Tuning the SU carburettor on the T-Series MG

Section 1 – Needle Selection

On the SU carburettor, recall that the diameter of the needle at each piston position from a closed throttle (Position 1) to fully open (Position 9) in one eighth inch travel increments, determines the area of the annular space between it and the 90 thou. Inch jet orifice. In operation, because of the highly turbulent flow through the annulus, the weight of fuel passed to mix with incoming air is precisely regulated at each piston position by the annulus cross section area.

The exact fuel weight is difficult to establish, but once a starting point needle profile is chosen (for instance by the car manufacturer after extensive testing), then for tuning purposes all that is required is a reliable method to quantify the difference between the original profile and any new proposed new needle. At any position, obviously, a wider needle will give a weaker fuel/air mixture and a narrower one a richer mix.

The following is a method to quantify an increased or decreased fuel mixture richness from a needle change to aid in the selection of a needle for carburettor tuning purposes, to provide improved power or economy.

Essentially it compares the annular area between a proposed needle diameter and the 90 thou jet internal diameter with some standard needle area. The rate of fuel flow through the carburettor is directly related to this area at each of the carburettor piston positions (P1 to P9 for the H2 SU carb).

I have dubbed the area, calculated below, as an “area index” A. For example, for the standard TC needle (ES) at the mid-needle position (P5);

From the Burlen Fuel Systems SU Needle Profile Charts (see Ref. 1 at the end of the article), the ES needle diameter at position P5 = 0.0770 inches.

Calculation steps

(1) Just use the three significant digits 770.

(2) Square this value 770 x 770 = 592,900.

(3) Take again the first three significant digits 593 (rounded).

(4) Do the same with the jet diameter = 0.0900 inches, thus 900 squared = 900 x 900 = 810,000.

(5) Extract the first three significant digits, as before, 810.

(6) The area index for the ES needle at position P5 is then:

Aes @ P5 =810 – 593 = 217. The figure of 217 is proportional to the actual annular surface area, and thus the fuel flow rate, for this needle at P5.

(7) Now repeat the steps 1 to 6 for a proposed alternative needle (say) EU at the same position P5 = 0.0785 inches:

Thus Aeu @ P5 = 785 X 785 /1000 = 810 – 616 = 194. As before, this figure is proportional to the actual annular surface area and the fuel flow rate.

(8) Comparing the two needles at P5, ES is richer than EU by 217/194 = 1.119 as an area ratio (AR), or as a percentage 11.9%. Alternatively stated, EU is leaner than ES by 11.9%.

This whole procedure may seem tedious when done for all the nine effective needle positions but it has the advantage of precision, It is reliable and more accurate than simply comparing needle diameters by eye from the Burlen needle tables which are related to the fuel flow rate by an awkward square root function.

The method is not that onerous once one has become used to extracting the three significant digits from the needle profiles and calculating the squares. The jet area figure is constant of course at 810 for all H2 carburettor needle positions and needle profiles.

For the ES/EU comparison the area indices and rich/lean ratios and percentages are as follows:

Click table for bigger version

Section 2 – Piston Weight

To meet Don Jackson’s call for a leaner fuel/air mix for the MG TC with the ES needle, in theory a lighter piston spring could weaken the mixture, but the standard blue spring is the lightest available. However, by returning to the origins of the SU carburettor, a weighted piston, as at first designed, could be employed instead (see below). A weighted piston could be used on its own or in combination with a needle change if required.

The standard blue piston spring for the TC and TD exerts the following down force on the piston at each needle position:

Added to the weight of the aluminium piston and needle (on my car 107 gm) this gives a total down force (T/force g) of:

A piston with a total weight of 239g, but with no spring, would thus on average equal that of the blue spring plus an unweighted piston but slightly richer at positions 1 to 4 and leaner at positions 6 to 9.

Two colleagues have weighted pistons with no dampers or springs, one with a TA at 200g (HV3 carbs), the other a TD (H2 carbs) at 165g. These are the extremes of weights I have encountered on T-series cars. Don Jackson mentions an 81/2 oz (241g) piston but with no car specified.

On my TC I have tried various weights from 230g to 165g but I find that a total of 175g (107g piston, with 68g weight) suits my needs at present.

The effect of a change in piston weight on the carburettor richness or leanness is related to the square root of the weight ratio:-

Weight ratio (WR) = square root (heavier wt. / lighter wt.)

Thus the average weight ratio (WR) for the standard TC versus mine is :-

WR = square root (239/175) = 1.169, i.e. 16.9% rich, or expressed the other way, my TC is 16.9% lean versus standard at Position 5.

Note that the use of a different weight can provide a sensitive method to richen or weaken any particular needle if experience shows that a current one is not optimum. For example if one finds that a specific needle is slightly too lean for a particular car then a piston weight of (for example) 175g could be raised by increasing the additional weight from 68g to (say) 75g. This would then richen the carburettor at all positions by the square root of (107+75)/175 = 1.02 or 2%. This is a much smaller change than could be obtained by a different needle selection and could be used of course either way to richen or weaken accurately the carburettors as required, by the addition or subtraction of weight.

The use of a weighted piston within the range 165g to 230g (or more) makes redundant the damper fitting on the standard carburettor. The momentary richness required for acceleration is accomplished by the inertia of the heavier piston, as in the original SU design.

To digress for a moment:-

On the question of how to fix the additional weight to an aluminium piston I have avoided drilling it by using a 1.1/2 inch length of light steel tube of about 1 inch diameter. I have slit it length-wise and opened it out slightly to be a snug fit in the cylindrical cavity around the central steel bearing rod. I have drilled two small radial holes close to the tube top edge in which I use wire ties to anchor an accurately weighed strip of lead (solder would do) wound around the tube top. In my case the whole assembly weight is of course the 68g which I need.

Needle profile and piston weight changes combined

As noted above, if required, a needle profile change can be used together with a piston weight. For example, a EU needle leaner than ES with an average area ratio (AR) of 1.19 can be coupled with a 175g piston weaker than a blue spring giving a weight ratio WR = square root of (239/ 175) = 1.169. Thus the leanness ratio (EU/ES) at position 5 becomes AR x WR = 1.119 x 1.169 = 1.308. The carburettor is then weaker than standard by around 30.8% (see Table 1, below). These are settings I have used successfully on my own TC. An expected fuel consumption with these conditions would be about 1.3 x 28 mpg = 36 mpg on a standard TC.

Click table for bigger version


To sum up, this is a method to adjust the richness or leanness of the SU 1.1/4 inch diameter carburettor for the TC series, including the TA (HV3), TB, TC and TD (H2), or the 1.1/ 2 inch (H4) of the TF. The method in principle could be used for most SU carburettors including the MG Y—Type or the MGA or MGB. For the H4 carburettor the Position range is P1 to P10.

Don Jackson‘s opinion is that the standard ES needle for the T- series is “far too rich” and the EU is better. My experience supports this and I am still experimenting at the time of writing with my own TC.

Ultimately the results obtained by using the method are dependent on the practical judgement of the driver regarding the improvements in car performance and economy to be made. The average MG owner is usually sensitive to the response of the vehicle in tests on the road and can decide by experience how to get the best from the car. The tuning method I have suggested can be applied as an aid in this endeavour.


“SU carburettor needle profile charts”, Burlen Fuel Systems, Salisbury, Wilts.

“Your SU Companion, Hints and Tips for MG Owners”, Donald Jackson, Panache Press, Burnley (1994).

John Saunders

Ed’s Note: Don Jackson’s little book is available from the Octagon Car Club – it is well worth having.

2 thoughts on “Tuning the SU carburettor on the T-Series MG

  1. ANDREW SMITH says:

    Hi John,
    I am afraid that Part 2 of your article is a bit misleading. You state that the “mixture strength” is dependent on the square root of the piston weight. This is not correct. It is true as you state that the fuel flow rate through the jet, at a fixed carburetter piston position, will vary in this way but so does the air flow rate through the choke area of the carburetter. The carburettor piston always floats on the difference between atmospheric pressure and the pressure in the choke area. So a differently weighted piston will float at a different position and so the effective position of the needle will have altered and thus its effective diameter depending on the needle profile.
    (For Simon Johnston’s benefit the square root dependence of both fuel and air flow is derived from Bernoulli’s laws of fluid flow first derived in the late eighteenth century).
    The control of the combustion chemistry of the engine is best expressed by neither the fuel flow rate nor the air flow rate alone but by the ratio of the two, i.e. air/fuel ratio. For the reason why this is the best expression please ask a friendly chemist about The Law of Combining Proportions (or just think about what H2O implies). Therefore you can see that the square root dependence of both components of the mixture will cancel out for the ratio. The change in mixture strength which you have observed is thus, as I said earlier, the result of the change in the effective needle diameter as a result of the different position of the carburetter piston.

    Andrew Smith
    (Not a T-Type owner but I have had an ohc PB for fifty years and counting. My qualifications for making all the above statements are a degree in Physics and twenty years professional experience in fire and combustion.)

  2. Steve Ball says:

    Although I have stripped and overhauled many SU carbs I have almost no SU tuning experience. I offer this comment based only upon contemplation of how they work;

    The SU meters fuel according to airflow by maintaining a constant vacuum over the jet by allowing the piston to rise or fall, and simultaneously working the lovely little tapered needle in and out. So when developing an SU installation for a new (or very changed) engine the first and most basic tuning step should be to see that when the engine is at maximum power the piston JUST hits the top of its travel.

    If the piston tops out at, say, 3/4 power the needle can’t move any more and so your mixture control beyond that point is by guesswork (extra flow will still suck out extra fuel, it doesn’t go dead lean.), not needle profile.

    If the engine reaches maximum power but the carb piston is only half way up (too heavy and/or too stiff a spring) then, okay you can tune it but the bloody piston is sat blocking the carb throat and limiting your power.

    It is the combined piston weight and spring pressure that control this limiting condition.

    My TC engine has had a load of tweaks all done at once and has never been run, so the entire carb tune will be wrong.

    My first request at the rolling road will be to replace the dampers with lolly sticks marked for max piston lift.

    I’ll then have her run up to flat out and (assuming mixture isn’t miles out) check on the lolly sticks that the pistons are at max lift at or just before max power and change springs until they are. THEN I’ll know that the needles are moving, and so can be tuned, over the entire range of airflow that the engine needs.

    Incidentally, if I carefully notch my lolly sticks level with top of dashpot and then further notch every 1/8″ the nine SU needle positions then if I get a lean spot on the power curve I can read off from the sticks which needle position it is that needs enriching?

    Whilst I have completely rejetted twin Webers on a Caterham successfully on a rolling road, as mentioned above I am thus far merely an arm-chair tuner regarding SUs and would appreciate any comments on the above.

    Am I wrong to think of carb piston weight and spring as being a first setting to get the engine’s maximum airflow within the meterable range of the carb?

Leave a Reply

Your email address will not be published. Required fields are marked *