharing our preferred ways is a good thing.
paulh4

Thank goodness the carb settings are as individual as the owners, no modern one size fits all here! At least we can all tinker without laptops.......I had to stretch the dashpot spring on a Commer 1600 camper van with an SU conversion to get it to run just so on Tuesday.
Peter Burgess Tuning

Just to throw another method in--Tend to use the CO meter now for idle mixture as it's not so hot running temperature dependant.
Do the jet height measuring equalisation -I set them at .065" initially and check the needles are set right, warm him up and aim for 4.5-5% CO. It's usually right in there at that. I find this method works well especially on engines with wilder camshafts, which seem to be able to mistakingly soak up over richer idle mixtures without any noticeable idle quality change--at least with a co meter it helps understand exactly what's going in there.

Len--have you seen the later online Ford Master Tech program, It's enough for you to walk away shaking your head in disbelief, nothing like the old original practical make it and do it course. All multi choice tick the box sh*t.
William Revit

CO for setting mixture is fine for twin exhausts where half the cylinders are on one and the other half on the other but on single exhaust like the MGB you can only average the two carbs, one might be rich and one weak. The book does say to use CO as the final stage on emissions tested cars but if rich only to adjust both carbs by the same amount in the same direction the minimum amount to get down to the limit then put it back after the test. That's what I had to do on the V8, although after a couple of years of that I just weakened them by a quarter turn and that was enough. Until I forgot to put it back one year if not two, and the next test came back with 0.2% and 190ppm.

'Bangers and Cash' last week had them picking up an ancient Crypton Engine Analyser. They asked an old boy who spent his working life on cars of our era when they were new if he had ever seen one and he said they had one but never used it saying "we used these" pointing to his eyes and ears.
paulh4

He probably doesn't use a tension wrench either then Paul--keeps it in the cupboard---

I do agree with using a CO meter, scope etc after a tuneup to check results though and for diagnosing possible issues.
William Revit

Along with other tools I inherited this from my father. The theory was that the average mechanic wouldn't be able to overtighten using it ... as long as they didn't use a length of scaffold on the end ...

paulh4

This thread was discussed between 16/01/2026 and 28/02/2026

MG MGB Technical index

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