CIS AFR at idle vs at WOT
 

The wide band sensor I bought last year apparently didn't work, because my Innovate LM2 would count up to 99 and stop. Bought another one, and it works. But I am puzzled.

I was shooting (based on some advice) for a WOT AFR of 13.2. I thought that if I set the idle AFR at that value, it would be that at WOT, since the frequency valve in this 82SC motor (with the O2 sensor not connected) should run at the same default setting at idle as at WOT.

But there is some flaw in this thinking. The engine was running at about 10.9 at WOT, and well below 14.7 at idle. I kept leaning the idle, eventually to about 15, but the WOT values on the track stayed pretty much the same, maybe more like 11.9/12. This is a race motor, so power is all I care about.

But why won't the mixture screw lean out the WOT values? Obviously I can't adjust it at WOT, so have to rely on the Innovate to see what is there.

Plugs looked, if anything, to be quite lean - bone white ceramic with a tiny hint of brown at the tip, though the read didn't follow the prescribed protocol and was taken in the garage.

So maybe I wanted to shoot for an idle AFR of 13.2, and let the WOT value fall where it may?

Hello,
I have a 1982 SC and have noticed similar behaviors - have an AEM air/fuel gauge installed. I disconnect my O2 sensor ( as per directions) to set idle mixture so that the Lamda system will not be correcting as it is designed to do. I have my idle mixture (open loop) set at about 13.2 - which WITH the O2 sensor reconnected and lambda system correcting fluctuates around 14.7 as it is designed to do. At wide open throttle (open loop) my A/F ratio is 11.2 – 11.6 depending upon ambient temperature. I also have noticed that there does not seem to be a significant correlation between idle mixture and WOT mixture, as I have attempted to lean out the WOT mix without much success. I realize that my WOT is a bit rich but the car polls extremely strong and since at idle and cruising the lambda system leans it to around 14.7 I am not concerned with having excess fuel washing down my cylinder walls, and being in Florida with the heat, I definitely am OK with it being rich at WOT. I am not sure, but I think I read somewhere that the air sensor intake funnel angle was designed in such a way that it prevents adjusting the mixture screw so lean at idle that there would be a corresponding overly lean condition at WOT. Although I fully understand that adjusting the mixture screw does impact the mixture throughout all ranges, it just does not seem to have a very significant impact on WOT mixture. I also seem to remember that before I totally rebuilt my CIS to remedy significant air leaks my WOT mixture was about 13.2 due to unmetered air getting in.
Anyway, just thought I would share my thoughts.

Dave

Shared thoughts are what I am hoping for. Thanks.

Did higher ambient temperatures translate, at least qualitatively, into leaner WOTs?

Walt,

I need to go look up the values but I believe the duty cycle is different for idle open loop and WOT.

Idle is 50% and WOT is 60%. I believe this varied depending on model year. I think 82-83 this was the behavior and 80 & 81 were different still.

There is a picture that is floating around here from the workshop manuals that shows the differences.

It is my understanding that duty cycle at idle is 50% and once the 1/3 throttle position switch is triggered,"open loop" duty cycle goes to 65%. I also believe that colder air is denser providing more oxygen therefore a leaner mixture, my a/f gauge shows leaner in colder weather.

Well, I was only part right about the throttle switch in the FV system.

Here is what Jim Williams measured. The throttle switch on these CIS systems is basically a double pole single throw (if I have that right - center to pole 1, or center to pole 2) switch. The center pole is to ground, so at idle one of the other poles is open, the other to ground. Mid-range I think both are open. WOT the other pole is grounded. One connects to the mini-brain directly, one to the acceleration enrichment unit.

Here is Jim's diagram for hot running:

http://forums.pelicanparts.com/uploa...1596252492.JPG

So 50% duty cycle at idle and part throttle (no O2 sensor), and a richer 65% at WOT.

Maybe I set the hot idle AFR at 13.2? I can do that. At idle 14.7 or thereabouts the plugs look rather lean.

Yes, I had to reset my idle AFR recently after an exhaust system change. 13.1-13.4 at idle was what I had in my notes and what I reset it to. I landed on 13.2.

As I stated in my post, with my O2 sensor disconnected "open loop" idle mixture set at around 13.2 , my WOT mixture is 11.2 - 11.6. I would be interested in learning what another car's WOT fuel ratio is when that cars's idle is also set around 13.2.

Wikipedia has this to say: "In naturally aspirated engines powered by octane, maximum power is frequently reached at AFRs ranging from 12.5 to 13.3:1 or λ of 0.850 to 0.901.[citation needed]

Air-fuel ratio of 12:1 is considered as maximum output ratio, where as the air-fuel ratio of 16:1 is considered as maximum fuel economy ratio."

Which suggests my WOT values of high 10s and low 11s are too rich?

But also that this may vary from engine (model?) to engine? So dyno tuning may be the only way to get it just right for your engine's max power?

No. With the sensor not connected and at warm engine state, when driving with throttle opened below 30° angle (cruising and idling) the duty cycle will be kept at static 50%. If you accelerate and throttle opens beyond 30° angle then the duty cycle switches to static 65% means enrichment of the mixture. So at WOT your mixture should be richer. But as you're very in the 911 engine concept, I guess you already know that.

You report a behavior which I also figured out already a year ago when testing my CPU based ECU.

At idle, engine was set to default 0.8% CO with lambda regulation switched off, means with static 50% like when unplugging the sensor if using the orig. ECU for doing the initial CO setup.
Then, when switching to lambda regulation (plugging the sensor) at idle a duty cycle of about 60% results which is normal as here the lambda regulation enriches just a little cause 0.8% initial CO is below lambda 1 where Lambda 1 actually (on this engine type) equals to approx. 1.2 - 2.0% CO.

Then "at idling" the 60% duty cycle results in a perfect, smooth stoichiometric AFR based engine running.

Now, .... with warm engine when driving the duty cycle immediately drops down to about 30% and if rising the rev to about 2000 RpM on a freeway or highway the duty cycle even goes down to about 10-15%! Means in AFR words: the sensor noticed in this case at approx. 2000 RpM that the AFR was noticeable too low = mixture was too rich. So this was hereby compensated by the low resulting duty cycle of 10-15% which makes the mixture stoichiometric again, means 14.7:1 (14.4:1 with E5 gas).

This means that the construction of the K-jet let's the mixture enrichen very much already at middle revs and for sure at higher revs.
So IF in that state of driving at medium/higher revs a duty cycle of 15% means an optimal combustion of lambda 1, means AFR of 14.7:1 (or 14.4:1 with 5% ethanol added fuel) then here your engine when're the lambda operation is switched off (sensor disconnected) with static 50% duty cycle will already result in a much richer mixture even before WOT.
Now in your case when hitting the gas pedal even more with a wide open throttle beyond 30° angle a duty cycle of 65% will push the enrichment even more. And thats why you get an approx. 10.9:1 AFR at WOT in your case.

Now we come to the next thought:
Yes, the idle initial CO adjust via screw does affect the WOT mixture as well, but proportionally in a very small way cause the air flow sensor funnel comes with diff. angles. A modification of the sensor plate initial adjustment here doesn't help neither. As here when adjusting the initial state of the plate in a way so far off that the WOT will result noticable leaner, then the middle load/rev states will lean out extremely and will be totally messed up.

Means, without lambda operation working on that engine type you can't modify the WOT AFR in a way without messing up the AFR of the middle load/revs state. Thats the very downside of the static 65% enrichment of the original state of the orig ECU.
I used in my ECU an option that beside the static 65% enrichment you can choose a "enrichen by XX%" in case of WOR. Cause I found out when driving at 2000 RpM and here a perfect 14.7:1 mixture would result in a 30% duty cycle then a higher duty cycle by "+25" will result as the best "switch" to obtain a perfect 12.5:1 WOT mixture when accelerating. Means in that case/example a duty cycle of 55% and not 65!.

Andrew - thanks, I was hoping you would have ideas about this.

Makes sense that the mixture screw will have less effect at part throttle, and even less at WOT. Sort of like the idle screw has a lot to say about the idle mixture, but just about nothing at WOT. Luckily, I haven't considered messing with sensor plate height, which seems like a good thing.

Since it seems the O2 sensor doesn't affect the WOT 65% duty cycle, am I on track by thinking that - for race track purposes - I won't get better performance by hooking up the O2 sensor? I could do that, since each exhaust bank has a bung, but I'd rather not, especially if it won't give me more power at WOT.

Assume again no O2 sensor (except for my Inovate wideband to see what is happening). What if I disable the WOT part of the switch? I think that would be green/red from pin 3 of the switch, either at T10/12, or pin 4 at the acceleration enrichment unit? No ground on that line (or would it be pin 3/T9/12 without disconnecting the temperature switch?). That way 50% duty cycle all the time? A bit off of the 55% at WOT you found good? Maybe better than 65, though?

I could compensate there a bit with the warm control pressure? Adjust it to a lower value within the range given? Or even a hair lower? Seems I could use the mixture screw as needed to get idle in a decent AF range if it got too high?

Usually I run US pump premium, which is 91 or 93 RON/MON octane, and normally has alcohol in it (E5?). I'd want my WOT AFR to be 0.3 lower with E5? I have not tried any dyno tuning to see exactly what WOT mixture gives the most power (and, given that you can't adjust that with the emissions screw, it would be a bit tough to do other than by making the control pressure adjustable - which isn't all that hard but I haven't tried that on a dyno). Looks like with your electronic bits you could make changes like that on the dyno, though - very nice.

But you view 12.5 as perfect for WOT, and earlier in this discussion there are suggestions going up to 13.3 or 4. Guessing (based on indifferent scales) from Bosch manual graphs, 13.2 looks like where they thought it was, though that would not have included alcohol in the fuel, which didn't happen when they were designing and testing the CIS.

Further thoughts earnestly solicited.

With my current lean idle settings, I'm going to have to be careful next time I am at the track to do a proper spark plug cut: WOT approaching the exit, then turn engine off, coast to a spot in the hot pit, and pull a plug. When I get around to it at the track garage, the engine has been idling or nearly so for quite a while. Last year (Innovate not working) even then plugs showed rich. This year, quite lean, but maybe due to bad procedure.

Walt

Reviewing the various Bosch manuals, I see that the air measuring system gives a momentary enrichment when you "hit the throttle." The initial extra rush of air causes the plate to rise up farther than it would at cruise at that RPM, which adds more fuel until the plate settles down and continues its more linear climb.

So perhaps rich spikes on my Inovate data indicate the start of accelleraions out of corners?

Walt

Don't forget that you can change the effective AFR with control pressure and system pressure.
Ian

Right - I've thought of that. I'm not keen on reducing system pressure - affects the injectors some? And would call for finding a thinner washer, or leaving one out, which might be too much.

Reducing control pressure I can deal with. I modified a previous WUR with two threaded holes, and a sort of raised bottle cap I could use to have the nuts on the two studs or bolts raise or lower the main pressure slug. To lower CP I usually, after loosening the appropriate nuts, used a punch in the longish Allens I had screwed in to tap down for higher WCP

Affecting the pressures here – no matter if control pressure or System pressure – is not the way to solve your problem with correct idle AFR but too low AFR @ WOT.

Because if you rise the Pressure to obtain a higher AFR at WOT ... this automatically will lean out significantly your idle. Because in case of a 930.16 engine the idle and WOT pressures both do stand in direct relation ... as no manifold pressure connected to the WUR affects the control pressure at acceleration. or WOT.

First of all, ... its the biggest myth that the lambda regulation "itself" on the 930.16 engine is responsible for any power loss.
As you said, hitting the gas for WOT makes the throttle switch turning the lambda control off.
So, if your are very into this lambda system, its even a wonderful part of the engine for tweaks & Co.

So what you can do is: Get your lambda control fully working and for the following CO adjustment do NOT unplug the sensor as you should do normally.
Now connect a DMM with duty cycle option or a dell meter to your test port. With connected O2 sensor, engine warm and idling turn the CO screw CCW till the duty cycle on your DMM is at 60% or 54° on a dwell meter.
By doing this you will get a leaner initial mixture which directly affects the AFR at WOT (65% duty cycle).
And .. as the lambda regulation now is doing its job, you won't suffer from a leaned out idle cause the lambda control brings the idle back to stoich! :-) So idle will be ok at 14.7:1 and WOT will be leaner ... here you can play a bit with CO screw for setting up the initial idle duty cycle, maybe istead of 60% actually 55% or even 70% is needed for obtaining a 12.5:1 AFR (or whatever you want) at WOT.
Let me know if this helps

Or ... if you got a wideband sensor controller like the innovate one, you can do the throttle switch relay trick.
http://forums.pelicanparts.com/porsche-911-technical-forum/246909-lm-1-wot-closed-loop-trick-freq-valve-cis.html

Or use the CPU controlled Digi SC ECU I showed you, where you can set up everything for obtaining an on target AFR@WOR with a narrowband sensor, ... Wideband sensors from diff. brands are supported as well.

Walt,

How long are you observing AFRs in the 10-11 range? I find that even tuning idle AFR in the garage is challenging because of the fluctuation. Blipping the throttle causes the AFR to bounce and then settle. Are you looking at log data or just eyes on the gauge?

If you’re seeing momentary fluctuations in the 10-11 range I think that is different from sustained periods of time in that range.

If log data, what is the sampling rate?

I thought that perhaps my experiences might be helpful.
When adjusting my a/f ratio at idle with the O2 sensor unplugged, my AEM fuel gauge hardly fluctuates at all due to the constant open loop 50% duty cycle, with the sensor connected in closed loop,at idle, the gauge fluctuates about .5 percent around 14.7 as it is designed to do as the O2 control unit is constantly reacting to mixture reads from the O2 sensor in its attempt to maintain 14.7
When my throttle exceeds 1/3 and activates open loop via the throttle position switch the mixture switches to around 11.8 and stays there as long as I am past the 1/3 throttle and accelerating or WOT.

I have recordings from the Innovate LM2. They are very squiggly on the track, even with a fair amount, say 2 seconds, of smoothing. And lots of spikes, which I think happen a) when I hit the brakes, b) when I go from cornering to WOT (lean spikes?), and maybe also c) when I blip the throttle downshifting. Haven't got it quite figured out.

On my Racepak data logging I can figure out where I am on the track by knowing the track - this must be the straight, see these two small dips in the increasing speed trace, etc. I haven't figured out the Innovate's log.

I'll use my "take a photo of what you highlight on the screen" function to make JPGs of a couple of representative ones. Idle (like waiting on the grid) is easy to spot, and is relatively smooth. Katie bar the door when I get on the track.

I can't go to digital etc control - rules don't allow it.

Andrew - my thinking was that if I set CP higher (starting carefully, just moving it to the leaner edge of the band on the graph), I can compensate at idle with the CO/mixture adjustment screw, plus the idle air screw. Because what I am hearing is that at high RPMs (on track typical range is from 4,000 RPM or maybe 5 depending on the track, to 6,700, with upshifts on straights at 6,250 (best fit to the torque curve and gearing). All WOT stuff, well above 30% throttle opening. I just want a stable idle for convenience before and after being on the track.

And I perceive I can get some power gain by keeping the WOT mixture in the 12.5-13.2 range. It for sure isn't there as things are - high 10s, 11s, occasional low 12s (where the trace isn't spiking like crazy).

Since at WOT (and all else above 30%), the O2 sensor contribution to what the FV does is nothing (right?), and the WOT programming built into the system gives a 65% FV modulation.

My second (right side header) bung is buggered - I installed a sensor without anti-seize, and damaged the threads. I'll have to see what I can do about that so I can hook up a narrow band sensor (I have one or two of those hanging around). How important is it to have the factory wiring and plug to hook up to the still factory wiring from the engine to the control units?

ok, but the compensation with the CO screw will also affect the WOT.

Yes, the lambda control driven by sensor reading is switched off and a static 65% duty cycle modulation of the FV is given.

With all my respect, but I think with all that procedere above just for trying to achieve a better AFR at WOT, ... at the end you won't feel a significant rise of power.
What has already been done else on your engine? (exhaust system, timing, cams, P/C Replacement ... etc?)

http://forums.pelicanparts.com/uploa...1596716690.jpg
http://forums.pelicanparts.com/uploa...1596716738.jpg
For total AFR control, I use the Innovate 02 kit, dumped the wur for a regulator mounted between the seat. Dial in whatever afr I want. In the morn...dial down to 15-20 lbs bust off as soon as you engage the starter...a few blocks up the road...dial up to 55. Need power? Just dial down to 12.8 in my case. The WUR is prehistoric..never in spec and the problem with most driveability issues.

Andrew - in the class I race my '82 US SC (well, that is what the engine and transmission ratios are) in, the air filter is free. The exhaust from port to tail pipe is free. I can set the stock cam wherever I want, and the stock distributor where I want. And adjust the CIS how I want, but not replace it or a key part (like the WUR). I can use high strength rod bolts. But basically the engine's parts have to be stock.

I've got the Henderson/Buckley exhaust. Loud as hell, but at least 10hp better than what Porsche used on those cars in the day, or the '74 headers or SSIs, etc. After doing all my engine assembling and tuning for years I paid, as a present to my wife so she could race it for a year a shop with winning cars to do a rebuild, and it set the cams (I had set them mid range per Bruce Anderson) and distributor timing - which has to deal with our pump fuels. The shop owner then test drove it to match a track record even though he was too tall for the fixed seat and didn't try to see what more he could do.

The car has more potential than I have as a driver (I'd like to think not much on some tracks, but on others I can see I am way off the best). I can't imagine that running as rich as it does at WOT isn't giving away something. Why would Bosch publish those nice graphs of AFR vs power, economy, and NOx if AFR at WOT didn't matter.

Walt, I think what you’re getting at is that if max power is ~12 and you’re getting to 10s, have you gone past the point of diminishing return and now you’re in fact getting LESS power?

That’s a fair question and the reason I was asking about time interval.

If you’re in the max power band on the straightaways then I tend to think don’t be concerned if you blip up to 10s when you let off the gas (for example). But if you’re consistently in the 10s on the straightaways then what is the net result?

Best I can tell, the only fairly stable readings are a) idle, and b) WOT accelerating down a straight, maybe can tell upshifts by spikes. One reason I believe this is that I look (occasionally) at the reading while on the straight. In an SC, on the long Brainerd straight where I have hit 130 (not this year) before turning in (no braking for T1 in a car which only hits this kind of speed), there are 2 upshifts (3d to 4th, and 4th to 5th, and plenty of time to glance at instruments if you aren't catching someone (the norm is for 991 RSRs or GTB cars to blast past). Not so beyond that.

And those are the over rich readings - what I see with my foot to the floor. So they have to correlate with what I see on the graph of what is recorded by the LM2.

When I get around to screen copying bits of the recordings I'll post.

For checking the frequency system I added a BNC connector to the Acceleration Enrichment unit, so can read on a scope. Alas, since this is PWM, I can't just put a frequency counter on it. Easy enough to spot 50%. As long as 65% is the only other value (O2 sensor not hooked up) anything different would be the 65%? I have a dwell meter, though it is old, was cheap, and I've been a bit unsure about trusting it. I think this is what I was seeing three years ago when I was heavy into fiddling, opened up the "ECU", added wires, etc.

If I get really creative, my LM2 has several data inputs for 5V signals. I'd have to go back to the drawing board on that, refresh recollection of what the voltage level involved with the FV is, etc. If it is 12VDC peak to peak, maybe a simple resistive voltage divider would knock that down to 5?

That would provide a FV trace on the Innovate graph. Might help identify where I was off the gas braking, show the (brief) time the throttle was in the cruise segment of its travel (trail brake?), etc.

But without the O2 sensor telling the system what to do about the pulse width, and then only at the cruise range of the throttle switch, I'm not sure what monitoring this will do for me. I know the switch works, and the FV responds as it is supposed to based on that switch. How is this going to help me get the WOT AFRs to a better place? The emissions screw alone has proved to me that it can't influence WOT enough to do this, pretty much no matter how lean I make it at idle. Plus I can see why that screw is less effective as the air flow gets up into the WOT range.

Hence the hunt for some other way without adding electronic stuff.

I mistakenly posted some verbiage and pictures at this location, thinking it was this topic. It was on CIS, but a guys problems. Take a look.

http://forums.pelicanparts.com/911-engine-rebuilding-forum/1069233-engine-stalling-off-acceleration.html#post10979277

Here is what Porsche had to say about the effects of the three wire throttle switch:

http://forums.pelicanparts.com/uploa...1597013521.jpg
http://forums.pelicanparts.com/uploa...1597013521.jpg

I see I was wrong about the sensor having no effect at idle. In a warm engine it does, but thats not what I care about. This seems to confirm that closed loop has no effect on warm (which is all track guys care about) WOT AFM. Any counter argument?

Andrew estimates that a 55% duty cycle at WOT open loop would give AFMs in the desired high 12/low 13 range. How might one go about doing that without just replacing the ECU (or its innards)?

To start with, assume no O2 sensor, so the base is 50% idle, 50% cruise, and 65% WOT. The 3 wire throttle switch could be used to disable the WOT duty cycle, which would remain at 50%. I suppose I could try that. I'd have to have the function of the 3 way switch right. Here is what I think it does in this regard: at warm idle or cruise (throttle below 35*) it is open. At WOT it is closed, and provides a ground to pin 7 on the ECU. Leaving the thermostat connection to pin 7 (opens at 15*C) alone, disconnect ECU pin 7. Check to see if WOT is 50%.

If this works, I see two likely alternatives, both using the WOT closure of this part of the throttle switch:
A) provide the O2 sensor input to the ECU (pin 2) a dummy 02 signal. Should be easy to provide a signal which varies at a frequency between 0.1V and 0.9V which will do the trick. I haven't figured out what that might be, though. The FV frequency is 70 Hz (or 69). But it is the on pulse width for that set frequency (which is generated in the ECU) which controls the pulse width modulation to the FV.

So that is question 1: what fake 02 signal would cause a 55% (or some other, based on dyno testing) duty cycle at WOT.

B) directly drive the FV. The same WOT switch would provide the ground to a relay controlling power to a pulse width generator set to 55% at 70Hz. To be safe maybe, the same relay could disable the signal doing this generated in the ECU and sent from pin 15. This would require figuring out what the signal voltage should be, and its current capacity.

Here is the tricky bit: The scope signals from the test pin on the ECU (pin 17, also found at the test plug in the engine bay) are nice square waves of about 6 volts peak to peak. Circuitry in the ECU must shape up the signal sent to the FV to make it look this nice. The scope picture of the signal from pin 15 to the FV, which is what turns it on and off, is much more complicated. In fact, I don't understand it beyond suspecting that the current needed to operate the FV is high enough to cause a big voltage spike, and maybe some of the rest of the form. Here is a CRT picture of what the two signals, taken on separate channels, looks like. Also one of my digital scope pictures of the same thing.



http://forums.pelicanparts.com/uploa...1597016515.jpg


http://forums.pelicanparts.com/uploa...1597016619.jpg

And another CRT:

http://forums.pelicanparts.com/uploa...1597016671.JPG

Plus what someone posted as what powering a fuel injector on and off looks like:

http://forums.pelicanparts.com/uploa...1597016794.JPG
http://forums.pelicanparts.com/uploa...1597016794.JPG

So question 2 is what specs do I need for a PWM generator set at a frequency of 70Hz and a duty cycle of 55%? Can I accomplish that with a 555? I've got a handful of those and the resistors and capacitors needed to control things (would need help figuring out values, though). Or something else?

Walt

Control Unit Accelleration Enrichment unit
 

The 81-3 US 3.0 CIS system has an additional electronic component in a cigarette sized pacage, attached next to the large ECU: The "control unit accelleration enrichment" box, which shows up on the wiring schematic for these models. What does it do? Looking at the Porsche or Bosch information on what throttle position has to do with fuel enrichment finally led me to understand its function. If I have doped it out right, when the throttle goes from idle, or just above 1 degree of opening (micro switch opens) and the engine is cold (maybe also warm, but not at operating temperature), there is a 2 second burst of a 75% duty cycle to the FV.

This must have been a useful add-on to deal with sluggish response on cold engines, maybe merely warm engines, when getting going headed out on the Autobahn. Rather than redesign the PCB of the ECU for this, Bosch just added another box. Sort of like how all the throttle air systems got added to the pre-ECU CIS.

So it would appear that this CUAE box is irrelevant for track driving, where your engine is always hot after the grid and a warm-up lap or formation lap.

Have I got this right? Ignoring this box when it comes to thinking about WOT AFRs simplifies things.

LM2 graph traces
 

http://forums.pelicanparts.com/uploa...1597017985.JPG
http://forums.pelicanparts.com/uploa...1597017985.JPG

About 2 minutes, which is about one lap, at Brainerd.
Idling on grid, then driving through hot pit to track.

It depends on the duty cycle which is needed for a 14.7:1 AFR when driving without acceleration.

Means:
When driving normal and the working lambda control shows a duty cycle of 20% (which is normal when driving at i.E. 2000-3000 RpM) which in this state is needed for Lambda 1 (14.7:1 AFR), then a jump up to 65% when accel., like the original ECU does, will result far off an AFR of 12.5:1!
See the point? Thats the very downside of the original ECU that the control actually switches to a static 65% duty cycle when accelerating. But in that generation of ECU where simple electronic components are included, no conditional programming was possible to just "add" a value of duty cycle percentage for a proper 12.5:1 resulting mixture. And thats what I meant: IF the ECU let the duty cycle result in 30% for hitting Lambda 1 when driving normal, then a duty cycle of 55% would result más o menos in Lambda 0.85, means an AFR of 12.5:1. So th 55% example above is not a static value where this will give a proper 12.5:1 at all RpMs/conditions.

For your interest of performance for a racing engine this accel. enrichment unit is not useful at all as it only acts when engine oil temp is below 35°C°!

If the initial CO screw setup meets the factory specs, means a CO of approx. 0.6% with unplugged sensor and a hot engine, then when sensor is plugged again, a resulting duty cycle is in the up to 60% range. So ... if engine is cold and by this the lambda operation is switched off, then the cold duty cycle of 65% would be not enough for a jerky free slight accelrating out of a cold idle. Thats why the unit pulls the duty cycle to 75% at 5° throttle angle and at 15° angle as well. And this at both states, means when deaccel. and accel.

I don't know this output from innovate. Which graph corresponds to what?

AFR scale on left. Top is roughly a lap on the track racing, maybe 1st or 2d lap after the start (I should capture one much later in the race, I suppose). Scale in the middle is time from start in minutes and seconds (snip for bottom graph didn't get time ticks, but same) at 2.5 seconds per gradation. Bottom is idling on grid, then modest driving through hot pit to get to track. Both about 2 minutes.

I assume lean spikes are braking and upshifting, where off the gas. Braking only 6 times per lap, with one lift. One upshift from 3 to 4 and 5 on a long straight, one from 3 to 4. One downshift from 5 to 3 (I don't go through 4), one from 4 to 3.

Innovate has an RPM channel, a voltage channel, and several other, undedicated channels. The RPM channel can give spotty results - you are advised to use a potentiometer as a voltage divider to calm the RPM signal down. Sometimes I believe the results, sometimes they have the RPMs way high - I know what my upshift RPMs are from the tach. I used another of the channels to record control pressure on the track, but didn't learn anything significant before the sensor for some reason quit three years ago. The guy behind Innovate knows his electronics and programming, but his documentation can be somewhat over the head of someone like me who knows the concepts, but hasn't internalized the steps needed. Plus he got tired of answering questions (engineers who write manuals tend to be irritated by people who didn't seem to read the manual), so I haven't been able to get much help there. And, having just bought another wide band sensor from Innovate, I suspect maybe the Holley etc carburetor company may have bought him out?

Andrew - there is very little part throttle on a race track. WOT, upshift, WOT, brake/throttle closed, WOT. Smooth acceleration often means you squeeze back on the gas a bit before the apex, or at it, or a bit after, maybe reaching WOT at about track out, though sometimes by the apex. But mostly foot to the floor, or idle. Plus the blip for downshifting to match RPMs.

When conecting an DMM or another electr. device to the rev counter, then at least use a 1mOhm resistor, so as voltage devider in combination with a 2mOhm resitor leading to ground. This gives you the "Hertz x20" calculation for the RPM. BUT thats the quickn dirty approach as the CDI internally comes with an open collector output on the TD pin.

Not shure if a leaning out that much is normal when just shiftig the gear. You did disconnect the seonsor, so you're not using the lambda control, right?

For fuel economy, many (all?) newer cars shut off fuel injection altogether when throttle is closed and engine speed is above idle (or some small multiple of idle, ramping injection up as engine speed nears idle).

What does the CIS do when throttle is closed? Presumably it thinks the engine is idling, and therefore provides the idling quantity of fuel?

If so, those lean spikes make some sense.

Does not explain super lean idle, is that on purpose, Walt?

Sent from my Nokia 7.1 using Tapatalk

1) The Innovate wide band sensor is in the bung in the left side exhaust after the collector. So no Lambda system adjustment.
2) The very lean idle is from trying to get WOT to go leaner (stop being so rich) and be at or closer to the AFR for max power. I think it nudged the WOT readings up some, but not there from what I see looking at the display on the straights when not otherwise occupied.
3) Porsche/Bosch built in a system to deal with what happens when you snap your foot off the gas. I think the center gizmo in the first picture does that, with the nipple connected to the throttle body below the butterfly, to provide a vacuum allowing unmetered air to lean out the mixture. The second photo shows where this air enters the rubber boot over on the left (the other tube in the boot is the breather tube for the oil tank).

http://forums.pelicanparts.com/uploa...1597125996.JPG

http://forums.pelicanparts.com/uploa...1597125996.jpg

The deacceleration valve does not "lean out" the mixture, it provides bypassed "metered" air (same conduct as the AAR and AAV) to the manifold to reduce the vacuum resulted by a fast closing of the throttle at high RpMs. This is only for emissions.

With lambda control deactivated and then reducing the CO value at idle in such a way to compensate an over enrichment at WOT is not the way. Thats why your mixture in your plot shown above leans out that much when deaccelerating. Just my two cents.
I really suggest you should reactivate your lambda so you can exactly follow your approach you do above by using such a lean initial CO value, .. but then with lambda control at idle or deaccelerating your mixture won't lean out that much cause the ECU brings the mixture back to stoich.
I have an innovate sensor/controller as well, ok a LC-2 but I guess yours also comes with an narrowband supporting 0-1v output (brown wire?) where you can try to connect this as sensor signal to the ECUs Pin2 (inner signal of the sealed green wire) terminal.
In case of that the ECU won't support this signal, go into the innovate sensor setup using the software and try to change the signal setting from sensors initializing and failure state from "0v" to "high resistance".

To answer the question, 81-83 US cars have a vacuum port on the distributor that retards timing at idle. I would not think it would it would come into play on the track.

There are other differences over the years and markets.

Just a thought, but don’t these wideband sensors require a periodic re-calibration process?

The manual of the LC-2 wideband controller i.E. gives here recommendations for the calibration intervals.

At the beginning I re-calibrated all 6 months, but here no change, so I increased the interval to every year – but lets see, I'm quite shure there will be no need to do the job in an interval less than every two years

Walt,

Are we ready for an off the wall comment? :D

Have you considered forcing the Lambda system to run in Mode 1 by grounding the 15C switch? Compared to the table in your post #6, that will cause the FV to cycle at 65% across the board, unlike the Mode 3 operation that has the FV duty cycle at 50% at idle and 65% (richer) at WOT. Note there will still be brief enrichment (75% duty cycle) when coming off idle.

That would give a starting point where both idle and WOT AFR are low, so there's room to lean out the mixture across the RPM range by increasing the WUR warm pressure (maybe even staying within high end of spec). Using that and the idle mixture might get where you want to be at WOT and have an acceptable idle AFR.

All this would make cold starts (actual cold weather starting) difficult if not impossible, but I understand your priorities for a track car.