Carb Tuning (balancing/mixture)

[Dr-G]

All, I've not tuned carburettors for a very long time, but tonight had a go at the E-Type's triple SUs.

I started by making sure the butterflies were fully shut at idle, and adjusting the idle circuit screws in conjunction with a flowmeter. Not much problem there.

Next I visually adjusted the split throttle linkages so that they all opened simultaneously. However, on holding the throttle open at about 1500 rpm and re-checking the flow, it was all over the place. I've almost got this configuration balanced now, by fine-adjusting the lever clamps by trial-and-error. The split in the linkages ensures that they are all fully closed again once the throttle is set back to idle.

Is there an easier way to do this part-throttle flow balancing, apart from assuming it's a visually simultaneous initial movement of the throttles? Doesn't seem very accurate, indeed I've proved it's anything but. There's no mention of part-throttle balancing in the manual, which is odd, becasue I'd have thought balance under throttle is as, if not more, important that at idle.

I'm also finding that when using a Colourtune, there's a huge range of adjustment of the mixture screws before there's any perceptible change in combustion colour. Apart from sticking with the book jet lowering turns and checking plug colour, is there a trick to doing this accurately?

Thanks.

[christopher storey]

You do not tell us whether this is a 3.8 or a 4.2, and there are significant differences between them because the 4.2 manifold has, relatively speaking, very large balance passages compared with the 3.8. Thus in a 4.2 altering either the flow or the mixture strength on any one carburetter has a not inconsiderable influence on the flow and/or mixture of the other cylinders . The effect is a lesser one on 3.8 cars, but is still present. It also must be understood that it is not just the jet settings that govern the mixture : the float level, if even slightly out of kilter, will skew the mixture strength not only on the 2 cylinders primarily fed by any particular carb , but also to a lesser extent on adjacent cylinders

You will get on better by using a CO meter - the Gunson one is perfectly adequate - and initially setting each jet about 65 thou down from the bridge , and aiming for a CO level of about 5% at idle in each pipe . When setting the centre carb, it is necessary to swap the probe over between the pipes because the centre carb affects the strength in both cylinders 3 and 4 , one of which feeds into each exhaust pipe

Once you achieve approx 5%, I then pull the dashpots off and re-measure the jet depths, average the 3, and set each jet to the average depth and then re-check the CO level

[Dr-G]

Thanks Chris.

It's a 4.2.The carbs were professionally refurbished by Tony Hebden in Sheffield in about 2015. The car hasn't been run since. The throttle spindle bearings were reamed, and all the float valves replaced and re-set to the correct heights (apparently one of them was from a Morris Minor). Bottom line is I'm confident that the carbs are now in pretty good condition.

ETA I do have a Gunsons gas tester, but I always found it a bit vague.

[malcolm]

I used a color tune (Gunson) to start, got them all to the colour recommended, and it ran like rubbish.
Fact is, the E likes to run a bit rich and using Colortune puts them all too weak.

[chrisfell]

There is a very good write up on tuning triple carbs on the jag-lovers website. If I can find a link I'll post it here.

[Dr-G]

There is a very good write up on tuning triple carbs on the jag-lovers website. If I can find a link I'll post it here.

Thanks - probably the one that tells you to adjust for highest rpm on a large, high resolution tachometer?

a) I've never seen such a tachometer

and

b) determining a miniscule difference in idle rpm with an uneven idle would be very difficult on my car.

[Dr-G]

So I had another go at tuning:

1) Set the balance at idle. Not much issue there.

2) Set the jets all initially at (IIRC) 0.065" using a digital vernier depth gauge.

3) Ran, checked & adjusted CO in both pipes:

Left pipe set to stable 5% with minimal adjustment turning all 3 screws a similar amount.

Right pipe way off, rear carb needed considerable enrichment.

Eventually got stable & repeatable 5% on both pipes, sanity checked with the colourtune in front, rear and a middle plug. Balance still OK.

4) Checked jet depths - rear one was visibly much deeper - turned out to be 0.055" lower than the other two (which were within 0.007" of each other).

5) Averaged the depths and set all carbs the same & re-started. Ran like a dog, predictably with neither pipe giving 5% (more like 1.5% and 7%). Repeated the mixture adjustment to stable & repeatable 5% and re-checked, with exactly the same results: Rear carb way lower jet, front two nearly identical. Balance still OK. Idled pretty well, even exhaust beats.

Checks:

A) Float forks - all spot-on.

B) Float dimensions - ditto.

C) Float buoyancy - identical.

D) Needle type - correct (all measured at root, middle and tip with calipers; identical).

E) Needle protrusion - all identical.

So...after several hours and a few litres of fuel, it runs OK, but there's clearly a significant issue with the rear carb (or cylinders). Any other ideas?

Thanks.

[Simonpfhc]

Following this with interest, as I’m thinking about buying a CO meter.

What I don’t understand with the process that Christopher suggested and you followed, is why, when you have both pipes running with a steady 5% CO, you then go back and change jet depths to a group median. If the output at the exhausts is correct, surely that means the jets must be set to the correct richness (depth)??
Tolerances in each carb will differ so that would explain the different result jet depths - am I missing something?

[mgcjag]

Dr G..its not easy to get a car running great unless you know that each carb is built and set up correct.......in my opinion i would start with the basics...check thoroughly for air leaks..then..remove each carb and thoroughly check it over....butterflies closing correctly..drop test good....spring ok..jets centered etc....then set them up as per the very good setup pages in tne Service manual....the car should then run good with only minor adjustments needed...iv done this more than once ........whats your name....please try to put it and your car model in the signature area...thanks..Steve

[MarekH]

What I don’t understand with the process that Christopher suggested and you followed, is why, when you have both pipes running with a steady 5% CO, you then go back and change jet depths to a group median.

The CO versus rich/lean relationship isn't quite a straight line so this approach is hoping to make the "worst" carburettor "better" more than it makes both of the better ones worse.

Following this with interest, as I’m thinking about buying a CO meter.

There are other metrics available and it's quite telling that modern engine tuners don't tune fuel injected cars by CO, but by what makes the manifold air pressure drop the most, i.e. the more vacuum you are pulling, the more efficiently your engine is running.

kind regards
Marek

[DWW]

As a point of curiosity how are your fast idle screws set?

[christopher storey]

I think, as Steve ( mgcg) has said, a check for an air leak on the rear carb/manifold and manifold to head is needed. A convenient way to do this is, with the engine idling, to play an unlit butane gun along the various joints : any variation in idle which this produces suggests a leak at that spot . The big difference in jet depths is highly abnormal

[Dr-G]

So this morning I thought I'd re-check the carbs. After warming the engine and CO meter up and getting a steady 2% after 30 minutes or so, I put the probe in, and got opposite results form yesterday...(front carb needed significant enrichement, and the rear was way too rich).

I re-adjusted to get both pipes at 5% again and then measured the jet depths. This time, the front jet was significantly lower than the other two (yesterday it was the rear one). Interestingly the rear two were both about 0.060" (similar to yesterday's front and middle optimum settings). While the carbs were being reassembled, I checked the piston drops and resistance, and all seemed well.

Re-checking with the CO meter basically showed that the results were very inconsistent, and required frequent re-calibration to check the baseline hadn't moved before any meaningful readings could be taken. It all got to be a bit of a joke, so I've put the Gastester under the "novelty toy" category, since it appears incapable of giving consistent results, which obviously makes relying on it for repeatable readings and adjustment impossible.

Anyway, I plucked up the courage to actually go out on the road again (first time in 13 years), and it seemed to run very well - pulled cleanly in all gears and a fairly consistent and smooth idle. I only did a couple of miles, but on returing home, I checked the plugs, and sure enough the front one was sooty black. I weakened the front carb by 1/4 turn, and the idle quality immediately worsened. Same as yesterday when all three were set the same, only this time I was adjusting the front carb to be weaker, yesterday it was the rear.

So...some progress, but I still have little understanding of what's going on with the jet settings. Combining yesterday's and today's jet depths which gave good results, 0.060" seems like a good average, but as I say, setting all three to that depth results in rough idling.

[Gfhug]

DrG, as suggested by Steve (mgcjag) read the manual and remove all carbs, check they close properly and then set them up properly. If your car has not run for so long then there’s a very good chance the rubbers in the carbs, etc. have perished and not working correctly. It is only by going back to a correct set up of distributor, carbs, etc. that you’ll get the car running well. Stop messing about and go back to basics, PLEASE!!

And has been requested more than once, add your name and car type to your signature area. Don’t wear out the patience of people trying to help you.

Geoff

[bitsobrits]

I know you indicated the carbs were rebuilt by a reputable person, but this sort of variability points to a variable air leak condition, such as at the carb spindles, or at the carb base gaskets/spacers. Are you able to get it to idle in the 500-600 rpm range? If not, again evidence of an air leak somewhere. You might also temporarily remove and block the vacuum port to the brake booster to eliminate that vacuum circuit as a variable.

[Simonpfhc]

The CO versus rich/lean relationship isn't quite a straight line so this approach is hoping to make the "worst" carburettor "better" more than it makes both of the better ones worse.

Thanks for the explanation Marek, makes sense.

[MarekH]

If you want to borrow a portable lambda sensor and gauge, I can send you one.
kind regards
Marek

[Simonpfhc]

If you want to borrow a portable lambda sensor and gauge, I can send you one.
kind regards
Marek

Generous offer Marek, thanks. I’ll send you a PM.

[PierreW]

If you want to borrow a portable lambda sensor and gauge, I can send you one.
kind regards
Marek

Hi Marek , just asking : where do you install the lambda sensor ?
Regards
Pierre

[MarekH]

It measures how much oxygen is in the exhaust gases, so the sensor has to go inside the exhaust, away from the influence of outside air. For people who have a permanent setup, a threaded bung is welded onto the exhaust and the sensor is screwed in not less than about one metre from the exhaust manifold. For temporary installations, e.g. for an MOT inspection, it is simply shoved as far up the end of the tailpipe as possible.

kind regards
Marek