MG-Cars.info

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

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 - Something I've learned about SU carburettors

We all know that setting up the SU carburettors can be a bit frustrating, but after 17 years with my B GT I liked to think that I had it taped. However, I've discovered something that makes such a difference that I'd like to share it with you all.

Assuming that you've adjusted the tappets, fitted new points, set the ignition and so on, the next thing is usually to balance the air flow at idle. I use a Gunson Carbalancer for this and it works well. Next comes setting the mixture and I can't use my Gunson Gastester for that because it samples only at the tailpipe. So in the past I've fallen back on my Colortune, the lifting pin technique, and generally listening to the exhaust beat. I've been happy with the results for years - but now I have something better.

After rebuilding the carburettors it is usual to set the jets directly level with the bridge and then open them out the same number of flats as a starting point. It occurred to me that if both carburettors were sucking the same amount of air, they should need the same amount of fuel, and therefore the jets should be the same distance down from the bridge.

So, I removed the dashpots (taking the opportunity to clean them) and using a vernier gauge measured the distance from bridge to jet. The difference was tiny but all the same I added both measurements, divided by two and set each jet exactly to that distance. When I'd reassembled everything I put the Gastester on the exhaust and found it slightly rich so I weakened both jets by one flat.

The difference on the road is amazing as the engine feels smoother than ever before and even SWMBO noticed it from the passenger seat. Now this is probably something that many of you already do as standard, but for those of you who haven't tried it then I recommend buying a cheap vernier if you don't already have one and giving it a go. Even without the added refinement of the Gastester you may be pleased with the result.
Miles Banister

Good one, Miles.

I came across a cheaper, simple and accurate way to balance the carbs, using a couple of bits of (soft) wire. See picture.
Adjust the ends to be level, when stationary, start engine, set to fast idle and adjust throttle stops till ends are level again, tighten linkages.

Simple.

Herb

Herb Adler

Miles-
See nearby thread "Need SU needle suggestion..."

Herb-
He is talking about mixture balance, not throttle balance. If you dare, go to MGA board and read our discussion re carb vacuum adaptors!

FRM
FR Millmore

Miles - out of interest what year is the car, does it have HSs or HIFs, and what were the figures before you made them the same?
PaulH Solihull

Paul - It's a 1973 with twin HSs. I've stupidly lost the piece of paper I wrote the measurements down on, but I'm almost sure that one was 1.16 and the other 1.19 and so I adjusted to 1.18 before backing off one flat.
Miles Banister

Hi Miles

We start at 60 thou with 1 1/2 HS carbs and 85 thou with 1 1/2 HIF carbs.

Peter
Peter Burgess Tuning

Hi Fletcher,

I am aware that he was talking about mixture balance. I was responding to his comment on using a Gunson Carbalancer, and an alternative to an expensive (anything greater than $0) item.

Herb
Herb Adler

Hi again,

Fletcher,

I have just been reading that thread in the MGA board.

What I can glean there is that the two wire, or similar rod type device is a perfectly valid way to balance the airflow through the carbs.
I have used the Unisynch, and found that it was somewhat difficult to accurately set the airflow, due to the bouncing around of the little ball.

Herb
Herb Adler

Thanks Miles - 1.16 etc. mm i.e. 60 thou, I assume. Out of interest I checked my roadster (which I last set-up years ago after they were rebuilt and haven't touched since) and got .0445 for the front and .0440 for the rear i.e. a twelfth of a flat difference which I was pretty pleased at. That would appear to be pretty weak though, but it does and always has run extremely well. The V8 were obviously lower at .1080 for the right and .0745 for the left, but I blame Halfords for that as unusually they had to tweak the carbs last month to get it to pass emissions. Normally I avoid that by weakening it before the test, but obviously I didn't do it enough this time. I did have a quick fiddle afterwards to restore the mixtures, but obviously need to do it again a bit more carefully.

The other thing to remember is that the benefit of the lifting pins (and air-flow meters) over measuring, like dynamic timing over static, is that it takes account of small differences between the carbs as every component has a tolerance, as well as wear in anything other than brand-new and perfect carbs.
PaulH Solihull

Hi Paul,

It sounds as though your roadster was pretty much spot on. It will be interesting to see if you notice any difference after tweaking the V8.

I'd always believed that the lifting pins were the best approach, but what I found suggests that my ability to interpret their results wasn't as good as I thought!

Miles Banister

Major problem is that people do not attend to the bit that says "Lift the piston 1/32inch". The piston lift does not begin when the pin moves, and the pin will lift the piston far more than 1/32". {1/32" = .031"}
Result is wildly out of tune mixtures.

FRM
FR Millmore

Ok I've bitten..
FRM I've tried the pin lift method many times and have given up in favour of Gunson technology which isn't perfect I admit. So, how do you limit the lift to 1/32=.031" on a pin that can lift about 1/8 or 4 times the amount needed or more?
I know the amount of lift makes a big difference to the results eg allows more fuel and air in and may run faster if rich and stall if weak but I just can't judge how much a 1/32" is when leaning over the wing with hands fiddling around under carb's between the body and the filters. And what tolerance is acceptable (if you can measure it!).
And, taking in your lesson on the MGA board ie how one carb affects the input to the other set of cyls, do you do one carb at a time or both together?

Me thinks the class needs another lesson...

MGmike
M McAndrew

Mike-
Calibrate thy paws!
Easiest thing is to use a piece of .030 wire under the piston with the filters off. In actual fact, .060 or so isn't bad, so a piece of welding rod is decent, moreso if you file it flat-tapered.
One carb at a time, usually, which makes it more sensitive. But, you do single carb engines the same so...
I actually usually use a small screwdriver, stick it under the lip of the piston, and twist to lift controllably.
Or I stick my fingers in both carbs and do it by feel, alternately, on things like MGA which are hard to reach or see into.
Keep the screwdriver or fingers off to one side so you do not interfere with airflow.
More than specified lift amplifies the effect, but destroys the sensitivity, so you use it to get "close". Then correct lift to finish.
If a lot of lift, like .250, does not cause drastic effects, that carb is not contributing, go back to synch.

Curiously, if one carb is very rich, and the other very lean, the "read" gets reversed somehow, such that you are adjusting it more in the direction it is already wrong. We never could figure this out, so if the results don't make sense, go back to basic setup.

Throttle connections loose, idle screws in one turn, jets down 10 flats or .065 is my basic setup. Fast idle about 1500 with choke out a half inch just before it starts to pull the jets down. This is usually far closer to correct than any car that comes in the door.

When you get it right, put the filters back and lift by feel. It should be the same or faintly rich if the filters are somewhat restrictive. Lean out one flat usually. With the filters on, 1/32 is about what you can be certain IS Some LIFT; if you did the filter off part right, you can use that response to judge lift.

The rpm change by lift is less than 50rpm max.
I like about 10rpm for the faint increase part; if you lift it .010 it will stay there.
Too fast idle obscures things.

All this is of course based on a better than scrap carb, correctly assembled. Contrary to popular opinion, worn throttle shafts are not usually a big deal; it does mean that you need to adjust idle a tad lean according to the lift result. Worn needles require slightly rich at idle, so worn but serviceable carbs tend to equalise needle wear with shaft wear at idle. I got to where I could tell how to compensate by how much wear, and how much to compensate for expected ambient temps.

Now you figured all that out, go find an XKE or better yet a three carb Healey on a hot day!

PS: I have a Colourtune, it is wherever my UniSyn is; both gifts from people who sold their cars. And I have been pretty much an SU specialist since 1966.

FRM
FR Millmore

I do my SUs the same as FRM and you get to know them pretty well. One little trick I picked up while tuning a T series on the dyno a few years ago, (but you dont need a dyno for this to work). With the airfilters off. Run the engine at a fast idle and pass the back of your hand past the carby intake so you can just feel the air flowing past your fingers. If the revs rise that carb is lean. You do need to keep the throttle position constant. You can then try it at higher revs. I usually do it when finished and just now and then a little tweek is needed. It mainly works if a carb is lean. I use a wideband A/F meter for overall mixture which is great but one carb can be balancing out the other. This is not replacing the lift the pin or piston but just a handy hint and remember almost any airfilter will enrichen the mixture a little so dont over do it. Denis
Denis4

The SU Toolkit (or just some bent wires like Herb illustrates) make it dead simple to see when you've lifted 1/32" or whatever.
Rob Edwards

Would it be possible to fit some sort of collar around the lift pin to limit the lift to precisely 1/32. I need to sort my carbs out this weekend so will check it out.
Steve Church

I doubt it, at least not easily. You'd need to know where the piston will be at that particular RPM and then set the limit at that +1/32" Impractical, if not impossible. Do Herb's wire thing and just lift about the thickness of the wire, more or less. It doesn't have to be _exactly_ 1/32", but if the book just said "lift a very small amount" then no one would know what a "very small amount" was....
Rob Edwards

"So, how do you limit the lift to 1/32=.031""

You guess! It's only a smidgen, you should be able to feel the additional weight of the piston when that starts to rise, and you should soon learn how little extra movement is needed to make the revs change according to whether it is weak, rich or just right. I doubt anyone was born knowing what it feels and sounds like, you have to train your ear and your finger by moving it between deliberately rich and deliberately weak, then as your ear gets more sensitive the two points at which you recognise weak and rich will get closer together i.e. less turns of the nut or screw. If you can't recognise the momentary rise of just right, and on HIFs it *is* very difficult, more so than HSs in my experience, then setting it mid-way between just rich and just weak should get you close. It's important with HSs to run the engine at 1500 rpm or so for a few seconds every couple of minutes to 'clear its throat' as fuel pools in the throat, one of the mods on the HIF is to provide a passage way from immediately after the jet bridge to under the butterfly to automatically clear this.

It has to be one carb at a time as changing one carb will then affect the other as they are on a common inlet tract, so you need two or three iterations after apparently getting the first one right, to get them both right.
PaulH Solihull

Miles - I've set both to .088, weather dire today but hoping to take it to Wellesbourne tomorrow to see the Vulcan bomber in action.
PaulH Solihull

Idle seemed lumpy as if the cylinders were unbalanced. Needed less choke on a cold start than before, but that could equally be both carbs a bit too rich. But on the lifting pins the left carb was obviously a lot rich, and the right carb a bit weak. Readjusted to the lifting pins I had to weaken the left by about 3/4 of a turn, and richen the right by about a quarter, smoother now. That does tend to confirm the dynamic benefits of the lifting pins as opposed to the static setting of jets by height, and it does also show that there is a difference between the two halves of the intake system. Whether this is a difference between the carbs - they are at least 100k old although have had new floats, valves, jets and needles in my time - or a small vacuum leak on the right side I don't yet know.

I tried my Gunson's Carbalancer on it afterwards but it has been iffy for some years now. When level the starting CO can't be reduced to 2%, it is always above. However if I tilt the unit to one side or the other it goes way down! Standing on it end and adjusted to 2% in free air it went up to 4.4% with the probe in the pipe. Removed again, and left for ages to stabilise, it needed calibrating again to get it to 2%, and next time up the pipe showed 3.5% So as I say iffy, which is why I haven't used it for years.
PaulH Solihull

Paul do you mean Gunson's Gas analyser instead of carb balancer ?
Dave
dave bignell

Paul, That's interesting, and not what I expected. I wonder if it makes a difference that this was on Vee rather than Bee? Does Vee have one carb for each bank of cylinders perhaps. However, it does rather cast doubt on my assumption that if both carbs are set the same then their mix must be the same. It looks as though maybe the best all round approach is to use the vernier gauge to set a starting point, and then follow up with FRM's small screwdriver trick with the lifting pin.
Miles Banister

Dave - yes, thanks for the correction.

The two banks are isolated from each other, unlike the 4-cylinder which has both carbs and all cylinders on a common inlet manifold. The V8 plenum has cross-over ports that result in each successive firing cylinder being fed from alternate carbs, to keep a steady pulse rate (for a given rpm) through each. This is because the firing order results in 4 firing immediately after 8, and 7 after 5. I presume the reason for that is a combination of best mechanical balance when the pistons are going up and down as well as firing.

So whereas on the 4-cylinder altering one carb has a small effect on the 'other' two cylinders, the V8 doesn't. But in both cases a different number of turns from the same starting point does indicate a difference in the two intake systems, for whatever reason.
PaulH Solihull

Note that the measured jet height necessarily assumes all other parameters are nominal and identical between carbs. Careful work with not badly worn parts usually does the job, but there are traps.
FI, "the carbs - they are at least 100k old although have had new floats, valves, jets and needles in my time - "
Question is when, and how does it relate to this issue. These carbs are most likely bias needle type, and needle replacement is recommended at 20,000mi. But the needle starts wearing flat on one side, on installation. If it is left in the initial position, it soon wears in to a comfortable fit with the jet and the wear rate drops; if the needle is turned in the piston whilst being cleaned or inspected, the whole wear-in starts over in a new position. This is a way to have substantial differences between carbs at idle, even at fairly low mileage.
The repeated wear-in process also wears the jet, so turning needles causes increased jet wear too, which gives significant mixture changes both at and above idle.
So, if the needle in one carb is turned differently from the other, both jet and needle can suffer noticeable wear, unpredictably differing between carbs and over time/miles.

Numerous other carb parameters could affect this, and so can non-carb factors - pretty much anything in the engine. I try to get the carbs themselves as balanced as possible in all respects, then if there are big differences between final required settings, I look for what I missed on/in the engine.

V-8 do not have the "banks" separate, rather the two end cylinders on one side are mated with the two centre on the other, in the usual conformation. This gives two induction "manifolds", just as the siamesed port four has, and in the same way, most (practically all) V8 manifolds do have a balance passage between the two. Some simply have a big plenum with the carb(s) dumping into it unseparated, the branching being post-plenum.
In some cases these passages are built into the carb itself, with two or four barrel carbs. I do not know what the Rover/MG SU manifold has. A source of perplexity can be exhaust heat crossover passages, if used and especially if not functioning correctly.

FRM
FR Millmore

This thread was discussed between 10/06/2012 and 26/06/2012

MG MGB Technical index

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