Amazon: Jet-setting

[Burdekin]

I have mine screwed right in also.

[arcturus]

Just as a matter of interest how would one test out which would be the most suitable needle for SU carbs in a modified engine? Just an academic question.

[HRA]

The simple answer is that I don't know, but the academic in me loves a "thought experiment".

I think you're going to need some sort of exhaust analyser, and some sort of remote way of adjusting the jet-height from inside the car. You'll also want a good vacuum gauge piped to just behind the throttle-plate. For safely you'll probably want a deserted private road, and an assistant or driver would help.

Start with a known needle and the jet set at a known height. Drive the car at a steady set speed (say 50mph in top gear) and look at the gas analyser. If the engine is running rich or lean remotely adjust the jet up or down to get the ideal mixture. Make a note of exactly how much the jet had to be raised or lowered. Make a note of the vacuum reading at the same time. All the time keep the driving conditions constant. Repeat the exercise for other speeds and conditions.

The vacuum should give you a good indication of the piston height at each condition and therefore the needle height. The amount you had to remotely adjust the jet will then lead you to the needle diameter you actually needed under those conditions (since you know the jet profile you started with). You've then got the information you need to specify your own needle profile.

My guess is that this will give you a needle that returns great mpg at all cruising speeds but might not have the best acceleration characteristics.

So, driving with your new needle, if you find the performance lacklustre during, say, moderate acceleration in 2nd, make a note of the vacuum gauge reading. You'll then know that at that needle-height you want a richer mix. Adjust the jet down until you're happy with that bit of acceleration in 2nd. You'll probably want several attempts at this since you won't be in that acceleration condition for very long.

Now you've got a compromise on your hands. Make the needle thinner at that point and you'll get great acceleration in 2nd, but if that same part of the needle comes into play in another gear at cruising speed you'll have lost some of your mpg.

Of course you could do all of this on a rolling-road. It would be easier at the cruising speeds, but maybe more difficult to decide if you're happy with the feel of the acceleration.

I expect that if I were ever to get the opportunity to try this I'd soon find that things aren't quite so simple and that there's much more of a "black art" involved. But alone on a desert island with a workshop, a car and my own private race-track that's where I'd think about starting. Then when that didn't work I'd start marinising the engine!

[Ron Kwas]

HRA;

Long hills with steady, unvarying inclines present a good load under which to conduct engine performance /carb adjustment tests...

I'd rather have a Wideband O2 Exhaust Sensor (but those are still pricey) than a Vacuum gauge (my foot gives me a pretty good idea of the level of load I'm putting the engine under).

Cheers

[HRA]

Ron Kwas,

I agree about the long hill, but I get too impatient driving back down again for the next run. In another life I used to run temperature-gradient tests in a lab. The 20 minute test itself was interesting, but the 45 minute wait for the oven to cool down again for the next run was always a killer!

I was thinking vacuum gauge leading back to the carburettor because it's low pressure that causes the SU or Stromberg needle to rise. A vacuum gauge should be a good analogue way to measure needle height without actually looking. But a wide-band O2 exhaust sensor does sound like a good bit of kit to measure whether the set-up is running rich or lean.

The only trouble with these thought-experiments is that as you develop them you end-up re-inventing the modern computer-controlled, multi-sensor, fuel-injected car engine. And then (for me at least) the "soul" goes out of the job and you might as well fit a modern engine...

[Ron Kwas]

HRA;

I've run plenty of environmental tests on equipment myself...

...but simply optimizing a carb set-up using the modern tools available to us doesn't by a long-shot mean that you "might as well fit a modern engine"...I'm not quite there yet!

Here's an idea: Monitor Dashpot position with an LVDT, and plot that against the Wideband O2 Sensor...and Viola, you have a closed-loop carb monitoring set-up to use during test drives (or on a dyne) with which to optimize Metering Needle profiles...!

Cheers