Modifiying SU carburettors

(Robin O'Connor) #1

A few years ago I seem to recall a production that covered modifiying SU carbs to flow better. I think it was by David Vizard. Does any body have a copy of the book and what its’ title is?
I seem to remember he smoothed out the bridge area and a radius on the pistons’ leading edge?

(Paul Wigton) #2

It can be done, but the results are only really noticable at high rpm, and at WOT.

Overall, for street use, stock, properly-set-up SUs work fine.

(Phil.Dobson) #3

The SU Carburettor High Performance Manual (SpeedPro series) by Des Hammill.
However I am with Paul on this. Unless you into the extremes of performance you will not get any gains from trying to modify the standard set up.

(Keith. P. series two roadster,,,UK) #4

You may notice that on full throttle, any radius work on the pistons disappears as the piston raises to its full height, so no advantage there, the advantages are at part throttle when the rear as well as the front of the piston is radiused, the airflow increases, atomisation is better, look at a section through an HD 8 to see how forming a wedge with the rear radius obviously increases the “suck”…readings with airflow meter before and after confirm this…

(Paul Wigton) #5

Indeed, and there is a good chance overall drivability will be negatively affected.

This reminds me of the story, where at some car club meeting, Ed Iskenderian was the featured guest.

A question posed to him was, “What’s the best cam I can use for best power, drivability and mileage in my ______?”

Ed, with a deadpan delivery, answered, “The one it came with, from the factory.”

(Attributed to our Mike Moore.)


(Roger McWilliams) #6

David Vizard has written numerous engine tuning books. You may be recalling his “Tuning BL’s A-Series Engine” or “Tuning the A-Series Engine” or “Tuning Standard Triumphs over 1300 cc” (which does discuss SU carbs in comparison to Stromberg).

(Robin O'Connor) #7

Thanks guys, those would be the books, unavailable new now by the looks of it, slightly conflicting thoughts, maybe i’ll Radius one carb and leave the other standard :slight_smile: :slight_smile: :):star_struck:

(Lovell) #8

Sort of surprised at this one.

SU’s have bett flow characteristics than Webers.

They also only have 4 moving parts.

Didn’t realize there was anything else that could be improved upon on those?

1 Like
(tony) #9

Hammill ( & others) make the point most vehicles come from the factory made for road use,
and therefore fuel economy, general response, and even quietness are compromise design factors.

He has advice for some things that can be done, which include changing piston springs and oil, minimising throttle cross-section, verify everything equal etc, but really only wants doing if the head is worked, and will definitely use more fuel

His chapter on Racing mods, he states clearly is not for road use

surprised when VW said they could reflash their computers in such a way that there would be no loss of power, or fuel economy, or increase in emissions

(Keith. P. series two roadster,,,UK) #10

Right, and if you’re happy with the basic horsepower thats fine, Hammil did the porting on my head, and the engine has been opened up to 4.5 litre, I found a very rich stumble with the UO needles on a 750 rpm tickover, increasing the air speed across the jet solved the problem with the pistons radiused on both sides, my engine will now tickover evenly at 500 rpm, and the pick up on hard acceleration from stationary is phenominal, I get 18/20 MPG {Imperial gallon}. 14.5 quarter with a 2.88 diff…

(Paul Wigton) #11

I got 18/20 mp US-gallon, on Tweety…dead stock carbies/engine.

Just a data point!

(tony) #12

not against more HP at all, so long as the car is completely useable for normal use, including smooth idle
sounds like u got an excellent engine…is it just bored out to 4.5, or stroked ?

will have a look at Hammills books again, but dont think he actually discusses radiusing the SU pistons ?

(Robert Wilkinson) #13

I replaced my factory UM needles with UO back when I had pancake air filters. I know they are almost certainy too rich for my tired unmodified engine (now with Mark X air box), but I’ve been too cheap to buy a set of UB or UE. But it was my understanding that the first few stations are about the same on all of these needles, so characteristics of tickover shouldn’t depend on needle type? If I’m wrong, perhaps I’ll swap back three of my nine brand new UMs.

(Robin O'Connor) #14

I have Hamills book and there is very little on modifications only setting up.

(Keith. P. series two roadster,,,UK) #15

Needles aren’t so expensive as a jar full of Weber jets !! could not seem to get a 700 rpm tickover with the engine after the port work,I had dropped a needle size but had to be the UO for full throttle with an ITG filter…Hamill was helping out locally at the time, and suggested this piston mod at about the same time his SU book was coming out, he had them radiused for me, we had set the uni sync sometime before, think it was set on ten, and as best I can remember, airflow increased the reading by a couple of points, this mod seems to tick all the boxes when it comes to airflow, losing the sharp edges, no detrimental effects on slow running or pick up, plugs run a good colour…this is a series three XJ head and block, 110 crank throw, viscous damper and Rob beers valve work and cams…sorry Wiggles, but I find it impossible just to drive bog standard, I once had 24 mpg keeping the car at around 60 when unmodified, and once, a reading of 25.53 on one ten imp gallon tank driving a 4.2 XJ down to southern Germany at a constant 60 behind a pair of MAN trucks…have a look at the slide on an Amal carb , cut to a wedge on the inlet side, they know what they are doing !! Worn mains can cause a rich tickover by the way, have new mains in a kit I bought but waiting for the right time to swop them…they have about 40 thousand miles on them…

(Paul Wigton) #16

FWIW: at a constant 65 mph, back on the ‘11 SOL, Tweety got ~26 mpUSg (3.8l), bog standard. That’s somewhere close to 28 mpg, Imperial.

There are many, MANY variables here, and am pleased you saw improvements on what appears to be a non-standard engine.

To repeat: on a bog standard XK engine, standard, well-setup SUs work quite well. Ceteris paribus, I think mods to the carbs will not necessarily bring about noticeably significant better performance.

My experience: when I was younger and more motivated—and had spare HD8s (same as the Jag) with which to work—I took my then Rover, did about 500 miles of baseline measurements, then installed a pair of carbs, modified similarly to what you reference.

Before mileage = ~25 mpg, combined: afterwards, ~26 mpg. Driveability and acceleration times were a wash.

Was it work the many hours of work? Nope.

Again, you have a modified engine, and clearly you’ve gotten differing and apparently positive results…and that is great!

I just suspect that, on a standard XK engine, massive and not-easy-to-do-properly carb mods may not work as well as how the extensive factory-tested carbs did.

In this particular case, Ill toss in with the factory engineers!

(Paul Wigton) #17

Ive seen it…:crazy_face:

I also think the engineers at Jag and Skinner’s Union were pretty sharp cookies, and—again, on a STOCK XK—had a wedged front of the piston made a significant difference in power and fuel mileage, our Jags would’ve had them.

The wildest mod I ever saw on an SU was on a friend’s H Production Spridget: the pistons were modified such that, at WOT, the bottom of the piston was machined so that it formed a perfect cylinder from front to back of the carb body. The butterflys were silver soldered to the shafts, and profiled so they barely presented much impediment to flow.

The car didn’t idle below 2000, and it was an “on-off’ switch, but, jeebus, that lil A Series engine (motor?..:stuck_out_tongue_closed_eyes:) screamed!!

(Keith. P. series two roadster,,,UK) #18

Well, if we didn’t experiment, we’d still be cracking nuts on rocks !! I try to go where a situation is reversible if it doesn’t work, as a twenty something year old, I would haunt the premises of Messrs Whitehead, Hawthorn, Coombs and Duncan Hamiltons place, and ask the same question, how can I make my Mk2 faster, the answer was always the same, “they are fast enough”…so, find a second hand Coombs, which I did, which proved to me that without turbos or superchargers, by subtle work here and there, help the breathing, change the cam profile and whatever else they did, you could get a 60s six banger car that would go straight to 140mph, indicated, and still be road tractable…that was what I was looking for…was there ever a spec sheet for a John Coombs engine ?? not that I know of, was it all in his head? I don’t know…I was on the phone for a hour to a guy a lot older than me a few nights ago who turned me out a set of pistons twenty odd years ago that lasted for ten years until head gasket failure, he rebuilds Bristol radials and makes the piston rings for most of the Spitfires that are in the air, he also made the pistons for the military 4.2 Jaguar engines, 4000 of them, but all we talked about was carburation…we discussed the three valve torsion actuated valve system on Archie Butterworths intended Formula ! engine, there is an ex Jaguar man in this group who knows the name, heard the name dropped on Nevilles day at Curborough, anyway, like you, he says, at full throttle, the piston is wide open , what more do you want?? but, as my mod seems to work ok, I’ll stick with it, …A series engines always amazed me with their lack of displacement compared to the performance you could get out of them …

1 Like
(Terry Sturgeon) #19

The modifications discussed here to the piston make little sense when assessed against the principles under which SU’s operate. They are called “constant depression” carbs for a reason. Bernoulli’s principle dictates that air flowing through a venturi, which is what the bridge to piston bottom is, accelerates to get through and creates reduced air pressure or depression at the point of greatest restriction. In an SU or indeed any carb it’s this depression that raises the fuel into the air flow. If you increase air flow (air speeds up ) without increasing the size of the venturi the depression increases and more fuel flows. American carbs/Webers function this way. SU’s don’t, and in fact can’t. The bell over the piston is connected directly to the depression in the manifold which is regulated by the throttle plate. The piston rises in direct proportion to the increase in the vacuum caused by increasing the opening of the throttle. More air flows in but the venturi is enlarged by the piston rising and maintains a constant air speed through the venturi with a constant depression over the jet. The needle narrows as the piston rises and more fuel flows. You can install heavier springs, and heavier oil to slow the pistons rise, thus enriching the mixture as you accelerate, but the force created by the depression in the bell is very high and quickly overcomes these measures. In other words modifications to increase airflow at the bridge at anything less than full throttle (where the piston physically cannot rise any further) are ultimately futile. If more air flows over the bridge at a given piston height because of the modifications it will pull more fuel, but if you want more fuel it’s easier to just install a richer needle. At partial throttle the carb flows all the air it can while still maintaining a depression in the venturi to raise fuel. The bridge/piston is not the impediment to air flow, the partially closed throttle plate is. Will “wedging” the piston help air flow in a meaningful way? Maybe but to what end? First you don’t want (need) more air at partial throttle, and at full throttle the bottom of the piston is tucked inside the bell opening. The carb throat only “sees” one valve open at a time, and stock and “big” valves are smaller than the throat. Does wedging help fuel atomization - don’t know, don’t really see how. Maybe it induces more turbulence in the air? Carbs don’t atomize fuel very well - any improvement would be very welcome.

(Robert Wilkinson) #20

I think you’re probably right to the “first order.” But fuel/air flow and atomization are so complicated that I don’t think that any modification can be ruled out by theory alone.

A related issue is that of pumping losses. One could argue, I suppose, that modifying the piston might improve air flow in such a way that pumping losses are reduced despite (or independent of) the constant depression principle.