81 SC O2 sensor issue

[aadrew10]

Hi Everyone,

I recently finished a head stud replacement, and have a small issue going on that I think is with the o2 sensor. The engine runs great and has lots of power, idles just below 1k rpm steadily, however I feel a surge from the engine when cruising at a constant speed anywhere from 2.5 - 5k rpms. As I push the gas pedal down the car accelerates, but still surges. It feels like once I get the throttle to 30% and the WOT throttle switch clicks over, more power comes on and the surge is gone. This surge in power feels almost like a turbo is kicking in. I think this increase in power is due to the frequency valve ignoring the O2 sensor and enriching the mixture.

I unhooked the O2 sensor, and the surging at a steady speed is gone. The power delivery is much smoother as well. I have a new O2 sensor and an old one, and switching between the two yields the same results. I am using a regular 1 wire bosch O2 sensor. I also have done away with the fancy O2 wire connection in the engine compartment since it had disintegrated.

It looks like people's opinions are that running the car with the O2 sensor unhooked is fine. However, if possible I would like to get the O2 sensor and related gear working properly to get better gas mileage while cruising.

I have an LM1 and the meter read between 14 and 15 air/fuel ratio. During hard acceleration it drops to 12. I haven't measured it without the O2 sensor connected.

To me it looks like I need to get the duty cycle of the frequency valve higher to run a bit more rich below 30% throttle. Does anyone know how to tune an O2 sensor system? Is that even possible? I wonder if switching brands of O2 sensors might have any effect.

Any thoughts are appreciated. I might be all wrong with my ideas here. I'm new to Porsches.

- Andrew

[James Brown]

you should run the o2 sensor but sounds like the lambada system working or the CIS is "tuned" to run without one. How did you connect the O2 sensor wire after the original one broke?

[psalt]

Andrew,

What is the duty cycle at hot idle closed loop ?

[aadrew10]

Sorry for the delay guys.

James,

I crimped the o2 wire directly to the wire that the connector came from. I covered everything with shrink wrap.

Paul,

On cold start, duty cycle is 35%. When well warmed up, at idle, the duty cycle is 87-89. I'm using a craftsman professional digital auto-ranging multimeter with duty cycle function. Could these readings be opposite the correct values? I say that because I expected around 65% on cold start. Hooking up the negative probe to the test lead and the positive probe to the engine gave 65% while cold.

- Andrew

[psalt]

Andrew,

You cannot get a good reading with a digital meter, you need to use an analog meter. The O2wire is shielded to protect the <1 volt signal, if you destroyed the shield, the signal may be degraded.

[aadrew10]

I'll try and find an analog meter. There was only shielding up to the o2 connector in the engine bay, then it became a single black wire between the connector and the o2 sensor. I removed the connector and crimped the single black wire to the shielded wire, being careful not to connect any of the shielding.

[aadrew10]

Hey Paul,

I'm not sure how to read the dwell meter, so here's a picture of it. The yellow line is where the needle was at warm idle. I suppose the lambda computer is trying to lean the mixture out a lot?

- Andrew

[psalt]

Andrew,

You are trying to read the duty cycle (time open). Dwell is a 90% scale and you need to convert to duty cycle by dividing by 90% of the 4 cylinder dwell. Your meter only has the 8 cylinder scale , so you need to double that and divide by 0.90. The yellow line looks like it is at 5 (5x2/.90=11). I need to know if the reading was cold, hot, open, loop or closed loop to give you more info, but 11 is never a good number.

You should see a steady 29 in that meter after start up, then a regular, fluctuating meter between 20 and 30 after it goes closed loop. Most car run best when adjusted to 15-20 (closed loop) on your meter.

[aadrew10]

Hey Paul,

When I first start the car the duty cycle is at 65%. After getting it all nice and warm at idle the duty cycle is around 11. I checked this twice, once after driving about ten minutes, and another after driving a half hour.

Since it was night time, while hooking up the dwell meter in the dark, it was quite a light show with my spark plug wires. I could pick out sparking in about five places on the left side only. I just replaced my original shielded spark plug wires with some non-shielded Berus from SGS Porsche in San Luis about 6 weeks ago. Could this be causing a weak spark which in turn is causing the duty cycle to drop?

- Andrew

[psalt]

Andrew,

Is it a steady 11 or is it fluctuating ? If it is not fluctuating, you need to figure out why it is not going closed loop. First troubleshoot the main temp switch on the right, then check the 02 sensor output, then check the lean and rich stops.

[dshepp806]

I think a digital VOM would work, as long as it's true RMS capable.....?

[aadrew10]

Paul,

The 11% was steady. My temperature switch checked out OK.

Here is a breakdown of the tests I performed with the car warmed up and idling:

O2 sensor connected:
Duty cycle at 11%.
O2 wire voltage between .8 and .6 volts.

O2 sensor unplugged:
Voltage at O2 sensor .9 volts.
Voltage at O2 connection from lambda box: .515 volts.
Duty cycle 47%

Rich stop:
O2 sensor unplugged and O2 connector grounded: Duty cycle crept up slowly to 78%. Then settled at 44%.


Lean stop:
O2 sensor unplugged and O2 connector at +1.5 volts: Duty cycle 11%

- Andrew

[tuned6]

Originally Posted by aadrew10

Hey Paul,

When I first start the car the duty cycle is at 65%. After getting it all nice and warm at idle the duty cycle is around 11. I checked this twice, once after driving about ten minutes, and another after driving a half hour.

Since it was night time, while hooking up the dwell meter in the dark, it was quite a light show with my spark plug wires. I could pick out sparking in about five places on the left side only. I just replaced my original shielded spark plug wires with some non-shielded Berus from SGS Porsche in San Luis about 6 weeks ago. Could this be causing a weak spark which in turn is causing the duty cycle to drop?

- Andrew

Replace your plug wires and try again!
You should not see arching from your wires. I'm guessing you have ignition problems and your disconnecting the lambda is in fact correcting or stabilizing the mixture so it feels smooth again but without the performance, proper metering... maybe the lambda is recording the miss and running the system lean to comp and giving you poor performance.
Your Lambda reads the 3 cylinders. I think a total miss will throw huge hydrocarbons into the exhaust which will make your o2 sensor lean out the mixture for the next turn, maybe way too much, giving you a lean miss and bouncing back and forth to no end until you get up in rpm and the ignition problems mask themselves.
Ignition problems tend to go away, become masked, at higher engine speeds when the spark weakens out... which in turn is burning all of your mixture so the lambda corrects properly and there is your turbo feeling..

[fireant911]

Originally Posted by aadrew10

...I crimped the o2 wire directly to the wire that the connector came from. I covered everything with shrink wrap.

I suspect that may very well be the problem. The wire is a coaxial wire with the signal wire in the middle surrounded by the grounding wire separated by a layer of insulation. If you attached to any of the grounding wire the signal is lost.

A doubt that the arcing you are seeing from your spark plug wires are to blame. I replaced my arcing and sparking shielded OEM wires with new shielded OEM wires and arcing and sparking was still occurring.

[aadrew10]

If my O2 sensor is showing .9 volts with it disconnected at 50% duty cycle, I would think this points to the idle mixture being set too rich. A guess would be that even with the lambda brain trying to lean the mixture out, the duty cycle stays at 11% because the mixture is still too rich. This is me thinking out loud so please feel free to chime in.

However, when I did the lean stop test, the engine idle slowed to 500-600 rpm, and didn't seem too happy. It idles much better when the O2 sensor is disconnected and the mixture is richened by the 50% duty cycle. Could the O2 sensor be the cause of this problem by giving an artificially high voltage? When I checked my afr mixture with an LM1, it was constantly around 14-15/1, yet the O2 sensor indicates a rich mixture (.8-.9 volts) causing the lambda brain to try and lean it out. My suspicion here is that the O2 sensor is not indicating the true fuel air mixture.

Also, I was careful to keep the shielding of the O2 wire separate, and I would think that if the shielding had gotten spliced into the signal wire, I would not have been able to do the lean and rich stop tests because the signal wire would have been grounded.

Any thoughts would be appreciated. I want to make sure I completely understand what is causing this problem before I try turning any screws.


Thanks,
- Andrew

[don gilbert]

Andrew, i recently tuned a car sim. to your prob. (could not get correct duty cycle, 02 was always trying to lean it out, mixture screw would not help, turned out once i put the gauges on it, the warm control press. was 41, way to low for a warm engine, causing it to run rich at idle. raised prss up to 55, everything straitened out, good duty cycle, car ran great. something you might check.

[psalt]

Hello Andrew,

I think you are on the right track. If the engine is hot, the sensor trips closed loop, and the duty cycle is pinned at the lean stop, the O2 sensor or the wiring could be at fault. The 0.9 at the sensor being 0.515 at the box could be a clue. Can you hear or feel the FV valve pulsing ? If the system gets a very rich reading, it should lean out the mixture and if the sensor signal was working, you should see the needle fluctuate. The O2 sensor is just that, it only knows O2, it doesn't know hydrocarbons from single malt scotch. What do the spark plugs look like ? If you have access to an LM1, measure the AFR before the cat. the factory spec works out to 14.16-14.33 AFR. I would try leaning out the mixture screw 1/4 turn, if the duty cycle does not respond, I would try a different sensor. Don's point about control pressure is a good one, the basic pressure settings need to be in range and the FV pulsing to get the correct mixture curve.

[aadrew10]

Just to clarify, when I unplugged the O2 sensor, the sensor read .9 volts. The other end of the connector read .515 volts. During this reading the O2 sensor was not connected to the box. The engine was running in open loop.

I'll post back with the control pressure readings, spark plug condition, and LM1 readings once this storm goes away.

- Andrew

[aadrew10]

I hooked up my pressure gauge today, warm control pressure read 51 psi. For about 30 seconds when I first put the fuel gauge on the duty cycle actually oscillated between 11 and 22 or so. The O2 voltage was .5-.8 volts.

I would like to bring my warm control pressure up to 55 psi. Could Don or someone give me direction on how to do this? I've been reading you can change it slightly by adding or subtracting shims to the regulator at the fuel distributor, but I've also heard this has more of a change on system pressure. I've also read about drilling into the WUR to turn a screw. I'd like to know the best way before I do anything foolish.

All my plugs looked good and consistent. The next chance I get I'm going to hook up my LM1 and get that data as well.

[don gilbert]

50-55 is in spec. i would leave it there, 55 will get you a little better mileage, i have my own homemade electronically adjustable on the fly wur, i run anywhere between 45 on a cool day (35-55 degrees) for max. performance, to 60 on the hwy for max mileage. do not mess with your shims on the primary press. regulator. Control press. is adjusted by moving the plugs in and out on you wur. there are lots of threads how to do this, but i would only recommend it to guys who really like to tune on their cars and understand completely how your k-jet system works. as you know, cis has no barometric or ambient temp sensor to adust the mix between hot and cold days, so and adjustable wur will help with that.