83 SC - another surging prob with a twist

[va_alfa]

I bought an 83 SC Cab a few weeks ago; flew out to St Louis and drove it home to Virginia over two days. This is my first CIS car, the car seemed to run fine and I don't recall any starting or surging issues. Clock, interior lights were inop at pick up.

Fast forward to yesterday, I was troubleshooting a few odds and ends including the non-working clock and interior lights. Found fuse 18 blown and replaced it; wallah, clock and interior lights now work!

I took the car out for a drive, nice sunny day about 68 degrees. Started up just fine, maybe a little surging at idle, didn't think much of it. Came to the first stop light and pushed in the clutch, idle drops almost to zero but recovers. Mmmmm something is not right.

To make a long story short, after driving for about 45 minutes I turn off the car. Then restart about 10 minutes later and it won't hold idle. The car starts but immediately dies. I try again with the throttle and can keep it running but as soon as I take my foot off the gas, dies again. I take it out for another drive, nursing it at idle and after about 5 minutes or so will idle again. Turn it off and back on, problem starts over.

I have searched this board for info but mostly find cars that run great with fuse 18 good and the Oxy sensor relay in operating condition. In this case, it seems that when fuse 18 is blown, my car runs and with fuse 18 good, start/idle problems exist?


Added: I have not confirmed that the oxy sensor relay is good/bad, I will try to do this today by starting the car, idling, and pulling the relay looking for a change in condition.



Calling all CIS experts....

[mreid]

Did you pull fuse 18 and verify? These are old cars and in my 1974, many wires were "repurposed". Here is a link to your schematic: http://www.pelicanparts.com/911/911_parts/Electrical/911_electrical_78SC_USA.pdf

The clock is Y on page 3. Perhaps trace the wires and see if fuse 18 powers something else or see if someone repurposed one of the wires. You never know...

[kodioneill]

Welcome to the jungle. These car are finicky you'll have to be patient with it but it may be a minor tune up issue or vacuum leak fuel pressure etc...

[SCadaddle]

Certainly no CIS Expert here, but sounds like what was going on with my 82SC when I first acquired it. The Oxygen Sensor relay and Lambda box under the passenger seat which control the frequency valve (FV) were inoperable and the previous owner(s) had adjusted the CIS mixture control to run without it. I bet that now yours is in play the mixture is just way off.

Take a number, have a seat. Tony will be here shortly.

[Bob Kontak]

Originally Posted by SCadaddle

Take a number, have a seat. Tony will be here shortly.

LOL

Agree with your assessment.

OP - Is the fuse still ok?

[va_alfa]

Okay, here's an update.....

Bob, the fuse is still okay.

Engine is cold.
I started the car with fuse 18 in place and the oxy sensor relay (OSR) installed. Car started fine and without drama.

I pulled out the OSR and there was a change in the engine tone but no surging. Throttle response was sluggish from idle.

I replaced the OSR and a change again in tone, Throttle response is spirited from idle.

The OSR is clicking when I take it out and put it in, so it must be functioning correctly.

Taking out the fuse and replacing it seemed to have the same effect as the OSR.

The car is warming up now and the characteristics are changing and effects are more dramatic.
In fact, the car was idling with the fuse and OSR in place and suddenly the car just died; no excessive surging, just died. An inexperienced observer (me) would think that maybe a temperature switch "clicked" the control into warm running mode or something. Anyway, read on....
With the fuse and OSR still in place, the car will no longer start without nursing the throttle. I start the car with throttle assistance and bring the RPM's down to 1000 gently and then release the throttle, the car idles for about 5 seconds, then surging begins, the oscillations get larger until the engine dies. Total time is about 20 seconds. If I drop my foot off the throttle at 3000 rpm, the car dies.

I remove the OSR, start the car again without throttle assistance; car starts right up and idles smoothly at 700rpm. While the car is idling, I insert the OSR relay and the car dies immediately.

I restarted the car with the OSR pulled, started right up without throttle assistance and pulled it in the garage. Throttle response is sluggish.
What's next.

Thanks in advance for the help!

Oh, and one more thing, when I came back into the house, my wife complained that my clothes smelled like I've been playing with the "old cars". In other words, car is running rich.

[Bob Kontak]

If you are ready for a mini-project help is here.

As noted, adjustments have probably been made to your AF ratio to offset the "dead" O2 system.

I would take the fuse out for the moment. Won't hurt anything and the car can be driven.

You will need an analog dwell meter if you don't have one. Maybe $50. Actron.

Find the test wire near the aluminum tray for all the relays in the engine compartment. Driver's side rear area. It has a cap on it and you will use the prong with the green wire leading to it to test the frequency valve operation.

What the test will do is tell you what the frequency valve is doing to offset a rich or lean mixture when the O2 system is in play. You adjust the CO/AF screw, using the dwell meter's output to dial it in.

I will be glad to speak with you offline as there are many, many viewpoints on this method and the equipment.

I will PM you my phone number and help you if you are interested. If it works, great and if it doesn't, you can post and say I am full of it.

The reason I am offering the offline help is that the test is weird, it's backwards from logic and there are so many opinions that it can get confusing and water down the clarity this method can provide.

PS: I like the Alfa. Not a 58 Testa Rossa, if you like sleekness, but way cool.

Check your PM.

[tirwin]

The OXS relay is on the same circuit as the interior lights. No interior lights, no relay. You fixed the lights, mixture probably needs to be adjusted back to normal. PO probably adjusted it not realizing the problems are related.

Listen to Bob.

My only note I would add is Jim Williams noticed the Actron dwell meters are not always accurate. I have one and I don't trust it. I suspect they have some quality control problems because Bob's seems to work and mine seem to make a good doorstop. If you don't have a DVM with a duty cycle function, invest in one. You can get an Innova 3340 for about $70 bucks and it is good for more than one thing.

Hook it up to the green/white wire on the test port Bob mentioned and a ground. 50% duty cycle is stoichiometric. Dial it in to 40-45% if you want it to run a little rich. Lower duty cycle on the DVM reading = rich, higher = lean. 5% is rich stop, 95% is lean stop. Where people get confused (as I did) is that the books say rich stop is 95%, which is technically correct. But the duty cycle function on the Innova DVM is reading the percentage the opposite way. It is confusing until you get your brain wrapped around the fact that it's just a difference in theory and how the equipment works. It's like saying the door was closed 25% of the time versus saying it was open 75% of the time. Same thing.

[Bob Kontak]

Originally Posted by tirwin

But the duty cycle function on the Innova DVM is reading the percentage the opposite way. It is confusing until you get your brain wrapped around the fact that it's just a difference in theory and how the equipment works. It's like saying the door was closed 25% of the time versus saying it was open 75% of the time. Same thing.

Bingo! Thanks.

That's where I figure a phone call can save a lot of typing.

The dwell meter will show that the frequency valve is fighting a rich mixture by delivering a lean duty cycle.

I did it backwards for years until psalt tuned me up, but thankfully, one can still dial it in to the correct range if you trust the meter.

Then you wont stink like an old car. :-)

[va_alfa]

Thanks Bob, I have your phone number. I will call you this week to talk about the process going forward. I'll be finding a dwell meter and 3mm hex tool in the mean time.
Alex

[tirwin]

Originally Posted by Bob Kontak

Bingo! Thanks.

That's where I figure a phone call can save a lot of typing.

The dwell meter will show that the frequency valve is fighting a rich mixture by delivering a lean duty cycle.

I did it backwards for years until psalt tuned me up, but thankfully, one can still dial it in to the correct range if you trust the meter.

Then you wont stink like an old car. :-)

psalt tuned you up... you tuned me up... guess it's time for me to pay it forward.

[va_alfa]

First, thank you Bob and Tim for your instructions helping me troubleshoot my CIS problem. I spoke with Bob on the phone for about 20 minutes and took about a page-and-a-half of notes. The conversation gave me a deeper understanding of theory of operation and the purpose of the subsystems. And your words gave me confidence to make this adjustment on my own. For those considering making this adjustment, I would also recommend reading the Factory Service Manuals and Dempsey's "101 Projects..." books to round out your research and understanding.

I borrowed a vintage Sun Tach Dwell meter and bought a 3mm allen from Ace Hardware; I had some time this afternoon to tinker with the tuning. The executive summary is that the adjustments made significant improvement to the idle and performance.

The details are as follows:
I removed the engine bay fuse cover and began searching for the connector for where the Tach Dwell meter will read the oxy sensor. I probably searched for "the green wire" for quite some time not really knowing what I was looking for. I finally found it (see photo), remembering the description that Bob gave me: "it's a connector that is about the size of your pinky finger."



After finding the correct wire pin, I started the car and connected the Tach Dwell meter but the meter still read zero. ...huh, why isn't this thing working..........does the meter need a battery?........dammit, I forgot to plug in the oxy sensor relay under the passenger seat... I plugged in the relay and got a reading of about 12; exactly as Bob had predicted.



Using the 3mm hex tool, and following Bob and Tim's instructions, I adjusted the mixture counter-clockwise to lean it out; turned it about 1/4 turn total. ...I practiced getting the tool in and out of the adjusting screw the night before, something that was not easy at first since I had no idea what to expect. After a few tries, finding the screw is easy... The adjustment definitely made a positive difference in the idle and throttle response.



I continued to make adjustments to the A/F mixture and monitored results. Bob told me on the phone that I would begin seeing the needle fluctuate approximately 10-15 degrees indicating the lambda system is controlling the mixture, I could now see this oscillation happening at a frequency of about 1 cycle per 2-3 seconds. Idle was very smooth; blipped the throttle and the idle fell right in without any over/under shoot; throttle response is peppy.

But I was still not at the target 40-45 degrees on the meter as I was instructed to strive for, so I continued to make very small counter clockwise adjustments to the mixture screw. The idle began to oscillate rapidly at a frequency of about 2 cycles per second and the idle did not sound as smooth. .....mmmm did I adjust too far.... I turned the adjustment screw clockwise to enrich the mixture and idle returned to very smooth with the lambda sensor controlling the mixture.

Along the way, I made adjustments to the idle speed; starting at about 600rpm and adjusting up to the target 950 rpm.



Satisfied with the results so far, I buttoned it back up and took it out for a test drive. Idles smooth, pulls strong to redline, no over/under shooting of engine speed as the car returns to idle.

Interested in comments about the rapid oscillating idle speed when I leaned the mixture further. Did I go too far (too lean) before I turned back or is there a "new and better world" on the other side.

[diy83sc]

Keep up what you are doing. Just adding my setup.

We still have cool weather where I am. Depending on your weather, be sure to get the car to full operating temperature before you make any adjustments.

I use the dwell meter the same as described and have my idle mixture dialed so the needle shows a slight oscillation between 28 and 29%. A few degrees either way and it's clearly rich or lean. The needle has a wider swing as the mixture is adjusted by the controller. The dwell meter helps narrow the setting and provides for a good baseline the next time I connect the meter. Before I set it, I checked my ignition timing and got my idle as close to 950 as I could.

[T77911S]

bob, good advise, but he still needs to verify fuel pressures.

[psalt]

Interested in comments about the rapid oscillating idle speed when I leaned the mixture further. Did I go too far (too lean) before I turned back or is there a "new and better world" on the other side.

Alex,

Yes, my experience is these cars run best with the dwell fluctuating below 40. There is no "better world" past the dark side. The late lambda SC's also cold start much stronger and run much cooler at idle in summer with the vacuum retard disconnected and blocked and the timing reset. The open loop WOT mixture has a lot to do with how much high speed timing you can run. Porsche chose to detune the ignition timing on the late high compression US SC because they were forced to spec it for 87 CLC fuel and they wanted to advertise improved mpg. If you run premium fuel and the engine is in good condition, you can usually run a few more degrees than 25 BTDC max.

Paul

[garment]

Just sent you a PM, pls have a look...

Originally Posted by va_alfa

First, thank you Bob and Tim for your instructions helping me troubleshoot my CIS problem. I spoke with Bob on the phone for about 20 minutes and took about a page-and-a-half of notes. The conversation gave me a deeper understanding of theory of operation and the purpose of the subsystems. And your words gave me confidence to make this adjustment on my own. For those considering making this adjustment, I would also recommend reading the Factory Service Manuals and Dempsey's "101 Projects..." books to round out your research and understanding.

I borrowed a vintage Sun Tach Dwell meter and bought a 3mm allen from Ace Hardware; I had some time this afternoon to tinker with the tuning. The executive summary is that the adjustments made significant improvement to the idle and performance.

The details are as follows:
I removed the engine bay fuse cover and began searching for the connector for where the Tach Dwell meter will read the oxy sensor. I probably searched for "the green wire" for quite some time not really knowing what I was looking for. I finally found it (see photo), remembering the description that Bob gave me: "it's a connector that is about the size of your pinky finger."



After finding the correct wire pin, I started the car and connected the Tach Dwell meter but the meter still read zero. ...huh, why isn't this thing working..........does the meter need a battery?........dammit, I forgot to plug in the oxy sensor relay under the passenger seat... I plugged in the relay and got a reading of about 12; exactly as Bob had predicted.



Using the 3mm hex tool, and following Bob and Tim's instructions, I adjusted the mixture counter-clockwise to lean it out; turned it about 1/4 turn total. ...I practiced getting the tool in and out of the adjusting screw the night before, something that was not easy at first since I had no idea what to expect. After a few tries, finding the screw is easy... The adjustment definitely made a positive difference in the idle and throttle response.



I continued to make adjustments to the A/F mixture and monitored results. Bob told me on the phone that I would begin seeing the needle fluctuate approximately 10-15 degrees indicating the lambda system is controlling the mixture, I could now see this oscillation happening at a frequency of about 1 cycle per 2-3 seconds. Idle was very smooth; blipped the throttle and the idle fell right in without any over/under shoot; throttle response is peppy.

But I was still not at the target 40-45 degrees on the meter as I was instructed to strive for, so I continued to make very small counter clockwise adjustments to the mixture screw. The idle began to oscillate rapidly at a frequency of about 2 cycles per second and the idle did not sound as smooth. .....mmmm did I adjust too far.... I turned the adjustment screw clockwise to enrich the mixture and idle returned to very smooth with the lambda sensor controlling the mixture.

Along the way, I made adjustments to the idle speed; starting at about 600rpm and adjusting up to the target 950 rpm.



Satisfied with the results so far, I buttoned it back up and took it out for a test drive. Idles smooth, pulls strong to redline, no over/under shooting of engine speed as the car returns to idle.

Interested in comments about the rapid oscillating idle speed when I leaned the mixture further. Did I go too far (too lean) before I turned back or is there a "new and better world" on the other side.

[tirwin]

va_alfa,

Glad you got it sorted out. Now go put that top down and drive it like you stole it!

[dfk911sc]

+1 great info

[7thPorsch]

This article is awesome and made my weekend, seriously, I really appreciate the tips. I have an 83 SC Targa that has been running great for 6 months. The shifting problems lead to a missing transmission bolt. Yes it was gone and she had only 1 of the 2 transmission bolts holding the tranny to the car. I had to helicoil the body and all is well, shifting is back to its usual self with 2 installed tranny bolts. However, now my car is exhibiting the symptoms that started this thread, ie runs great but will not idle when warmed up. I must have tweeked a hose when I finally replaced the transmission mounts ? This makes no sense ?

I check my dwell angle and its 14. I do have a cat with cat fuse (interior lights work) and cat relay installed. Should I also just change the AF to 45 degrees even with a cat installed ?