innovate LC-1 as a narrowband on my SC

niner11 · 04-18-2008, 03:45 PM

This thread has lots of good info in it but I am trying to do something much more simple

http://forums.pelicanparts.com/showt...t=innovate+CIS

If Klaus is still around it would be great to here from him but anyone that has insight please chime in. I have been running an LC-1 in a megasquirt II turbo setup (datsun) with great results and I'm thinking with the price of gas I need to run my 82sc closed loop these days.

About a year ago my SC had idle oscillations so I unplugged the O2 sensor and the car ran great. I have heard that low end torque might even be a little better but I have no complaints and the oscillation is definately gone since the car is open loop.

I'd like to take my LC-1 and install its wideband sensor but use one of the simulated narrowband outputs to communicate with the CIS frequency valve. Planning on setting idle AFR at about 13.1 to 13.4 to mimic what the correct CO would be. Not sure on these numbers but an old thread said John Walker recommended these in leu of using CO.

Has anyone recently done this? I dont even plan on installing an AFR gauge...just going to plug in laptop and set my CIS mixture then monitor the car to insure its not too lean. Also, can anyone confirm that CIS (with frequency valve) will run closed loop after warmup at all times but WOT. Thanks, Chuck

sammyg2 · 04-18-2008, 04:04 PM

I did exactly that with an LM-1. I used the wideband to read the mixture and the analog output to run the lambda computer.
It doesn't react quite as quickly as a modern electronic injection system but it does pretty good.

LM3929 · 04-18-2008, 04:06 PM

watching

niner11 · 04-18-2008, 04:12 PM

Sammy, do you recall if you reprogramed the simulated narrowband output of your LM controller (guessing it is setup the same way as the LC-1 is using the LM programmer) or did you just use the default settings? Thanks, Chuck

jbrinkley · 04-18-2008, 05:06 PM

the lc1 manual says the first output default is NB
i only know that because I have it sitting here.

niner11 · 04-18-2008, 05:33 PM

I read that earlier today but wasn't sure if all nb's are the same.
Here are the defaults....does it look like it has the same values as the factory Nb?

jbrinkley · 04-18-2008, 05:47 PM

yes .1 to 1.0 or close to

psalt · 04-19-2008, 03:33 AM

Niner11

The voltage of the stock sensor is around 500 mv at lambda, not 1 volt, which is near full rich. The slope of the voltage of the stock sensor is so steep, it's really just a switch to dither the duty cycle of the FV back and forth around stoich for the sensor.

If your goal really is "the price of gas" you are out in left field with a 13.1 AFR. You should plug in the stock lambda ECU and not be so anal about the idle. The idle mixture is irrelevent outside of emissions, you want to focus on the AFR at WOT under load. The advantage of a wideband controller comes in when you defeat the stock open loop throttle switch and use the wideband to adjust the WOT mixture to 12.5-13 AFR for best power. To improve mileage, you first need to remove the cat, then tune for 15.5-16.5 AFR under light cruise load with around 40 BTDC ignition timing. Use the wideband to determine your cruise AFR and adjust the FV duty cycle to come to your target. Best lean cruise economy occurs way beyond the range needed to feed the cat. GM did this on early closed loop EFI systems, but abandoned it for emissions and warrantee.

Paul

jbrinkley · 04-19-2008, 03:55 AM

who said stoich was 1 v,
the picture is self explanatory.

here nor there really, I still wouldn't bother.

niner11 · 04-19-2008, 04:00 AM

Paul, good info thanks. The reference to 13.1 or 13.4 Afr was really just something that I had heard to shoot for when setting the CO. Since I don't have a gas tester I was really just trying to use the LC-1 to set the mixture of the CIS and then let it run closed loop. Hopefully close to 14.7.

The Cat is removed and I am running SSI's for 74 and earlier so I imagine that their would be no problem running in the 15.0 or 15.5 range at cruise. If I set the "CO" using the afr it will be rich at idle (13.1 or so, open loop) but will it then run stoich in closed loop with the O2 sensor pluged back in?

The link in the first post is pretty interesting but I'm not sure its worth all of the trouble for the fairly limited benefit. If anybody has done this or if Klaus can elaborate on his setup I am all ears. It would be more efficient and probably produce more power at WOT.

psalt · 04-19-2008, 04:24 AM

It would be more efficient and probably produce more power at WOT.

No, this is eating your cake and having it too.

The stock system is closed loop at idle and below 35% throttle. There is a delay built in to reduce the fluctuating idle, but it does not eliminate it. There is no active idle control on an SC and you shouldn't expect it. If you really want to learn about CIS lambda, hook up the stock sensor and plug in an analog dwell meter to the green/white stripe wire in the test plug in the engine compartment. Extend the leads so you can see the meter inside the car. Then you can see the default cold open loop duty cycle, the warm closed loop duty cycle and the open loop default after 35% throttle. When you set the mixture, you are only really effecting the open loop, post 35% throttle mixture, because the system will pull back to stoich under that. These cars run best with the duty cycle set to 30-40 dwell and the timing set to 7-10 BTDC if you have the octane. Too much time is wasted on idle mixture and settings. The purpose of an idle is to keep the engine running when you are NOT driving. The focus should be on high speed mixture and timing under load.

Paul

niner11 · 04-19-2008, 04:51 AM

Many thanks! As far as eating cake and having it too I think that is what Klaus was doing in the other thread. They had a relay of some sort where CIS would run closed loop off of one of the LC-1 outputs up to WOT. At WOT the second LC-1 output (which was programed differently) would send a signal to the freguency valve for the desired AFR. Pretty neat idea but my SC runs well as is and I was just trying to run at stoich in the lower power settings without the idle oscillations that I had seen in the past.

Vipergrün · 04-21-2008, 09:14 AM

Following Sammy's lead, I am doing this with a PLX AF controller. Have the narrow band output going to the inner core wire on the green wire leading to the lambda brain. I have idle right at 14.7 and am at 12.8 - 13.2 or so WOT. When I let off the gas I get the typical exhaust popping and AF in the 17-19 range. Don't know what, if any impact that has. Car fires up and idles very well even on 0 degree days

I have my advance set to 33-35 degrees at 6K RPM's, which I believe translate to the 7-10 degrees at idle mentioned above. I always run at least 91 octane.

psalt · 04-21-2008, 10:33 AM

Brad,

How do you set two different AFR targets ?

Did you defeat the throttle position switch that causes the stock system to go open loop ?

Vipergrün · 04-21-2008, 10:46 AM

Paul, let me first say that I am far from being a CIS guru. I simply adjusted the mixture to get me at 14.7 at idle and bit more lean WOT. Before installing the PLX and before touching the mixture, the idle was off. As soon as I connected the PLX the idle smoothed way out and was easy to start. Also the WOT was 11 before making any changes to the mixture. Maybe I got lucky, maybe the PLX is compensating for something that was not working properly. I had run for a long time with the O2 sensor disconnected. I really got the PLX in hopes of moving to an EFI solution in the future.

psalt · 04-22-2008, 03:21 AM

Brad,

If your CIS lambda is stock, the PLX is doing nothing at WOT. The throttle switch grounds the ECU pin after 35% throttle and the system goes to a default duty cycle and ignores the O2 sensor signal. If you want the wideband to affect WOT, you need to disconnect the switch. If you want two different AFR targets, you will have to use two outputs and relays.