CIS FD balancing flow to injectors

[Walt Fricke]

Here are the milliliters of fuel coming out of my six injectors when I put plastic water bottles under the injectors, which were pulled out of the engine:

11
10
10.5
10
11
9.5

Biggest difference from 11 is 86%. Smallest is 95%.

Assuming I ought to get these closer together, am I right that screwing the individual adjusters in (right or clockwise) will raise the flow for that injector?

Any estimate of how much of a turn on the screw is required to make how much flow change?

Getting the #6 injector into the bottle is a PITA.

[77Targa0912]

I've read that getting them within a 20% range of each other is the best you can do.

One cis injector expect said out of a 100 injectors you might only find a handful that flow exactly the same. So that can get expensive really quick.

[Walt Fricke]

Well, as to the injectors, other than cleaning off deposits with chemicals, and backflushing the screen on the inlet side, I don't think there is anything you can do to match the injectors themselves. They aren't like electronic ones (maybe those too are only matched by picking ones which match out of a batch).

But you can control the fuel flow from the fuel distributor - there is a little screw under a cover screw for each of the six injector outlets. I've just never messed with these, but others have to get the output - distributor plus injector - to come out closer than when they started.

[fanaudical]

Walt - Where are those adjust screws located? Do you have a picture or diagram? I don't recall ever seeing those.

[E Sully]

Walt, On the original style from my '73.5, there are shims under the springs and top hats that would have to be changed for this. I have never taken apart a "Modern" fuel distributor, but it would seem that might work. As long as you keep track of the changes with these screws, it seems to be worth trying.

[tabasco]

Fanaudical : the screws are on top of the FD, next to the fuel lines going to the injectors.
The early FD don't have those adjusters. It started with the SC maybe ?

[Walt Fricke]

The cover screws are removed with a 3mm Allen. The screws under them appear (in diagrams) to press on one side of a cup the top of the springs inside rest on. I will have to put my bottles back on, and test after turning one of the screws.

[RarlyL8]

The individual adjusters work on a percentage of the total. If the total fuel flow from all 6 injectors adds up to 100cc and you increase one injector by 10% the result will be an increase of 1.67cc for that cylinder and a reduction of 0.34cc for each of the remaining 5 cylinders. Keep this in mind when altering the adjustment.

[fanaudical]

Got it - thanks. I have a '75 and have no such adjustments on my FD (have to use the shim method, which I'm familiar with).

[Walt Fricke]

The actual adjusters also use the same 3mm allen. I tweaked one a hair. But this information about how they interact sounds like it makes balancing things a bit harder. My plan was to tighten the adjusters where the volume was below 11mm (which is just an arbitrary first test value based on how long I lifted the plate) to tweak them up to 11 mm. I guess I can figure out how to do this. As long as they are all pretty close to equal at the end that should be good, as it is not the absolute flow you care about here, but the balance. So if a 10mm line comes up, and a 11mm line goes down, that ought to get them closer?

The absolute flow you can adjust with the single plunger/slit cylinder adjustment.

[RarlyL8]

Yes it can be a tricky procedure so take it slow. The example I gave is simplistic and only holds if small changes are made. If all the adjusters are opened up the total volume/capacity does increase.
Optimally one would tune each cylinder via AFR and CHT. Non-equal length intake runners and unequal cylinder head cooling lead to cylinders that are lean or rich. Using AFR and CHT to tune would smooth this out. I know, not practical but a consideration given you may tinker with the fuel flow.

[Walt Fricke]

Well, tweaked 2, 4, and 6 down (screwed in) a hair, 6 tried to get a hair more as lowest flow.

New results, with doubled fuel squirted on notion makes measurement more accurate:

21.5
24
22
22
23
21.5

So highest changed from 1 and 5 to 2.
Lowest went from 6 to 1 and 6.

Spread narrowed from 13.6% to 10.4%

Will see if another tweak will reduce the spread a little.

[Walt Fricke]

Tweaking 2 up and one down gave this:

28.5
29
29
28.5
30
29.5

So the max spread is now down to 5%. The ones I adjusted to increase did, and the one I turned a bit the other way went down a bit, so at least those qualitatively moved in the right directions. But the ones not adjusted moved up in the hierarchy. If I had any faith in my ability to turn a small screw any less than the hair I turned, I suppose I could raise the two which are5%low. The others would drop a bit, and they'd all come close to meeting in the middle. But I don't have confidence I can just nudge those.

I put all these in a spread sheet, which does the percentages for me keyed off whichever flowed the most. That way I am not just guessing.

Maybe time to call it good?

[T77911S]

i just did this on my 930 over the week end.
me, i would only adjust one at a time and i would have only adjusted #6.

also, you need to do multiple tests and the more volume of fuel the better the results. i have a small postal scale and i was weighing mine. i was testing at around 6oz's. just workied out that way.
you also need to do multiple tests before making adustmets.
i did one that represented idle. i stuck a tie rap between the plate and bowl, put a battery charger on the batt, and ran the pumps for about an hour. it took that long to get 6 oz's.
(i have looked at my sensor plate at idle so i new about how far to open it).

next test was WOT.
i was looking to see if any one cyclinder was lean both tests or rich both tests. #2 was lean.

next i swapped injectors. it balanced things a little better but #2 was still lean.
i also did an idle test with no injectors. i should have done one mid throttle or WOT but i did not.

i tweaked #2 just a tad. it does not take much. what ever you think is enough, turn it half that.

then i did 2 tests with the plate open about 3/4 of an inch. they were much better. i did not do another idle test but i should have.

i was wondering if they interacted as brian said and i can see how they would because i looked like the others changed a little also.

i did not do anymore idle tests. they took too long. this was an all day ordeal and i was tired of messing with gas.

here is how i would do it differently and this is because the more tests, the better you (or someone else) will know what to do.

do 3 tests every time.
one at or near idle, one at part throttle and one at WOT.
i would do these at least twice as a base line before making ANY adjustments.

do the same tests with no injectors. this may be helpfull or it may not. i only did the one and my results from lean to rich were not nearly in the saem order as with injectors.

next i would swap injectors around. i labeled my results 1-6 for lean to rich. then i swapped 1-6, 2-5 and 3-4.

this is not something i suggest anyone do and do it with great caution. i did it because my #2 was consistantly low and that is the only one i adjusted. i am also close to sending my FD in for a rebuild. my idle DID improve but i still have other problems.

[Walt Fricke]

I started out using six test tubes, which only held about 15ccs. I squirted into plastic drinking water bottles, and poured into the test tubes. Measuring height was too difficult, so I bought a 25mm tube with stand, graduated every 1/2mm (took me a while to realize each line wasn't one mm, it was a half). But allowed me to squirt more. I got overconfident and squirted a bit more than 25mm. The tube held more, but had no gradations up there. So I'd pour some back into the plastic bottle, measure, pour out into my collection jug, and measure what was left. Two measurements means two opportunities to mess up reading the meniscus. And the car is in a sloping driveway, so I'd eyeball the tube to try and get it level.

However, I couldn't resist one final tweak of two lines. I ended up with four lines exactly equal, and two lines 1.5% low. Called it good. The car had given no signs of rough running due to small variations in output. I was chasing a more significant issue, but decided to take the time to go through about everything.

I think that it is better to test with injectors in, and to leave them in and adjust. The alternative would be to rotate the injectors around at least one step, if not more. That ought to give you some idea of how close to each other the injectors are. But leaving them in place is going to allow you to balance the combination of the flow from the FD and from the injector.

Ideally, all injectors would be balanced, as would all of these upper chamber flows. That way you could replace any injector with one of equal flow without having to do all this again. But you'd need a lot of injectors to find one that was an exact match, I think.

To top it off, I could not get the #3 screw cover off - it is back there and hard to see if you have your wrench coaxial, and I got the feeling I was starting to strip the inhex. If the engine was out I'd have tried harder, and used a variety of tricks to deal with this. But it wasn't. Luckily, I got the results I did without having to adjust #3, so that was to the good.

A friend had done this using graduated baby bottles, or so he told me. I couldn't find any, though if they came in 1/2 CC graduations, and were long enough so that the injector couldn't spray fuel back out (my test tubes would have done just that), that would simplify and speed things up. Squirt fuel until the bottles were pretty full, withdraw then and write down how much they held, pour out the gas, and you are done. If you thought your whole system was clean, the gas can just go right back into the tank. I didn't think that, and was squirting injector cleaner at times as well as gas, and didn't trust that the gas didn't contain some deposits which were being flushed out, especially at first.

I do like the instruction given: turn about half of what you think is as little as you can.

[T77911S]

you really dont need to know "numbers" when doing this. IE, you could take a clear jar, put tape on the side going from top to bottom. pour one bottle in and mark the tape #1. do the same for the rest.
you really just need a way to compare each cyclinder. i happen to have a small postal scale i use.

this is all a learning curve, maybe an expensive one. even thought i tried to be as careful as i could and get accurate results, there are things i would do differently next time.
for me the big problem was dealing with the metal lines and getting something in there to catch the fuel. this made it very frustrating and took away from were my attention needed to be.

one other thing. always make sure the system is primed before doing a test. IE, do one test and just pour it out. if you have had injectors off, this will get fuel back in the lines and injectors.

[Walt Fricke]

Thin plastic water bottles removed most of the frustration of getting the injectors into containers. Only #6 on my NA motor called for occasional name calling, and I had to scrunch the bottle a whole lot.

Measuring in a graduated tube worked fine. If I do this again, I'll try to find a larger capacity graduated tube.

I did purge/prime the system after first checking volumes with no injectors. Part of running injector cleaner through the FD.

[PcarPhil]

Originally Posted by RarlyL8

Optimally one would tune each cylinder via AFR and CHT. Non-equal length intake runners and unequal cylinder head cooling lead to cylinders that are lean or rich. Using AFR and CHT to tune would smooth this out.

Yep.

Getting the injectors to flow the exact same quantities is a nice place to start but with variations in the intake, cylinder head cooling, and exhaust I feel matching individual cylinder AFRs and CHTs is the way to go. You may find that you'll be adjusting the FD again to get better matched AFRs and CHTs which I feel is far more important than trying to get the fuel injector flow to match exactly!

An infrared thermometer aimed at the individual exhaust pipes near the exhaust flanges helps with this.

Just my $0.02!

[T77911S]

you do this at idle?

if temp is too hi do you richen or lean?

[PcarPhil]

Originally Posted by T77911S

you do this at idle?

Yes, idle is a very good place to start. Ideally you'll be checking AFRs/CHTs/EGTs at whatever your target RPMs are. For instance this could be at cruise and/or at WOT. Of course this can be difficult to do but there are ways to do it (checking temps on a dyno, installing CHT and EGT sensors, etc.)

Originally Posted by T77911S

if temp is too hi do you richen or lean?

Richen. (EDIT: Depending on which side of stoich the AFR is already on)

Another method of equalizing cylinder AFRs is to read your spark plugs. If you can get all 6 looking exactly the same you're on the right track!