Adjusting Mixture on Lambda CIS with CO meter

[Nickshu]

What is the spec for mixture adjustment on Lambda CIS (with O2 sensor) system for the CO reading at the tailpipe?

I have searched the old threads and found everything from 1-2% to 0.4-0.6%.
From what I gather (and correct me if I'm wrong), the setting on NON-lambda CIS is 3-3.5% but with an O2 sensor working the setting is much lower???

Confused.

[Peter Zimmermann]

1. You never read the CO% at the tailpipe on a lambda SC. Check the side of your cat, you will find a test port that is just forward of the cat body, on the flange pipe. It faces, approx, the inner sidewall of the LR tire. Take your exhaust sample there.

2. CO spec for a USA '82 car is 0.4 - 0.8%; before cat, oxy sensor disconnected, engine at running temp.

[Nickshu]

Thanks Pete. Only thing is my 911 has SSIs on it and no cat with a two in one out stock muffler. Should I measure the CO at the tailpipe then with this setup?

[don gilbert]

If you cant get the plug out, you can hook up a dwell meter to you diagnostic plug. Quick way to get a reading of o2 function.

[psalt]

Nick,

The reason the spec is so much lower is the benefit of closed loop control. The main reason modern engines have a service life several times what used to be the norm is proper fueling. Excess fuel, when it is not needed to produce power, washes down the bores and dilutes the oil. There is a reason there are so many later SC's that are +200,000 miles. Follow Don's suggestion, you will learn to verify/set the mixture with a dwell meter and verify that the lambda system is working correctly open and closed loop. An 82 SC usually runs best with a closed loop dwell of 30-40 with the vacuum retard disconnected and plugged. Remember, when you adjust the mixture on a lambda SC, you are really only effecting the open loop (cold running and >35% throttle) mixture. The system pulls the idle and cruise back to stoich, but you get a better cold idle warmup and WOT mixture.

[Nickshu]

Thanks psalt. I have a dwell meter but could never get it to swing properly. Wondering if it's set so rich the o2 can't compensate. At one time the mix was set by a shop with the o2 left disconnected so I suspect very rich. Gas mileage is also horrible. Going to set the mix with my CO meter then retest the dwell. The car has a new o2 sensor installed on SSIs.

So just to verify, I should disconnect the o2, then set to 0.4-0.8 CO, then reconnect the sensor and check dwell using the test port? I will warm up and set timing first.

[don gilbert]

I would first check to see if your ecu is powered up. The relay under the pass. seat has a high fail rate. Can you hear your freq. valve buzzing? If all is good power wise, I would turn the mix down just a little a a time, (counter clockwise) while watching your dwell meter, if all is good, the needle will start swinging when you get close. Ever 02 car I work on with 02 unplugged, has mixture set way to high.

[psalt]

Nick,

What does the dwell meter show ? You want to know that the system is properly functioning before you adjust anything. Lambda CIS is different from Basic in several ways and if you don't understand them, you will never get factory drivibility and performance. The way the ECU dithers the closed loop mixture is by varying the lower chamber pressure with a variable fuel pressure regulator (FV) that is really just a single EFI injector. The higher the lower chamber pressure, the leaner the mixture. Increase the duty cycle of the FV, more pressure is bled off, and the mixture is richer. Unfortunately, when the system fails, (fuse, ECU, FV failure) no pressure is bled off (blocked) and the mixture is too lean everywhere. Compensating with the mixture screw cannot give you the correct mixture curve. If you have found the test port and cannot get a reading with the dwell meter hooked up , try a different dwell meter. When the engine is cold, the mixture is irrelevent and you should get the default FV pulse (58 dwell, 4 cyl scale). This is your first step, if you can't get there, fix the problem. Once you know the ECU and FV are working correctly, drive the car for 10 miles, hook up the meter and check the closed loop mixture. If the needle is not dithering, try disconnecting the wire to the 15C switch (Bentley is wrong here). Adjust the mixture screw until the needle dithers rhythmically between 30-40 dwell, making the final adjustment in the rich direction. In closed loop, you are reading the correction factor, a lower duty cycle means the mixture is rich and the FV is raising the lower chamber pressure to lean it to stoich (just the opposite). When adjusting the ignition timing, disconnect and plug both vacuum hoses at the distributor, and set the distributor to 25 BTDC @ 3000 rpm. Idle timing @ 950 rpm should be around 5 BTDC, if the distributor is working correctly. Reconnect the orange vacuum advance hose to the front, but leave the grey retard line disconnected and plugged. Reset the idle speed to 950 rpm with the thumbscrew.

[Nickshu]

Originally Posted by psalt

Nick,

What does the dwell meter show ? You want to know that the system is properly functioning before you adjust anything. Lambda CIS is different from Basic in several ways and if you don't understand them, you will never get factory drivibility and performance. The way the ECU dithers the closed loop mixture is by varying the lower chamber pressure with a variable fuel pressure regulator (FV) that is really just a single EFI injector. The higher the lower chamber pressure, the leaner the mixture. Increase the duty cycle of the FV, more pressure is bled off, and the mixture is richer. Unfortunately, when the system fails, (fuse, ECU, FV failure) no pressure is bled off (blocked) and the mixture is too lean everywhere. Compensating with the mixture screw cannot give you the correct mixture curve. If you have found the test port and cannot get a reading with the dwell meter hooked up , try a different dwell meter. When the engine is cold, the mixture is irrelevent and you should get the default FV pulse (58 dwell, 4 cyl scale). This is your first step, if you can't get there, fix the problem. Once you know the ECU and FV are working correctly, drive the car for 10 miles, hook up the meter and check the closed loop mixture. If the needle is not dithering, try disconnecting the wire to the 15C switch (Bentley is wrong here). Adjust the mixture screw until the needle dithers rhythmically between 30-40 dwell, making the final adjustment in the rich direction. In closed loop, you are reading the correction factor, a lower duty cycle means the mixture is rich and the FV is raising the lower chamber pressure to lean it to stoich (just the opposite). When adjusting the ignition timing, disconnect and plug both vacuum hoses at the distributor, and set the distributor to 25 BTDC @ 3000 rpm. Idle timing @ 950 rpm should be around 5 BTDC, if the distributor is working correctly. Reconnect the orange vacuum advance hose to the front, but leave the grey retard line disconnected and plugged. Reset the idle speed to 950 rpm with the thumbscrew.

Thanks psalt. The Dwell meter I have only has 4cyl and 8cyl scales (Acteon). My understanding is you just multiply by 2 on the 8cyl scale to get the 4 cyl scale?

I know the FV and ECU are working because when I lift the sensor plate w/ the ignition on (car off) you can hear it buzz immediately.

I'll try the cold dwell right now and report back. Thanks!

[Nickshu]

Cold dwell is fixed at 18. (9 on 8cyl scale)

When warm the dwell swings from 12-18 (6-9 on the 8cyl scale).

I tried leaning it out some but the car started sputtering after about 1/4 turn on the mix so I turned it back. When it was leaned out there was very little change on the dwell.

Wondering if my dwell meter is junk. I bought it new at my FLAPS but it is not an expensive one.

[psalt]

Nick,

Sounds like you are not getting a good reading, the closed loop reading should change with a tiny movement of the mixture screw. Make sure you are getting a good connection with the green/white wire and not touching the other two pins. If the dwell meter only has two connections and no +12v, others have had problems with it. See if you can borrow a meter with three connections and test and use the fuse panel next to the test connector for +12v and the engine for ground.

[Nickshu]

My meter has only two wires, one green (to the plug) and one black to ground.

I have searched Ebay and the like for a three wire version but cannot find one. I asked my local FLAPS if they rent them but they don't even carry them at all.

I was able to set the mixture as Peter Z recommended. Set to 0.7% CO with the O2 sensor disconnected using my CO meter. Then when it was reconnected mixture was 1.2% CO. Seems to run fine at that.

Any leads on a three wire dwell meter...where to find one?

[psalt]

Nick,

Something like this is all you need.

Sears Engine Analyzer for 12 Volt Ignition Systems DIY Tune UPS Solid State | eBay

[Nickshu]

Cool. Thanks! Just bid on that one.

Now...anyone else who is reading this thread please don't bid against me!!

[mca]

I also have a 82 SC with SSIs but I don't run with my O2 sensor connected. I never liked the "hitchy" feel I got from it - those small corrections that I noticed while cruising. Keep in mind that I also have 9.5:1 compression ratio and 964 cams so my car certainly isn't stock.

At idle I set my CO to 2.5% (13.56 afr) with ethanol-free gas. I've found this to be ideal for my car. Here's why ... while cruising my afr is anywhere between 14.4 and 14.7 with no "hitchy" feel and at wide open throttle (WOT) I'm at 12.5-12.8. These ranges are based on numerous runs I did on a 75 degree day while logging the data via my LM2. Also, my highway mpg is right around 25mpg at normal highway speeds. On my recent cross country trip where we were running at around 80 mph, I was seeing around 22 mpg.

One thing I've been wondering about lately:

For gasoline fuel, the stoichiometric air–fuel mixture is approximately 13:1, but E.P.A. regulations raised the ratio to 14.7:1 to allow the use of catalytic converters.

The E.P.A raised the ratio. This does not mean it is ideal. Therefore if the real stoichiometric air-fuel mixture is 13:1 (3.7% CO) ... this (comparatively rich condition) wouldn't "wash down the bores and dilute the oil" as Psalt suggests. Am I understanding this right?

These guys know a lot more about this stuff than I do so I'd certainly suggest following their advice. I'm just a weekend mechanic who likes to tinker. But I do feel like my car is dialed in pretty perfectly - I am getting great mileage and great performance.

[Nickshu]

This might work for your car with its upgrades but for most folks I would not recommend driving a lambda CIS car with the O2 sensor disconnected. There has been much discussion here about this. One of the most misunderstood things about lambda CIS is that it's not just K-basic with an O2 sensor added. There were several changes to the system for the O2 sensor setup.

Here's some issues with it:
-It reduces the HP output of the car. This has been verified in dyno runs (there's an old post here)
-You lose WOT enrichment via the throttle position sensor
-It's often done to mask vacuum leaks which are the true source of your "hitchy" feel and inability to get the mix set right with the O2 sensor plugged in.

Sure it still runs Ok with the O2 sensor unplugged, and IF your car has lots of vacuum leaks and other issues then it will often run better with the sensor unplugged. But this is just a bandaid.

....I've been down this road. Drove mine for several years with it unplugged. Then I sorted everything out and hooked it back up and it was like getting an extra 20HP.

But it's your car and you can do what you want. Just sharing my experiences in 12 years of owning and tinkering with my 82SC.

[don gilbert]

I've always talked my group into letting me sort their lambda systems. They are leary, afraid they will loose hp.??? The 02 system only works below the 35 degree throttle switch. (idling and light throttle cruising. After it passed the switch, system goes into closed loop, with freq. valve going to full duty cycle, thus inriching the mix. If you are getting pulsations at cruising,,,, you can usually tweek the mix a little, you are alway going to get pulsations at idle, just the way the system is, but I kind of like it,,lets me know my 02 is working its butt of to save me gas at idle and cruising.

[psalt]

For gasoline fuel, the stoichiometric air–fuel mixture is approximately 13:1, but E.P.A. regulations raised the ratio to 14.7:1 to allow the use of catalytic converters.

The E.P.A raised the ratio. This does not mean it is ideal. Therefore if the real stoichiometric air-fuel mixture is 13:1 (3.7% CO) ... this (comparatively rich condition) wouldn't "wash down the bores and dilute the oil" as Psalt suggests. Am I understanding this right?


No, what you are reading is nonsense, the EPA can not change fuel chemistry.

If you want to learn about this subject I suggest starting with Bosch's "Gasoline Engine Management".

[mca]

Nickshu,

Can you share the link that shows increased HP from a O2 sensor? I've never heard this before.

Thanks,
Craig

[mca]

No, what you are reading is nonsense, the EPA can not change fuel chemistry.



If you want to learn about this subject I suggest starting with Bosch's "Gasoline Engine Management".

Well dang. Someone needs to update the Wiki page regarding afr.
Air