1983 911SC - lambda question - Missing 02 harness
 

Hi guys, I decided to try to replace the 02 sensor in my car today to make sure everything was as close to possible as baselined before I tried some enrichment tuning with a dwell meter. Much to my surprise, there is nothing attached to the engine side of the 02 wire, just the large black female plug. I inspected the harness and it appears someone may have tried to eliminate the wiring, though I also have no idea what I'm supposed to be looking for. Interestingly, the test port is still there. Can someone point me in the right direction of where to start?

I believe the frequency valve is still working as it definitely runs rougher when I unhook the relay under the passenger seat. I'm not sure about full throttle enrichment - is there a way to test for this? I haven't had any problem starting the car, whether hot or cold (recent temps in 40s). It fires up, idles around 1750 for a moment, drops a few hundred more and settles around 1000 (which is where I have it set) after 3 or 5 minutes.

Prior to starting this project, I hooked the dwell meter to the green/white pin on the lambda test port. I got a reading of 10 on a very, very old Actron (not sure if this means 90 percent duty cycle?)CP7605 meter, and the car seemed to be running richer based off my nose. There was a distinct change in the sound of the engine when I grounded and ungrounded the dwell meter. Any thoughts on this?

I've read quite a bit on the forum and suspect know I'm looking for the single white plug (which has probably crumbled) connected to a green wire. Can someone point me in the direction of this wire? Will it be coming through the rear firewall, or is it near the test port in the engine compartment fuse panel?

Thanks!

Thanks to additional posts, I found what's left of the plug clamped to the frame rail under the fuel filter. Looks like it will be a fun one to fix!

There are a couple of guys that were making the connector. Likely find their names in the O2 wire threads. (Can’t remember right now)

Louie85 used to make the connector. Don’t know if he still does.

Thanks, guys. I found Louie85 who is making a batch soon. Any thoughts on the swell meter reading? Seems extremely low but I suppose my meter could be junk also. With the o2 sensor unplugged, I should be around 50 percent with a warm sensor, right? What about the fact it
the mixture seems to riches when the meter is connected?

Thanks for your help!

With O2 unplugged, the sensor has no impact on the Lambda system.

You will be at the default duty cycle if your remain at idle. I think that would be 50%/45 dwell but I am not sure if the factory defaults a touch on the rich side.

There are temp switches that will richen the mixture (increase duty) when cold. Let it run a while then check out dwell.

Also off idle switch, and WOT switch will change duty as well. I think there is a 30% throttle switch, too.

With respect to changing idle. Are you absolutely sure you are not touching one of the other test leads with your probe/alligator clip?

The Actron is a little wierd. Bob and I actually tested the same Actron and it worked for him and not for me so we suspected something is a little different on the signal output on the test pin for the 82 and 83's. I could never get a reading over 10 also. You can still hook it up to a digital duty cycle meter or even better an oscilloscope and see what's going on. A lot of the better multimeters like a Craftsman have this feature. One thing to look for is that some of the duty cycle meters trigger on different parts of the wave so you might need to take your measurement as 100 minus what you are reading to get your true number. If you have access to an oscilloscope you can just capture what the square wave actually looks like and easily figure out your duty cycle.

If it's at operating temperature and the O2 is plugged in you should see a very slight dither on the tach (we're talking maybe 50 or 100 rpm). If that's happening then the electronics are adjusting for the stoich as it should. If you are way off on the base mixture it can't adjust so it will peg the duty cycle scale in one direction.

You can get a hand held digital oscilloscope for under $100, or one to work with a lap top, and it will be adequate for the job.

http://forums.pelicanparts.com/uploa...1542684876.JPG
http://forums.pelicanparts.com/uploa...1542684876.jpg

Not Junk at all, just maybe because many dwell meters are built differently. I also tried once a simple dwell meter and got a reading of 10-20 which was totally odd. When I used a self powered dwell meter It showed correct reading.

As mentioned by Bob, if the sensor is unplugged then it doesn't matter if the sensor is warm or cold. Then it goes into open loop mode, means when engine temp is above 15°C, then the "duty cycle" is 50% (not the dwell). The dwell meter then shows 45.
If the engine temp is below 15°C and the sensor is disconnected the duty cycle is 65% (means about 58 dwell). Also when accelerating or at WOT (when throttle is open beyond 30%) then also the duty cycle changes to 65%.


You can also fix the plug/socket of the O2 connector in the rear left engine compartment by simply changing it to a BNC style one.

https://www.koax24.de/fileadmin/_mig...gruppe_000.jpg

Its even more comfortable when pulling off the plug for O2 adjusting etc.

Guys, thanks for the excellent information!

With respect to the change in idle (and I think enrichment) with the meter connected, there’s definitely a possibility the alligator clip could have been contacting one of the other pins. I’ll check again, and also check my multimeter for a duty cycle setting.

With respect to the o2 harness connection, was there ever a consensus on the shielding and connection of the o2 sensor to the female plug, and of how to hook up the new male plug to the harness?

Thanks again guys!

I like the idea of a BNC for its ease of use. But why am I remembering that the wire from the connector to the sensor is just a single wire? The male connector end doesn't look like it is anything other than a single rod, not like audio jacks and the like which have tip and ground,, or tip, ring, and ground? The stock sensor gets its ground from being screwed into the exhaust, doesn't it?

If you purchase an aftermarket sensor, you have to cut the wire and splice it to the Porsche wiring, as the aftermarket sensor just has a single wire, and the Porsche connector looks like kind of an odd ball - I don't know where you could buy just the male connector (or the fixed female part, for that matter).

The Compete wire from the ECU up to the Plug in the left rear engine compartment is a simple shielded wire, where the shield ends in the plug but the single signal wire goes ahead through the socket to the sensor.

So BNC is a very nice and reliable option.