Engine stumble/surge when warm

[joecrum]

I've been trying to diagnose a stumble/surge issue on my '82 SC and need some help.

The problem:
When the car has been driven around 40 minutes (highway), I get a stumble and surge situation. Sort of like the fuel is getting restricted momentarily. It happens beginning around 2500 rpm. The idle is steady around 950, though sometimes it idles much lower, but still steady. It never stumbles when cold. Sometimes the problem goes away, but it seems to have slightly less power.

Before the problem:
The car has always run strong and steady. 3 weeks ago, I replaced the fuel filter, spark plugs, and air filter. After that, the car ran stronger than it had in a long time. One week later, the problem began.

Some ‘relevant’ history:
About 45k miles ago, I had replaced the Beru plug wires with the same. When I was installing them, I cracked one of the plug ends. It didn’t completely break, just a crack. I taped it up with high temperature electrical tape and it seemed to work just fine. When I recently changed the plugs, the crack propagated all the way through the connector and the wire was exposed. I did my best to tape it up and the car ran great, for a week.

Also shortly before this problem, I had to replace the battery due to a broken terminal.

What I’ve done:
When the problem started, I was SURE it was the wires due to it only happening when warm. I replaced them with Clewett wires and was sure it was solved. Once it was driven about 40 minutes (highway), the stumble/surge was back. By the way, the engine temp rarely gets above the first red mark during these 70 deg days.

I have also swapped the fuel pump relay with a black one, but it didn’t help. I’ve gone through several tanks of gas, so I don’t think it’s the gas quality.

My suspicions:
-It must have something to do with the engine being warm.
-My old fuel filter seemed quite clogged, so I wonder if the new one got quickly clogged also. I don’t know what temp has to do with that, though.
-fuel pump?
-spark plugs?
-vacuum leak?

Your help and suggestions are greatly appreciated.

[Paulporsche]

What type of CD box do you have? Some posters have said their Permatunes start to act funny when they heat up.

Also have you checked your fuel and control pressures and mixture @ that temp?

BTW where did you get the high temp electrical tape?

[ruf-porsche]

Could be the control pressure from the warmup regulator is not correct. Have you connected a set of gauge to the system to verify that the pressure is within tolerance? Could also be the auxillary air valve.

[joecrum]

Everything is stock on the car, so the CD is original.

I haven't checked any fuel pressures yet and have never touched the mix.

The 'tape' was 3M Scotch 130C Linerless Rubber Splicing Tape that I found at work. It doesn't have adhesive, but is thick and tacky and sticks to itself really well. Somehow, it worked for a while.

What does it take to check the pressure, anyway?

[mca]

Originally Posted by joecrum

the engine temp rarely gets above the first red mark during these 70 deg days.

Please say you meant white mark. Right?

What kind of ends are on the Clewett wires? Are they Beru?

I only ask b/c I put some new Beru wires on my car and got misfire (felt like stumble) at higher rpms when the engine was warm. Put the old Bosch wires back on and it was fine. Still not sure why the new wires created this situation - could be that one of the wire ends was open and it only became problematic when the engine got warm (heat cycling).

Interested to see how this pans out.

[Paulporsche]

You should put a red relay back in. The black one will eventually burn out.

You can check your fuel and control pressures w/ a CIS FP gauge. JC Whitney has one for about $60. If you search you will find lots of info on how to do this. Ricks911s has posted some videos on this.

The mixture can be checked w/ an air fuel meter or @ a shop w/ theirs. Many Pelicans have both these gauges. Perhaps someone local can let you borrow theirs.

I think you should also check your fuel pump draw when warm. Maybe it is wearing out, drawing more current, and pumping less fuel. You can also check the fuel delivery by pulling the injectors an having them spray into a container.

Is this something that has started recently, or have you had this problem since you've owned the car?

[joecrum]

Yes, the first WHITE mark.

The wire ends are Clewett also. I still sort of suspect the wires/plugs because it only happens when warm. Also, it didn't happen until I recently changed plugs, so I thought it might have something to do with that.

I did put the red relay back in. BTW, do the relays fail completely or do they slowly start to go. The car doesn't stall and always starts fine (even when warm).

The issues just started recently.

Another thought...when I got the car (about 7 years ago) the idle would rise and fall significantly and the OXS light was on. I replaced the O2 sensor and the problem went away completely. That was about 45k miles ago. The light hasn't come on again (I may have disconnected it due to the annoying ticking sound). Could a bad O2 sensor do this?

[psalt]

Hello Joe,

I am not clear from your description when this is occuring, during acceleration or during a steady 2500 cruise ? If you have a stock functioning 82SC with CIS lambda, an analog dwell meter will tell you whether your 02 sensor, ECU and temp sensors are working and if your mixture adjustment is ideal. Seach for CIS lambda and you should find everything you need to know to diagnose the system.

[Paulporsche]

The relays usually just work or don't work.

Someone w/ more knowledge of 02 sensor cars (which includes a lot of people EDIT: like psalt!) will have to advise you on this, but I THINK that when the o2 sensor isn't working, the system reverts to a full rich situation. In that case, the surging you are getting could be from a too rich condition. Since the engine requires a richer mix when cold, it may be fine then; and when the engine has warmed up, then it is too rich. Do you get any other too rich symptoms, like black plugs, sooty exhaust pipe, or idle hunting/revs dipping below 950 and then returning?

[joecrum]

psalt, it occurs at a steady cruise as well as during accelleration. Sometimes, it seems to go away on it's own, too. It happens almost every time I drive more than an hour. My commute is just under an hour and it usually feels like it's just starting to happen when I get to work (or home). I'm almost sure it will happen sooner when the weather gets warmer.

It doesn't hunt during idle, just sometimes idles really low (around 5-6k). I haven't noticed any other rich symptoms, though I will check the plugs this weekend.

Maybe I'll try disconnecting the O2 sensor next time it happens.

I realize I may have a lot of diagnosing to do to sort this out, but I keep hoping I'l find a simple 'smoking gun', since it was fine until I changed the plugs, etc.

[psalt]

My advice would be to focus on narrowing down the problem, conditions under cruise and acceleration are so different the same cause is unlikely. Simply hooking up an analog dwell meter and observing the reading for a few minutes will tell you more about the condition of the fuel system than any internet posts. Did you change the type of spark plug ? Coil failure usually is a heat related problem, swapping a known good coil would eliminate that possibility.

[psalt]

Paul wrote,

Someone w/ more knowledge of 02 sensor cars (which includes a lot of people EDIT: like psalt!) will have to advise you on this, but I THINK that when the o2 sensor isn't working, the system reverts to a full rich situation.

Hello Paul,

I can assure you that no one on this list has a complete understanding of the lambda system, the signal to noise ratio is very low. When the O2 sensor is not working, the system does not go full rich, it goes to the default setting, which is in the middle of the range. A bad O2 sensor will have no effect on cold start, cold warm up, or acceleration. Fuel mileage, emissions and converter life will be effected. When the relay or ECU is disabled, the system goes full lean, the engine will run like crap and no amount of mixture adjustment can bring back the proper fuel curve.

[joecrum]

It's hard to believe the cruise and acceleration conditions I'm experiencing are not related. When I feel the condition coming on, I'm usually just cruising steady around 3k and it starts to slightly stumble/surge. When I accelerate, the condition is more pronounced. The cruising and acceleration stumble always occur at the same time, never independently.

As suggested, I will be getting a dwell meter and start from there.

(By the way, it is so great to be able to ask questions here! I can't imagine Porsche life without all your help! THANKS!)

[Nine9six]

Originally Posted by joecrum

psalt, it occurs at a steady cruise as well as during accelleration. Sometimes, it seems to go away on it's own, too. It happens almost every time I drive more than an hour. My commute is just under an hour and it usually feels like it's just starting to happen when I get to work (or home). I'm almost sure it will happen sooner when the weather gets warmer.

It doesn't hunt during idle, just sometimes idles really low (around 5-6k). I haven't noticed any other rich symptoms, though I will check the plugs this weekend.

Maybe I'll try disconnecting the O2 sensor next time it happens.

I realize I may have a lot of diagnosing to do to sort this out, but I keep hoping I'l find a simple 'smoking gun', since it was fine until I changed the plugs, etc.

"It doesn't hunt during idle, just sometimes idles really low (around 5-6k)."

For most...Me included. 5-6K idle is just a tad high
You prolly meant 5-600 rpms, huh?
If you are truely hoping for a simple smoking gun; for grins and giggles, why not put the old plugs back in and drive it a while, and see what you can see?

[JFairman]

all your descriptions point to a lean afr after the car has warmed up.
low idle and surging during slow accelerations are indications you are running lean. could be vacuum leaks or could be the steady cruise control pressure from the WUR is a little high once warmed up.

first try disconnecting the O2 sensor plug so the lambda goes to a default steady duty cycle.
do not disconnect the plug from the frequency pulse valve or you will make the AFR's go leaner.
the pulse valve fine trims the afr's by returning fuel sytem pressure from the bottom side of the 6 seperate differential valves to the gas tank in a usa fuel head with lambda.

when the motor is cold the temp sensor in the top of the passenger side chain case cover bypasses the o2 sensor readings until the motor warms up.
once it is warmed up the lambda system is working all the time trying to keep it around 14.7:1 afr, and sometimes a little leaner during steady low load cruise accept for when you floor it.

when you floor it the full throttle position sensor switch on the throttle body shaft cuts out the o2 sensor readings and puts it into open loop mode to give the engine a little more fuel.

there's more detail to it, but i'm trying to explain it simply

[joecrum]

Right, 500-600. Good thing I don't make mistakes like these in my engineering designs!

JF, I'll try what you suggest.

Any simple way to check for vacuum leaks?

[longhornchris04]

JFairman is probably on the right track. At idle & part-throttle cruise (when warm) the lambda system keeps the AFRs correct for max cat converter efficeincy by running closed-loop. During warm-up or when the throttle is open past ~30% (i think), the lambda system is supposed to go open-loop and richen up the mixture.

This is done with a frequency valve. You should check the bently guide, but I belive the duty cycle should be around 50% for closed loop operation. This gives it the max range to enrich or lean out the mix. If your CIS is out-of tune, this will compensate.

When open loop (and I may be backwards here) it runs a fixed duty cycle of ~70%, resulting in a rich mixture. This aids in warm-up and is generally better for the engine under load. As a note, the lamba isn't the only part of CIS that provides a rich mixture during warm-up.

Now, how this could relate to your problem:

1- base mixture set incorrectly, but within the correction range of the lambda - as long as you stay closed loop (part throttle) the system provides proper ratio but once you go open loop, the fixed duty cycle is too lean.

2- TPS sensor malfunction - once at operating temperature, the system doesn't go open-loop on acceleration. The "ideal" ratio of 14.7:1 isn't good for an engine under load and can cause stumbling.

3- Vacuum leak - same deal as incorrect mixture. During warm-up its masked by the rich warm-up cycle, and at idle/light operation by the closed-loop control.

If you don't already have it, get a copy of the bently guide ($100 new, well worth it) and some tools (dwell meter to start, CIS pressure gauges are also a plus) so you can get some measurements.

Whatever you do, don't just disable the system. Without it running open-loop your engine will go way-lean and cause all kinds of trouble. If you really want to isolate the lambda, disconnect the exhaust O2 sensor, which forces open-loop and see how it behaves. Open-loop mode is supposed to be rich, but functional (i.e. not good for emissions but shouldn't be a huge detriment to driving). The catch is the system is balaced for having the frequency valve operating at all times, and disconnecting it will cause serious issues.

Good luck. (goes off to thank having non-lambda CIS, for now at least)

[Paulporsche]

psalt,

Thanks for inputting the correct data. As I hope I pointed out I wasn't sure about my comment. I knew from previous threads you were much better informed than I am about the lambda system, which is why I referenced you.

[psalt]

Any simple way to check for vacuum leaks?

Hello Joe,

The best method I've found is using an unlit propane torch with a rubber hose. Hold the hose end near the usual suspect ( injector sleeves, orings, runner boots, throttle shaft, air boot) and listen for an change in engine note. The idle can rise or drop based on the mixture setting. You can also tell if the injector sleeve orings are bad by simply wiggling the injectors, if they are original, they are leaking. Vacuum leaks have a big effect at idle, a smaller effect under light throttle cruise and virtually no effect at WOT. The difference between cruise and acceleration is pronounced, a fuel delivery problem will show up under acceleration and disappear under light throttle. Also cylinder pressure at cruise is low and weak ignition shows up when you open the throttle and increase cylinder pressure. I would focus on the changes you made, especially the plugs and wires, before going fishing.

[Paulporsche]

I recently found 3 things wrong w/ my vacuum system:

1 A hole in the metal pipe leading to the AAR and AAV. It was rubbing on the top of the oil light sender and the crankcase cover.

2 An open port on the backside (toward firewall) of the throttle body

3 An atmo line instead of a vac line attached to the thermovalve.

I mention these because they are not usually mentioned in lists of typical vacuum leak locations.