Bill Ball

So, what you describe is after warming up, the car will die, and during the process it acts initially like there is a rev limiter.

When you try to restart it without waiting 30 minutes, how exactly does it behave? You say it will die if you drive off. Does it die immediately after releasing the key from the start position? Does it idle? Does it idle but poorly?

I have seen problems from fuel overheating, but not in stock S4s - just a heavily modified car with a fuel cell, and that was only after extended driving at high altitiude (>5000 feet. How high is Vail? Sounds high.). As the fuel heated up, approaching 135F even on 80F days, the external Bosch 044 would slowly fail to maintain fuel pressure. We had remote fuel temperature and fuel pressure gauges, so we could watch it happen. The pump would get progressively noisier and the pressure would fall slowly over a couple of minutes...50...45...40..35...30..DIE at 135F fuel temp in the tank. Also, with our wideband O2 we could see the mixture lean out. The injectors would fail to squirt at somehere around 28 PSI. We eventually resolved this with a host of fuel system changes to keep the fuel cooler and the pump(s) from having to suck so hard.

Note that an 044 is overkill for a stock car and will result in much more fuel being recirculated through the hot engine compartment and returned to the tank at whatever RPM. This will heat up the fuel. However, the pump will let you know if this is the problem by getting very noisy as it fails due to vapor lock. An intank pump also makes this much less likely to fail. 30 minutes just doesn't sound long enough for this kind of problem to develop, but if you are at high altitude and it's hot out and you have an 044 with no intank pump...just maybe that's it. Once this issue started, we would dump 5 gallons of "cold" gas into the fuel cell and get maybe another 20 minutes down the road before it would heat up and fail again. And, of course, the lower fuel level, the faster it would heat up and fail.

I think as a policy, 928s at high altitude should not have 044s installed and should have intank pumps.

finally!

ck fuel pumps and filters, mine did that last year. Drive then stop, fuel pump would overheat and not work until cooled back down..

Tom in Austin

Interesting comments Bill, first time I've ever heard anything good about the in-tank pump. Maybe Porsche had a good reason for this after all ?

Spun

If I start right after failure, it will try to start, maybe a second it will run then die. The longer I wait, the longer it will run....

Vail is 3200 ft. The pump is loud and always has been. I replaced the one on my BB 89 because I got tired of the noise and went to the 87/88 pump with no in tank pump. No issue to day.

I do have a second 044 sitting here. I can swap those and try that... or maybe just go with the other pump that works great on the other 89 and I drive it hard in all temps...

Though, now that I write, i have had a couple odd issues with BB that made it seem like I had no power... turned it off then back on and I was golden...

Hmmm... the bloody pump is very new and I did not have an issue until I did the intake, but, whenever I can get the tank drained down, I will swap the pump as well...

What happens when the MAF starts to fail? What are the symptoms there?

worf928

Did you remove the powder coating from the water bridge at the sealing surface underneath the hex-head part of the TEMP-II sensor?

Spun

Just did... I will drive it tomorrow and test...

Bill Ball

Ah, I think Dave got it. It explains why this just happened after you did your powder coating.

It didn't fit what I was talking about. That was more just for general information. 3200 feet is not really that high. Don't bother with an intank pump unless you find there is really a fuel overheating or other delivery problem. I'd stick with the stock pump rather than an 044, but, like I said, I don't think 3200 feet is high enough for this kind of issue.

Landseer

Nice diagnosis

Speedtoys

Weve talked about PC and various bits of the intake at your house before.

I had issues with the flappy movement, had to file down the PC at the E-clip ends of the rod, and chased down leaks anywhere where there was water and something to hold it in...including temp2 surface, etc.

Bill Ball

Yes, those areas and any sealing surface should be masked.

Spun

The great thing about problem solving is that it takes all the ideas possible to run down the one that did it. I drove it today with no issue yet, but I have not really tested it yet... we shall see on teh drive to Phoenix tomorrow for sure.

Dean_Fuller

So is this a PC issue? I want to understand incase I PC mine in my next life.

Speedtoys

We'll see, but anything expected to be a ground, a seal (temp sensors seal water in..etc) or where the flappy shafts are...ya..you want "clean".

I wonder how many flappys out there on PC'd intakes are stiff to open, or dont open well at all because of the space the PC takes, that the mechanism needs to rotate under the E clip.

BPG_Austin

Following this thread with interest and noting consensus on mask areas.

BG

worf928

I prefer a troubleshooting mechanism that includes a experimental step that leads to incontrovertible proof or dis-proof of a theory.

In this case that would have been to operate the engine until it began misbehaving and then run a test lead (two alligator clips on a length of wire) from the hex-base of the Temp-II sensor to a ground point like a (uncorroded) cross brace bolt. If the engine settles down with the test lead attached then the theory is proven correct, if not then the theory is proven incorrect.

The issue in this case is that the TEMP-II sensor has two temperature sensors - one for each male pin - each of which grounds to the car's body via the hex-head base of the sensor body through the crush ring to the water bridge to the water bridge bolts to the heads to the block to the engine ground strap.

Powder coating can be high-resistance and therefore interfere with the ground circuit. Initially, there's enough metal in the sensor to give a reasonable resistance reading. But as the engine runs, the ECU's circuits suck electrons out of the metal of the sensor body and with the high-resistance of the powder coating electrons can't flow nicely through the sensor to the ECUs from the body.

Basically, the LH and EZK ECUs begin getting a bogus temperature reading after a few minutes because of the high-resistance ground path. They think it is -20 degrees outside and mess with the fuel ratio and (maybe - haven't looked at Sharktuner maps in far too long) timing. Then, after a couple of minutes the 02 loop turns on trying to trim the air/fuel ratio and all idle-Hell breaks loose.

If you PC the water bridge you need to make sure to remove the PC from the sealing surface underneath the TEMP-II sensor and from the washer surface underneath at least one of the water bridge bolts.

Last, some PC can LOOK like bright aluminum when you Dremel it and provide a good ground path until it goes through some temperature cycling.

A multimeter resistance check from the TEMP-II to a ground point is not enough of a test.