CIS AFR problems

[Walt Fricke]

I'm having no end of trouble getting my '82 SC engine to run lean enough at WOT. I'm aiming at 13.2 AFR. I measure this with the Innovate LM2 wideband AF meter. This is a US engine, and checking with an oscilloscope and by feeling the frequency valve with my hand, this part of things is working. This car is track only, so the I don't use the stock narrow band sensor, and thus the frequency valve is (or should be) always at its default values: idle, part throttle, and WOT. Part throttle is not important, but having a decent, even if a bit high idle, is.

Last year the AFRs were a bit lean - low 14s. I turned the mixture screw a bit richer. This spring, the AFRs tended to be in the 10s to 11s, which is too rich at WOT.

I also hooked up a pressure sensor to read hot CP. The trace in the Innovate is way too wiggly - there is something not right with Innovate's data storing system. But with smoothing the hot CP trace (black trace on just one graph) seemed at least stable. 2.5 volts equates to 3 bar (a bit low).

So I adjusted the hot control pressure upward from where it was on the gauge in the paddock - 3.2 bar, which is a bit below spec, to 3.6, and then to 3.9, which is a bit above spec. Alas, this seemed to have no or little effect on the AFRs.

So I got to work on the mixture screw. After leaning it out 1 and 1/16th turns (which is enormous) I moved the WOT AFRs up into the 12s, as best I can tell from glancing at the Innovate when I could on the track, and from the recorded trace. The downside of this is that the idle stinks - the car won't keep running without depressing the throttle some. In addition, the AFR at idle is 22 or 24, and I saw a spike on the recording up to 30. Fiddling with the idle air bleed screw did no good at these settings.

The system pressure is good, indeed if anything a bit high. It is 5.1 bar, which is the top of the spec although not the adjusting value. I saw a 5.2 during one of the many times I was back in my pit adjusting things after a lap or two. I've wondered if maybe I removed whatever shim is present in the pressure setting assembly, and backed the system pressure back down to the bottom if the system pressure spec, this would help with leaning things out.

Attached are a couple of Innovate data presentations. Innovate seems somehow not to have figured out how to get decent RPM readings, so I have disabled that even though it would be nice to correlate RPMs with what all the wiggles and spikes on the graph are doing. Most of these graphs are at the maximum smoothing of 4 seconds, because with no smoothing they look like a punk rocker's hair. However, the periods where the readings are more or less level or slowly trending up are most likely the longer straights.

I know that by depressing the slug to raise the hot control pressure I messed up the cold start side of things (it rose from an in-spec around 2.2 - 2.5 to 3.2 last time I looked with a cold engine at around 20 degrees C. But it is easy to change that once I settle on a hot control pressure (which, with my modifications, is also easy to change.

This is my first June run with no smoothing.



This is an early May DE, with smoothing and volts for the CP.

[img]

This is from idle through on track, with smoothing.
http://forums.pelicanparts.com/uploads26/1st_run_from_idle_thru_on_track1496623537.JPG[/img]
This is 1st run with 4 seconds (max) smoothing - the 2 is a cut-off 12 AFR on the scale on the left.



This is the last run, with the mixture screw leaned by over one full turn from the first run:



I'm going to test to see if, with the engine not running but the fuel pump running, and I loosen one of the injector lines, any fuel comes out. I suspect I have turned the mixture screw so far counter clockwise that at rest the plunger is not high enough up for any fuel to flow. That would explain the lack of idle, but leave me somewhat shy of what to do next - increase hot CP even more?

Because I am over rich, air leaks are not apt to be part of this puzzle. I suppose the Innovate readings could be just plain way wrong, but that also seems unlikely. The tail pipe shows a little dark soot when wiped. Last year, when it seemed I was leaner than I wanted (but not dangerously lean), I would get a hint of gray from the tailpipe wipe. All of which is consistent with the Innovate working. I suspect the Innovate introduces some random noise, because the AFR readings from past dyno sessions don't wiggle around lie this at all. And when I store data from my CP pressure sensor on the Innovate the readings dither, but when I just look at them with a VOM, they are very steady. A scope shows a frequency around the average DC level, which is only present when the Innovate is connected, so maybe the same is true for the AFRs. But the smoothed averages have to be meaningful.

Anyone had issues like this? Mind you, I'm not aiming at stoich, and want to run a bit rich in the low 13s at WOT, but 12s and below are giving away power even though the engine sounds OK and feels OK. Feeling OK isn't adequate (although feeling not OK is cause for even more alarm).

[don gilbert]

The 3mm screw only adjust your idle mixture. When turning it, you are just raising or lower the flapper valve from it rest position at idle. Has no effect on afr after you touch the accelerator.

[don gilbert]

Only way you are going to change afr after your 30 degree throttle switch is with control pressure. Fuel pressures look good. Check your freq. valve duty cycle with a dwell meter. You can also check to see if your 30 degree (wot) is sending a signal to it. Should go to a full duty cycle when you go past the 30 degree switch. If not working rite, check relay under pass seat next to your lambda control box.
I have the innovate kit to,,and I also have a adj. wur controlled from the dash, afr gage and control press gage in my dash, so i have played around with this alot. 11 at wot is pretty much standard with the factory 55 lb setting. Built in protection to keep you from burning up your engine on a hot day. Really loves it on a cold day tho. Just the limitations of the factory wur. Moving the plug up and down gets to be a hassle to get it "just rite" for hot and cold starts. I finally decided to go a diff. way as I am very tune oriented. Adj. set up is the only way to go if you are going to get serious about tuning K Jet CIS.

[Walt Fricke]

Thanks, Don.

However, my understanding is that the mixture screw affects the AFR throughout the range, as it adjusts the relationship between the control plunger and the air metering butterfly. With the screw turned rich, the plunger is higher at any given position of the butterfly. With turned lean, the reverse. Of course, the geometry of the throat of the air metering housing (wider as you go up) affects things, but why would the screw only affect idle?

Idle is the only time you can watch what happens while adjusting the screw on a running engine, and it is generally only used to set the idle - and in particular to set the CO for emissions, which are (or at one time were) measured at idle.

I checked the freq valve with an oscilloscope. Nice square waves. Next try at this I'll record them. It should work just as well with the engine off but fuel pump running, so I can see what changes as I depress the throttle. I believe I saw just that my quick and dirty look, which means the three prong switch on the throttle body is doing what it should. It is trickier to read a scope trace, because you are looking at a constant frequency pulse, but one whose on (high) and off (low or ground) proportions within the same overall period vary. The off, I believe, keeps the FV on longer and thus richens the mixture, but I get confused about this part and may have it exactly backward. The duty cycle with a dwell meter integrates all this and gives, essentially the percentage of high vs low in a complete cycle.

Funny thing is, I am too rich. If the TB switch wasn't working (and it is pretty simple, has almost no current running through its contacts, and I had it apart last year just to see) I'd think I'd be running lean, not rich.

But obviously there is something I am missing, or something isn't working right.

Tony from Philadelphia keeps telling me that the 030 WUR I have is for '74-77 2.7 CIS, and I should have the 090. But the various values are the same for both pretty much - about the same cold and hot CPs, and same system pressure. And while the 2.7s used the vacuum part of the WUR, the 81-83 US SCs do not have an actual vacuum connection from the TB to the WUR. I'm pretty sure Bosch used the same round brass aneroid barometer type double diaphragm at the base of the inner spring on both, but the most it could do pretty much is make a little adjustment for ambient pressure differences - nothing like what applying a vacuum to either the upper or lower chambers does.

So I am dubious that it is the WUR affecting things.

If the FD was screwed up, I'd expect the CPs to wander all over, but they don't.

[antares]

… gee love to help you out but just getting setup to rebuild and tune my 79 3.0 CIS system.

I have been looking for a good gas co2/co/no2 analyzer so is this Innovative LM2 wide band A/F unit the one to get for my car injection system ? … Also, If you have dual exhaust does this LM2 unit have a controller that can read two banks with 2 separate 02 sensors one for each exhaust ?

Any pointers would be appreciated … Dave

[wreckah]

could be plenty of things wrong with your CIS, hard to judge from here. The lambda CIS is quite complicated, and there is not much info out there of what the lamda brain and the lambda enrichment realy actually does.

don't rule out false Innovate readings, it needs rock steady 12V and good ground...also, the slightest airleak in the exhaust will throw off readings.

turning the mixture screw WILL change the AFR's over the whole RPM, as you statet. It's a comon misconception that it only affects idle.

I would start with making sure you have spec fuel pressure when warm. Leave cold start out of the equation for now. Then stop adjusting the WUR, and go to the next component.
Another thing, i don't think you can make this lambda CIS work right without the O2 sensor input. You can simply use one of the analog outputs of the innovate to go to the lambda brain. Program it so that it mimics a narrow band sensor. I believe one of the two outputs already is programmed to do so.
Then check the FV pulsing through the test port, and see what it does under various circumstances.

Of course make sure that all your other components are OK to begin with...injectors, seals, etc...

[wreckah]

and i thought 22.4 was the highest AFR that the innovate will display. that is equal to fresh air. I'm thinking something is up with your innovate install.

in the bitzracing manual, there is info about installing a capacitor to smooth out the 12V supply.
http://bitzracing.com/docs/WideBand_O2_Installation_Guide.pdf
(do not follow the ground connection info, as the LC2 is different than the LC1).
Make sure you have a perfect ground for your control box.

Make sure your innovate control box is NOT in the engine bay. It will become too hot and stop working.
Mine is under the passenger seat (where the lambda box was).

[T77911S]

Originally Posted by don gilbert

The 3mm screw only adjust your idle mixture. When turning it, you are just raising or lower the flapper valve from it rest position at idle. Has no effect on afr after you touch the accelerator.

if you are saying the idle mixture setting does not effect AFR at cruise and WOT then that is not correct.


I did LOTS of testing of timing effects on AFR's and engine temp and different AFR's vs CP and across the RPM range using an LM-2.

not actual numbers but say you set the idle at 14.5, 2500 may be 15 and WOT 13.5.
drop the idle .5 and the numbers across the RPM drop the same.

this is how CIS was designed. you set the idle for what BOSCH specs it to be and the design of the system provides the correct AFR's for a given RPM.

you can change the CP then readjust the idle for 14.5, then you will still have the same AFR's as above but what changes is the transition points. (not totally proven in my testing but based on knowledge of CIS and how it is designed).
take a look at the AFM bowl. it has three sections to it. 1, an idle section, 2 a mid RPM section (this opens up more) 3 the WOT section( this gets narrow again so the plate moves more to increase AFR).
so when you change the CP this changes at what RPM the transition from idle mixture to lean cruise happens and from when lean cruise to WOT happens.
this is why bosch sets a spec on the CP, so that these transitions happen and the RPM bosch designed it to happen. this is what I love about CIS. set the idle mixture and everything else happens when it should, plus I love the engineering that whent into it. same with mechanical injection but that's another long story.

on my 77s I could see the lean cruise start to go rich around 4k RPM. given that car also had the vacuum controlled WUR to make it richer at WOT. that's another story also.

I have done this same testing on my 930. timing and AFR's and LOTS of mixture vs CP settings. the mixture always follows the idle setting no matter how much I change the CP. I have done + and - 1bar CP changes.

idle timing.
less advance at idle leans out the mixture and makes the engine run hotter but ONLY while idling. so if you want a richer WOT and cruise, reduce the timing and you can richen the mixture up across the board.
the opposite is true for timing at WOT.
more advance and the mixture will lean out at WOT and you get higher temps.

to the OP.
I don't know what your idle is but it sounds like you have it idled hi?

when you have problems and you don't have things set to factory design, I would go back and set the timing to spec, verify advance is working, se the CP to spec, then set mixture to spec and start there. if your advance is not properly working that could be making it too rich at WOT.
also, what plugs are you running? with a track car you should go a little a little colder on the plugs. use bosch coppers and probably a 5 heat range. pull a couple of the plugs and check them after making the changes.

some timing tests I did where to set the timing 15 degrees ATDC and watch the engine temps go up at idle, AFR also leaned out. then I would drive the car steady and make sure the temps came back down and they did.


I always ran my vac retard on my car. the idle sounded deeper, I could run it richer and the vac retard came out at about 1000 or 1100 RPM.
same with my 930. the idle sounds much better, smoother and deeper with vac retard but I cant run it for other reasons.

[don gilbert]

I'm looking at real numbers on a gage for many yrs. If you try to adj. your wot mix by adj. that screw..Thats the wrong way. That screw is for IDLE mix adj.. Do you have the #2 signal wire from the Innovate hooked up to your lambda controler? I like to adjust the idle mix using a dwell meter,,trying to get as little occilating of the needle as possible. After that is set..adjusting the screw for wot is not how you adj. your wot afr,,it is controlled only be control press and freq. valve duty cycle.

[T77911S]

correct. if you are adjusting based on WOT that is not the way to do it.
go back to specs and set things up the way it was designed and then CHECK WOT.
if the WOT is too rich go from there but don't fix it by adjusting mixture.
the system was designed to work a certain way based on a starting point of idle mixture.

been thinking, could be the mechanical advance is stuck.
I would also be pulling plugs and checking them.
also do a calibration on the innovate (did not say which one) if you can.
I have gotten some weird readings out of mine before.

['78 SC]

Walt,

I know you've checked all the obvious stuff and have been methodical with the CIS adjustments, so I'll try a different tack - do you know the CDI is good?

I recently went through a long, frustrating process to fix a cold running problem with my '82 3.0 CIS track motor. I went through all the checks and settings (air leaks, WUR cold and warm pressures, Lambda dwell, throttle switches, temp sensors, AAR, decel valve) but could not get a good idle or even pull away when cold. Cut to the chase, I had my spare CDIs rebuilt by Bob Ashlock and all the running problems went away when I installed one of them. Immediately. Spark plugs have never looked better.

At that time, I was running the Lambda open loop, like you. My WOT AFR was ok, but the cold running was miserable.

I use an AEM wide-band to monitor AFR and have been happy with it. One issue with these oxygen sensors - all they measure is free oxygen. They have no idea what the hydrocarbon, CO, CO2 levels are. The AFR is calculated and that calculation is based on a combustion model that assumes good ignition. This will cause them to indicate lean when there is a misfire, even if the mixture is eye-wateringly rich. That doesn't sound like your issue, but I'll put that out there a limitation of these sensors.

[Walt Fricke]

Dave - I can't say I am completely satisfied with my LM2. It has the capacity - if you buy the right flavor - to run two O2 sensors. Mine will only accept one as far as I can tell, but it has a jack for the other one (leading me to fumble around if I plug the sensor into the non-working by mistake). You can't get any help from the maker - no e-mail, no user forum, didn't return my phone call.

I'll see if adding a 4,700 microF capacitor to the 12V supply will calm it down some, which would make it handier.

My Racepak track data recorder has a nice feature - when you shut off your engine, you lose only maybe 20 seconds of data - you spend more time than that getting back to your spot in the paddock so don't care. The Innovate seems to require that you stop the recording before you shut down the car - the manual intimates that this may or may not happen. But it shouldn't.

You can program its auxiliary output channels - if 50 pounds of oil pressure from a sender is 2.31 volts, you can make a 2.31V input output externally (if I have this right) as 50. But you can't have one of the four input channels show up on the display as 50 somethings - it will show up as 2.31. Maybe I just don't understand this. But I can't find any screen or procedure to program an aux input on the screen - or as the saved data to display on your computer.

And there is the RPM issue. The manufacturer suggests putting a potentiometer across the input, upping the resistance until you get no display, then backing it down until you get one. They have doubled the suggested resistance of the pot - without looking I think from a 5K to a 10K - you can see on their update to the manual. Didn't work for me, and the internet chatter suggests for some others also.

But it works just fine for some users, and fundamentally seems a good system, so maybe the key is keeping stray voltage spikes and frequencies out of its supply voltage (and ground). I've got mine plugged into the cig lighter, but can easily enough hard wire it in with a switch - I could even run a wire directly from the battery cutoff switch (this is a race car) to the unit, along with a ground from the battery ground, and add the capacitor right at where the unit's power cord plugs in.

You will note that Steve likes his AEM. I don't know that one, but Google surely will turn it up easily.

[don gilbert]

Had my Lm for yrs, no problems whatsoever. Does need to be calibrated ever once in a while.

[antares]

… thanks fellas, the reason I ask is for when tuning my stock 4 -bbl R.M. 1957 Buick which has dual exhausts as well as my SC … just simply attempting to find a unit that will not require a lot of arts n crafts time fumbling and marching around trying to patch and piece it together so it will work or having to stand on my hands to please it … seems a lot of the units I have looked into require just that or you have to go on a parts safari in order to get it to work, when the unit should of come with that in the first place .. How about a unit you just poke into the exhaust and has one 3 position switch that you can read CO2-CO-No2 , no plotting, grafting, printing etc… but guess that would make things too easy …. :{)

[Walt Fricke]

All - thanks for the suggestions. I'm going to back down the mixture screw to Bosch's suggested starting point - 1/2 turn from the "no fuel from injector line" point. I'll readjust the cold CP to the rich end of its spec just to make sure starting is easy and I don't need a squirt of ether. I'll leave the hot CP where it is - the little bit above the rather wide range Bosch/Porsche show as the acceptable values (3.4-3.8 bar) And then I'll use my scope to see if I can verify that the throttle closed micro switch, and each of the three states of the three contact throttle switch, have the expected effect on the pulse width modulation which controls the FV.

I will double check my distributor advance. However, when it was frozen (which is what got me screwing around with the CIS three years ago, because I attacked the wrong system trying to fix it at first), the symptoms were very severe - the engine would not accelerate at WOT. When I disassembled the distributor and fixed it, then the FV system relay, which had gotten wet from cool suit chest water spills, rusted through the spring on the contact inside, so I was trying to get the system working without the FV without knowing it. When I fixed that, things were more or less back to normal last year, though I didn't hit the magic 13.2 neighborhood. So I started messing with the mixture screw.

This year it was raising the hot CP, which in theory should allow me to get any ARF I want, and shouldn't lean out the hot idle beyond what the idle air bypass screw can handle.

I can check the two temperature inputs as well - the warm and the hot switches, which affect how the FV system reacts, and dope out what default values are, and ponder if I could defeat one or both to be permanently open or closed. Jim Williams has a large table showing what various states mean or do, which should help me.

But the fundamental issue here seems to be (I've got to check my plugs to correlate it with the tailpipe wiping)that I can't get the WOT to run lean enough. Most issues guys have had seem to be they can't get it rich enough (all this assumes you aren't running the narrow band sensor and using the FV to control emissions and improve fuel mileage).

What does anyone think about adjusting the system pressure down to 4.9 if there is a shim in there whose removal would accomplish that? While I have the push valve out I can inspect the little O ring on its nether end - I understand that those can fail with frustrating effects. I don't suspect it, but who knows.

I could pull the WUR apart again, but there is nothing to indicate that its pressure diaphragm system is impaired. When the screen was clogged a year or two ago, the symptom was control pressure = system pressure, which made for very lean running. Here my slightly imperfect pressure sensor data suggests that the hot CP is not wandering around when the car is on the track.

[Walt Fricke]

Steve - when first I had problems I took my CD box to a local shop which had an old style CD box checker. It tested out fine, though this meant an LED or some such blinked. Not a very sophisticated test. I do know that the tach signal from the 6 pin box is just fine, for what that might be worth (it comes through a separate circuit inside the box). I have an old failed 6 pin box a guy gave me in the late '80s which I disassembled and realized I had no way to check the SCR. aybe I should send that to Mr. Ashlock and spring for his making it good so I have a substitute to swap in.

['78 SC]

Walt - I never suspected the CDI when troubleshooting the CIS. The only reason I swapped it out was it began to fail at a DE where my son and I were quadruple stinting the car. Apparently it got hot enough that it would "latch up". Even then, cycling the ignition got it going again. Other than that, it seemed OK. I was pleasantly surprised the refurbed CDI fixed the cold running. Isn't there an old saying, "90% of carburetor problems are ignition"?

Bob runs checks before the repairs to identify issues. You also get a sheet with the output voltage at various RPMs for the repaired unit.

FWIW, I ended up with all my CIS settings pretty much middle of spec. Maybe a smidge rich at idle to get a 13ish AFR at WOT.

The AEM meter can output an emulated narrow band O2 sensor signal. I hooked that up to the Lambda system and the car runs great (well, it did right up until the 5-2 shift). Not a big deal on the track at WOT, but I was driving the car to the track and could see the improvement in cruise and general drivability.

[Walt Fricke]

78 - I'll follow up on the CDI. It is original as best I can tell with this '82 motor in my car (not the original motor for the car), so has a lot of time for internal parts to be affected by heat cycles and whatnot. But in terms of diagnostics, having a spare CDI box in the trailer is overall a good thing - running issues, swap boxes to see if that fixes it. Plus you can deal with it when the SCR or some other component fails (over enough time perhaps, like some other things, this is guaranteed to fail)and you get zilch from the box.

The Innovate also has a way of simulating a narrow band sensor, but since I reduced the insurance to comprehensive only and don't street drive it (changing the headers so I could have a muffler is rather too much work given alternative vehicles)I have had no interest in doing this. My troubles with Innovate may, as with many things in life, just be of my own doing or lack of understanding.

[wreckah]

Originally Posted by don gilbert

The 3mm screw only raises or lowers the flapper valve from its resting spot (idle). Once air flow picks up the flapper, the 3mm screw is suspended. That is not going to change your afr past rest no matter how much you turn it. The screw is not touching anything as soon as the flapper picks up air flow, so it has no effect.Big misconception about that.

i don't want to disturb this thread too much, but this is wrong i think,

watch this at 11 minutes, the 3mm screw affects the mixture over the whole path of the air flow sensor. The plunger is continuously in contact with the AFS through that 3mm screw.
(The resting point/screw of the AFS is reached from within the airbox, and is a totally different thing. It sets the base position of the AFS and that's it.)

https://www.youtube.com/watch?v=a4fJAfXYxWk

[don gilbert]

Your are right wreckah...I am wrong..but still trying to adjust wot afr with the idle mixture screw is incorrect. I'm going to delete that part of my thread so as not to spread mis information. What I said in the thread was "the screw does not affect afr at wot"...it does indeed, but this screw is for idle mixture only, wot afr must be adjusted by control pressure.