Has anyone ever witnessed a situation where there is an apparent imbalance in Air/Fuel Ratio (AFR) between cylinders in an engine?

I have begun the tuning process of my engine – see link for history:

https://www.thirdgen.org/forums/diy-prom/401124-street-rodder-iroc-350-a.html


As noted in the link, my situation is a rich idle due to a new cam, aftermarket prom, and other significant modifications.

I have begun tuning. I had bought an infrared thermometer to closely monitor exhaust header temperatures to make sure I was not running things too lean. Upon first round of tweaking, I ran the engine for approximately 10 minutes – long enough to enter closed loop mode. I monitored header temperature and noticed it roughly stabilized to 275F. I then decided to check all header pipes to ensure I was getting a good average. To my dismay, I found 6 pipes were running ~275F, however two pipes were running 500F+.

Several years ago when I first completed the engine rebuild and fired up the engine for the first time, after all of the initial ‘burning smells’ of a new engine, two pipes on my new ceramic silver header pipes burned excessively hot to where the ceramic coating actually turned a dull grey. Ironically it was the same two pipes that presently were running hot.

Intuitively, one would think this temperature imbalance may be caused by a lean AFR on these two cylinders, which logically could be associated with weak injectors. Since the time of the original ‘burn in’ and the recent run, I have removed all injectors and have had them ‘professionally’ cleaned for a second time at S&P. I was told that the injectors were functioning properly. If the injectors were truly good, then the next hypothesis may be that the wiring or connections to these injectors might have excessive resistance which may be causing weak injectors, leading to a lean AFR in these two cylinders.

One thought I had was to come up with an inline injector connection jumper scheme which would allow me to monitor current to each injector, which in turn would clue me in on any potential electrical related issues…

I considered potential valve issues, however I kind of reasoned these issues out…

Any suggestions, thoughts, or recommendations?

Regards,
JAR_Tudor

 

Questions to start out with:

1) are your header tube temperatures different only at idle? If you hold the engine at say 2000rpm for a minute or two, do the tube temps even out?

2) any physical clues in regards to which two header tubes are the hottest (i.e. are they the last two on the fuel rail? closest two to the throttle body? closest two to your power brake booster fitting or other vacuum fitting?)

I'm presuming those two cylinders have a good hot spark, the plug wires are on the cap in the right order (firing them 90* late and sending the burn out the exhaust would certainly make them run hot!), and stuff like that has been checked.

if the problem only occurs at idle, then it's probably an idle air distribution issue. If you search for "idle air distribution" here on DIY PROM there is some good reading.

ultm8z battled similiar issues for a few years and found all sorts of contributing issues along the way, including a bad ground. Other issues recently mentioned include wrong plugs (too short for the plug hole in the head). basically anything which could delay the flame kernel becoming established and firing the chamber off could send out still-burning gases into the exhaust.

Also, what are your IAC steps at idle? you want to try and minimize them as much as possible.

Finally, there have been some mentions of idle air distribution problems with early TPI intakes, and that GM made a casting change. I can't find anything on it at the moment, but it shouldn't affect an '89 vintage intake I wouldn't think. perhaps someone else can jump in on that one.

 

First off, EGTs drop from being rich of lean of Stoic..

If you want an well balanced engine, then you want 8 ecms, with an IR intake system, all running closed looped.

Getting 8 cylinders to contribute evenly is one of the challenges of engine.

Hook a scope to a WB, and you can see slight variances with most any engine.

 

I ended up doing an elaborate idle-only IAC manifold (similar to LT1's cast-in IAC passages) in an attempt to alleviate much of these A/F distribution problems on my Miniram.

See this thread for details.

https://www.thirdgen.org/forums/diy-...ghlight=ULTM8Z

I too did all of the same kind of diagnostics, searches on this site, reading and more reading, asking so many questions that it's a wonder these board members didn't get sick of me and ban me!!

While this modification did drastically improve the idle speed A/F results data (i.e., left-to-right BLM's, exhaust tube temps), the idle is still rougher than I would expect it to be for my cam profile. The jury is still out on what could be causing it, but my latest round of diagnostic data is still pointing to cylinder #2 as having some sort of issue.

But, 91L98Z28 is right about the ground issue. Prior to this manifold mod, I was battling a very severe misfire on the #2 cylinder. Part of the problem as I discovered was that I had my main engine ground connected to the serpentine bracket rather than the cylinder head. It was producing a nice little light show when I viewed the running engine in pitch darkness. The spark energy was having a tough time finding it's way back to the battery ground and therefore much of it was arcing out of the plug wires to whatever grounded metal was nearby. That fact, and a discussion with an electrical engineer friend of mine who is also a hot rodder prompted me to move the ground to the cylinder head, which cured about 75% of the misfire. I'd say the manifold mod cured another 15% of the residual rough idle. The last 10% is what I'm after. However, 100% cured IMO doesn't mean idling like glass though, but rather it should idle very smoothly with how mild my cam is.

 

While that is true, Nothing is perfect
(Not until you tweak it for ten years anyway)... To Grumpy's comment.

I would also be concerned with such a drastic difference.
Which cylinders?
Check the valve lash on those two cylinders to be sure they are where they should be.

 

91L98Z28,

I measured temperatures primarily at idle - i'll try 2,000 RPM this weekend.

The cylinders are 4 & 6 (right/middle) - opposite side from brake booster, etc.

I will go through wiring order & check spark plugs again this weekend. Engine seems to run relatively smooth above idle so I kind of assumed this area was O.K. Probably good to verify everything one more time...

IAC steps are running in the 140's.

Soooo much to learn - jumping in barefoot with both feet at once...

Thanks,
JAR_Tudor

 

That's pretty high. I think the max is 160, so there's not much room for adjustment by the ECM in case of a stall-save condition.

I would adjust your throttle blades so that the ECM only has to use about 0 to 10 steps to hold the idle. I've found that the least amount of steps I have, the more stable the idle. I actually shoot for 0 in park which results in about 15-20 in gear.

 

A couple of other posts snuck in before I submitted...

I salvaged the wiring loom from the '89 IROC and retained grounding locations as stock (other than modifying it from MAF to SD) - I went through wiring fairly thoroughly - but like everything, there's always a chance...

Forgive me for being too naive, but is there much valve lash with hydraulic lifters?

 

AFAIK, there shouldn't be any lash with hydraulic lifters. That's for solid lifters.

The way I usually do my rocker arm adjustment is to idle the engine and thne go around each rocker, back off the adjusting nut until I just hear clicking. Then, tighten it back down until the clicking just goes away. At that point, I put 1/2 to 3/4 turn on the nut and then move on to the next rocker. It can get messy, but I have a special set of rocker covers that I cut the tops off of and I use some cardboard oil splatter shields to minimize oil splashing. That seems to do the trick.

 

But you can "over-do" them and hang them open a bit.
seeing how this is a new cam, there's a possibility they are out of adjustment.
There is only about 0.060 of adjustment to the lash before it runs out of travel. That can hold the valves open a bit when the plunger extends upward. On a 3/8-24 stud it equates to 0.042 per turn so at 3/4 turn you should be halfway.
If the lifter did not "pump-up" when the adjustment was done, theres a chance it is too tight.
Just something to check. I agree with checking the other grounds etc that were mentioned as well.
If anyone knows how to locate a miss...

 

Jump on those 140 steps of IAC first. Fully heated up (drive for 10-20 minutes), then shut all the loads off (no headlights, no radio, turn EVERYTHING off) and open up the throttle body blades until you get under 15 steps of IAC position.

That alone is going to change your airflow distribution at idle, so do that first (not to mention idle control should be way better, my 91 Z28 came stock at around 70 steps and it made a huge difference to drop it down).

 

Something to think about. If you search the Vette and LT1 forums for split blms and look at some of the things they talk about a lean #2 cylinder is one of them on a LT1. In fact the bins in 95 added individual cylinder fuel control to help balance out the cylinders and if memory serves me right #2 had a 10% fuel difference.

 

I was unaware that the #2 cylinder is a common issue even to the LT1's. I'll have to research that a little...

That's correct about the tailored fuel triming. I have the that bin and there's actually a 17% difference from front cylinders to back cylinders. But yes, 10% difference to the center four cylinders which have essentially no correction at all.

That could end up being my ultimate solution- go to an OBDI LT1 PCM for sequential EFI and tailor the fuel to each cylinder. Though, that Optispark is still a stumbling block.

 

You gotta problem.
Your relying on the ecm do to ALL your idle control, and using a simple hammer/ bang, closed loop system to then try and fudge the AFRs, towards, AVERAGING a Stoich AFR. (you didn't state if your running closed looped or not from what I've read), not to mention idle speed control, that's half based on timing, which does effect AFRs.

 

140 IAC steps signifies a serious issue. The IAC is essentially fully retracted. Even with the computer in full control, you shouldnt need any more then 30-40 or so. It sounds like there is something other then the tune being wrong.

 

If you look at TC and get an EE definition there are 2 tables for individual cylinder control one for idle and one for off idle. I have one definition that I personally read from the computer off of a stock 95 Buick Roadmaster. Cylinders 1 and 2 are 1.1 and 1.07 respectively at idle off idle the are both 1.03. Cylinder 7 has the lowest value which is .9 at idle. That should give you an idea of how varied each cylinder will perform.

 

And that car has a weeny cam in it! You can imagine what the distribution would be like on even a mildly cammed car like mine....