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 MGA - Air fuel ratio gauge

Has anyone installed an air fuel ratio gauge to their MGA/B? What do they think, findings? One of our guys has just fitted one and is experimenting with jet setting on his Weber (around $300 on the internet).
Mike Ellsmore


The A/f sensor is a lambda sensor that measure the oxygen in the exhaust and it is normally used mainly to optimise the operation of catalytic converters and emissions. There's a good wikipedia article on A/F sensors.

When I used to map racing engines in the 80's with electronic fuel injection, Jag XJR5, RS 200 etc, we didn't bother with many sensors except throttle/ air flow and manifold pressure because we optimised each engine for power output at every throttle setting and load. With a carburettor its always a compromise because you can't tune them for all throttle positions and loads. With production cars you can't map each engine uniquely hence more sensors for feedback and control.

My point is that it needs to be a trial and error experiment. Aiming for a particular A/F ratio also might not give you what you want, ie best economy, best power etc. The A/F sensor will just give you some indication of how close to stoichiometry the engine is I think but may not help optimise the performance.

I expect modern racing engine management is much more sophisticated now but emissions and fuel economy and transient performance are more important and developed than in my time.
John Francis

I haven't got an in car gauge but I do have a transportable A/F ratio meter and it's worth it's weight in gold--you can't set up properly without one
A mate has a wideband O2 sensor gauge in his motorkhana mini thing with a weber and was able to really sort it out using the gauge---the advantage of it was that the reading is instant so on the weber you can see if it needs an earlier/later emulsion tube or if it's a main jet issue straight away
Good thing about them is that the gauges are switchable from lambda readings or A/F ratio to suit what you're used to---and things like a little missfire will show instantly where the older sniffer units like mine will just show a gradual leaning from the missfire which if you're inexperienced can send you searching in the wrong area

If you do go ahead and get one ,don't forget that MGA/B engines like it a tiddle on the rich side on max power if you want it to live a long life, specially if there's some ethanol in there
William Revit

Thanks Willie, that was my thoughts as well. It would be nice to know the A/F ratio as you hammer up the back inclined straight at Sandown (the scene of a major disaster enginewise some years ago). My car has been setup with A/F on a dyno after the last engine rebuild 12.2 :1 at full wack but it would be good to see in real time. That said they arenít needed for standard A unless you are a bit of tech nerd like a few on this BBS.!
Mike Ellsmore

Hi Willie and Mike
Always good to know where my knowledge and experience is lacking! Is it particularly relevant to setting up Webbers? I'm not familiar with them at all?

John Francis

Reading your post, I doubt that you're lacking in knowledge or experience at all--
Interestingly I have a friend here that has an ex rally RS200 ,quite a machine,a real race car if ever there was one, I'm yet to have a ride but it will happen
Webers are interesting carburettors and very adjustable
Just a quick crash course
The venturies can be removed from the carb. body so venturi size can be selected to suit the needs of the engine, then we have the main jets and air bleeds which are bottom and top of the emulsion tube
As well as selecting the correct main jet, the timing of when the main starts working can be altered by selecting an earlier or later emulsion tube which doesn't change the main jet mixture, just when it comes in and then the final full load top end mixture can be fine tuned by changing the air bleed size to make the main jet suck harder or less at the top end. Also the idle jet comes into play as it also fuels in the transition period between idle and main
Unfortunately emulsion tubes and the transition number on the idle jets don't run in number order so a good chart is a must unless you've got a good memory
They can be a challenge at times but when you get them right they're magic
There are some simple initial setup formulas to follow which usually get you really close for engines with one carb. barrel per cylinder but on MGs with two cylinders per barrel a lot of operators make the mistake of following the chart which quotes engine size--eg 1800cc but the chart is for two carbs whereas the MG having only two inlet ports only runs the one carb. and the whole setup, venturies,jets etc needs to be larger

Starting to sound like a Weber salesman--

With the O2 sensors the older type operated at 0.5V for stoich and had a very narrow operating band and would really only operate as a on/off sort of signal at that point (14.7:1)but with a later wideband sensor the range is wider and allows them to be used to drive a gauge which can read from 8:1 through to 20:1
William Revit

Hi Willie
That's really interesting. I can now understand why a sensor would really help with setting them up.

Incidentally I never had a ride in an RS200 but did get one in the Group B Astra which never raced before they pulled the plug on Group B rally cars. A colleague also had a ride with Malcom Wilson in a Metro 6R4. It was an incredibly hairy, clenched buttocks, experience!
John Francis

Well- there you go, what a coincidence , my mate Tommo has one of Tony Pond's cars,an '86 championship car I believe, it's at the doctor's at the moment having its gearbox rebuilt, it's a non synchro dog box and something went wrong with the selectors or something and won't change gears

We had one crash big time in Targa Tas a few years back, the owner/driver ,John Potter had it wound up and about to select top gear and the screws that hold the steering wheel onto it's hub pulled out and the wheel came off, apparently they were the wrong size thread or too short or something and he had only recently had the wheel fitted---not a happy man
William Revit

Good day Mike And William [!!] --communicating via the U.K. !!
I have an exhaust gas analyser-I built it up---never used it. A Jayco's Kit project. It uses coloured diodes as indicators

I wonder if my normally asperated -- EFI experience with fuel tables might be of interest, to the present discussion. And what fixed my issue---
My EFI system circuitery was designed to run only 2 injection nozzles.
They fired once each only , in each cycle-into the 2 induction ports.
Literature indicated that there were going to be unbalanced mixture issues on the M G 's 5 port cylinder head--due to the 1432 firing order and the 2 induction ports servicing 2 cylinders each.
. This un-balance was later so proven to be true, after installing 3 plugs and tempory test lambders into the M G A's exhaust manifold, and comparing the exhausts.
SO What to do to balance it up?
A carburettor responds to the vaccuum created in the induction -- the metered fuel charge is sucked through into the combustion chamber.
The EFI works differently-- The computer fires metered fuel into the induction chamber--It becomes atomised and then sucked in.
Say--If no-1 cylinder receives the first available charge into its combustion chamber--then No-2 cylinder will only receive the left-over remaining fuel charge on its induction cycle.. So, the fuel charges differ.
By adding a second Driver into the ECU's circuit--each injector was made to fire twice in each cycle. So the fuel was then available, as the second cylinder required its charge from the same port.
The Tables in the ECU are altered normally- with a programmed Lap Top-We do these "on the road" & on a Dyno--
Great--results- instant starting hot or cold---no choke cable- the engine does not ever " run-on" 32mpg economy.--and balanced combustion. -
I.W. Cowen.

Ha Ha Should have proof read it----


Damned Computers print what ever they wish to!
The laugh is on me! I guess the experienced guys noted the mistake!
I.W. Cowen.

We're old enough to overlook a typo Ian
Interesting the no1 and no2 cylinder setup

In the firing order no2 fires before no1
I'm not sure what would happen with single shot fuel injection, it would all depend on the timing of the squirt I guess as to which cylinder dipped out
but with a carburettor no2 would suck it's load in first then no1 would start sucking it's load while no2's inlet valve is still open and no1 robs some of no2's load------the bigger the cam the more of a problem this is---With SU's and a manifold with a balance tube it's not too bad but with a Weber with no balance tube and a big cam the port robbing at lower speeds is enormous, The flow from the carb gets sucked, stopped and sucked again and no1 takes the easy road and robs no2 to the point where you can pull the two centre plug leeds off at idle and it make no difference
Scatter patern cams overcome this at higher revs but down low if it's a big cam and a Weber it's going to happen----no big deal really
William Revit


We, Lucas, developed the very first fully sequential engine management system in the 1980s so each injector fired once per cycle at the right time. Interestingly it was our work on fully electronically controlled diesel injection systems that enabled us to have the technology and software processing to do it on petrol engines. On a diesel you have to time the injection precisely in time and in fuel quantity. The Metro 6R4 was one of the first cars to get the system.

The most difficult issue we always came up against was making sure the interference from high powered spark systems didn't interfere with the electronics. We used to have to play with various suppression techniques and plug lead and cap solutions.

John Francis

PS Lucas got the initial patent on having a map for EFI ignition and fuelling and made a fortune around the world on licensing it. In those days Lucas was very innovative. It was good times to work there and the machine shop often helped make bits for my MGA!
John Francis

Thank you "William", for respecting both of our ages.
I should have been more explicid---The end 2 cylinders always run lean. We found that---and the reading supported this--since 2 and 3--together--have first opportunity with the fuel charge, it leaves 1 and 4 out in the cold starving! ! .
Early on with the project, I had a Jeep derived stepper motor, controlling the Idle. A messy hulk of Aluminium. Lots of pipes and wires. Did away with it-- Since the linked throttle bodys are throttle butter fly adjustable--It was nicer to balance the inductions--A 'la S U style so obtaining a nice idle. The Air Cond. works at idle without stalling. Hearing the injectors rattling while balancing the butter flies is interesting. I built a special manometer to effect the adjustments.
The original throttle bodys used 8 years ago were were Ex- 1 1/2" Syzuki motor cycle , units.
Moved on now--using injectors angle fitted into the ex-bells of larger dia. HS6 1 3/4" S U carburettor bodys.
I have used a connecting Vacc. tube---after the 2 throttle bodys ---on the Vacc. side, to set up the ECU's MAP vacc---the fuel return valve vacc, and also the Vacc. brake booster.
I am having a special cylinder head prepared at Windsor, N S W to make it run nicely. The results are going to be interesting.

John F. My first foray into EFI was with an Ex-land Rover hot wire manifold system built up unto an Ex-SD1 Rover V/8 engine --all built into a fab. M G B Body shell. This inspired me to advance on with an EFI system on our M G A Coupe' Great experiences.

Life is too short 1 !
I.W. Cowen.

I note that you mention a fuel return vac.
Do I read that as you are running variable fuel pressure reliant on vac.----
William Revit

This thread was discussed between 19/03/2020 and 21/03/2020

MG MGA index

This thread is from the archive. The Live MG MGA BBS is active now.