Balky starting - any ideas?

[deilenberger]

Cayenne S (V8) 2006, 56,700 miles. Newish to me.

It's back. It seemed to go away after cleaning the throttle-body (as suggested by a member here..).

Symptom is:

1 - Cold start (hasn't been run in hours, cold or cool outside): Starts flawlessly. Turn key, it's running.

2 - Not entirely cold start (was run, got to normal temps, stop, park for 30-60 minutes) but fairly cool environmental conditions (below 50F): Turn key, it tries to start, and sort of stumbles to a start. I usually have to give it a bit of gas via the pedal to get the RPMs up to where the starter disengages. Once started it runs fine. No other symptoms, no SES, no stalling.

I've done:

1 - Air filters (not that I suspected these as a cause - mostly 'cause it has 56k on it and the 60k service is coming up..)

2 - Plugs - (put in the correct Bosch 4-prong jobbies - great price at Amazon for them, and again done because the service is coming up.) There was a slight stumble coming off low RPMs in A6th - the plugs made this go away.

3 - FI Cleaner - Ran a dose of Techron (FI cleaner - very good stuff) with a tank of fuel.

4 - Cleaned the MAF's - with CRC MAF cleaner.

5 - Cleaned the throttle-body - using solvent on a rag (rather then spraying)

It seemed to clear up after cleaning the TB, but now (after several days) has started reoccuring.

Things I've found out:

1 - The goofy auto-crank on first start attempt isn't the best thing in the world when this happens. The starter tends to stay engaged and then be over-run by the engine. If I could, I'd disable this. If it wasn't for this, I could pretty much simply ignore it.

2 - Looks as if other people have experienced this - and fuel pumps and engine temp sensors have been mentioned as possibilities, but I can't find any posting where these actually were the solution. Most threads start out with these, and the owners are getting them done, then they never come back to report the results.

3 - It has been intimated here that the fuel pumps DO "charge" the system if the door handle is pulled when the car is COLD. I can't find confirmation on this, except people reporting when doing the coolant pipes that they feared someone opening the door while the fuel injection system is open since it would spray gas all over. That could explain the perfect starting when absolutely cold if there was a bad check valve in the fuel system allowing it to depressurize, and the time that it sits and doesn't start OK is due to the engine not being cold enough to trigger the system pressure charging. (This makes me think this could be fixed by programming by Porsche so the pumps charge whenever the car is cranked like normal vehicles do.)

4 - According to a friendly Porsche tech I chat with - he felt the TB needed recalibration, and there may be a flash for the DME to address this issue. Anyone KNOW about the flash for sure? He's in California, and I'm in NJ, so I can't ask him to run the car on the diagnostics tool until July when I'm out there.

What I'm looking for:

- Anyone have the fuel injection pressure leak-down spec's? I could invest in a fuel pressure gage and check how long it takes to have the system bleed down pressure.

- Anyone else experiencing this? (In the BMW world where I came from - there is a very common problem every spring where certain models start becoming 2nd-crank starters. It seems related to the seasonal change in fuel formulation from winter to summer fuel. It has been posited by BMW that bad check valves in the fuel pumps cause this - but on my now departed '02 M3 - it continued doing this seasonally for several years after having the pumps replaced under warranty.)

- Anyone with firsthand knowledge/experience with the same problem, who has managed to resolve it, and how it was resolved.

I'm giving strong consideration to buying a DuraMetric so I can watch what is going on as far as fuel-pump activation, and temperature sensors, and see if there are any non-OBD-II codes stashed away in any of the engine management modules.

Anyone - Buhler?

TIA!

[bigbuzuki]

3 - It has been intimated here that the fuel pumps DO "charge" the system if the door handle is pulled when the car is COLD. I can't find confirmation on this, except people reporting when doing the coolant pipes that they feared someone opening the door while the fuel injection system is open since it would spray gas all over. That could explain the perfect starting when absolutely cold if there was a bad check valve in the fuel system allowing it to depressurize, and the time that it sits and doesn't start OK is due to the engine not being cold enough to trigger the system pressure charging. (This makes me think this could be fixed by programming by Porsche so the pumps charge whenever the car is cranked like normal vehicles

For starters, have you actually heard the fuel pump prime when you open the door, first cold start?

Fuel pump fuses number 13 (secondary pump) and 14 (primary pump) in the engine compartment fuse box can be pulled one at a time, isolating the individual pumps.

It's common to pull the primary fuse if the primary pump fails and allow switching to the secondary pump to get the vehicle moving again.

Here is a post from RFM in the past..it may help.

Edited April 12, 2011 by bigbuzuki

[deilenberger]

For starters, have you actually heard the fuel pump prime when you open the door, first cold start?

BigBuzki, thanks much for the reply.. the silence here made me think I was yelling into the wilderness..

Actually - yes - I have (and I'm a bit deaf so that's an accomplishment.) And I'm guessing, despite my theory above, that this isn't the problem, since when I went to start it AFTER hearing the pump prime last night, it exhibited the usual symptoms.

Last night I again cleaned the throttle body - with official "CRC Throttle Body Cleaner".. and a microfiber rag. Got some more dirt out of it - and it made no difference (I think it was Einstein who said the definition of idiocy is repeating what didn't work before and expecting different results.. I'm perfectly capable of idiotic actions.)

Fuel pump fuses number 13 (secondary pump) and 14 (primary pump) in the engine compartment fuse box can be pulled one at a time, isolating the individual pumps.

It's common to pull the primary fuse if the primary pump fails and allow switching to the secondary pump to get the vehicle moving again.

Here is a post from RFM in the past..it may help.

http://www.renntech....post__p__141028

I can certainly see that being helpful IF a pump fails.. but since I've exhibited no symptoms that I can see of a failed (non-pumping) pump (it does start, just balky sometimes; it isn't limited by fuel - it can easily run up to redline under a load) I didn't see where that would be of use. My pump failure thoughts were more along the line of a leaking check valve - not a non-pumping pump.

I'm curious about the pump system.. are there actually primary and secondary (ie - low pressure feed to the high pressure) pumps or is it plumbed so the pumps are simply run in parallel with whichever has fuel running (thinking of the Cayenne on a side slope where all the fuel runs to one half of the tank..)? The appearance of the pumps would make me think they both are of equal capacity.. I have stashed that post away though.. (probably should put a copy in the owner's handbook.)

The symptoms I'm getting are simply a slow to start situation (balky), and when it does start, it stumbles a bit until it clears itself out, then it runs fine.

The problem reminds me a lot of when there were ugly carbon deposit problems on intake valves back in the '80's on BMWs. The deposits would soak up fuel as the engine was running, then if the engine sat long enough hot - the fuel evaporated out of the carbon, causing a VERY rich condition when you went to restart. You had the same symptom - balky starting and then stumbles, followed by running fine once the rich mix flushed out of the intake. BMWs cure was (1) walnut shell blasting of the intake valves - really! (2) reformulated fuel - the BMW "test" became the standard adopted by all fuel manufacturers and eventually the government (3) treatment with FI cleaner - repackaged Chevron Techron. That did clear up the problem - eventually. It took a few years though.

Last night I added a second dose of Techron to the fuel - and this AM - I did an "Italian Tuneup" on the way to work a bit - playing around in Manual mode. That did help back in the days of badness with BMWs.. perhaps it might do some good here if the valves are carboned up. I have no idea how the truck was driven before I got it - I've got just about 800 miles on it now, so it is possible it was getting short trip duty in our very urban environment - and the conservative (fuel economy I assume) shift points the Tiptronic picks rarely let the engine get much above idle driving in NJ.

Thanks again - good to know someone is out there.. :thankyou: And good to bounce ideas around. Sometimes it jogs things for me..

Edited April 12, 2011 by deilenberger

[deilenberger]

BigBuzki,

The link you provided did make me recall thinking it could be an engine temp sensor problem. I "won" a DuraMetric on Ebay last night (sorry to all the other bidders..) and expect to see it by the weekend, so I will be checking this. If the temp sensor is reporting a cold engine to the ECU/DME when it's actually warm, that could account for the balky starting (too rich a mixture.)

Another slight clue - I *think* I can avoid the problem by very slightly depressing the gas pedal as I start (which is really unnatural for me since it means left foot on the brake, and I've been driving stick-shift for the past 50 years.. still do in my BMW M-Coupe..) The engine starts better I think (haven't tried it enough times to say conclusively) races a bit to about 1,000-1,100 RPM, then settles back to normal idle at around 550-600 RPM as I release the pedal. That may indicate the rich-mix on starting as being the root cause, or not.

:cheers:

Thanks again!

[wvicary]

You might have a problem with the EVAP purge solenoid which is causing an uncontrolled vacuum leak during start up. You really need to check for codes so wait for the Durametric to show up and post any results.

[deilenberger]

You might have a problem with the EVAP purge solenoid which is causing an uncontrolled vacuum leak during start up. You really need to check for codes so wait for the Durametric to show up and post any results.

I'm hoping it gets here soon.. :)

Can the EVAP purge fail (I assume open) and not throw a CEL?

[mudman2]

It does sound a bit like the check valve

[wvicary]

You might have a problem with the EVAP purge solenoid which is causing an uncontrolled vacuum leak during start up. You really need to check for codes so wait for the Durametric to show up and post any results.

I'm hoping it gets here soon.. :)

Can the EVAP purge fail (I assume open) and not throw a CEL?

It can throw a code but not trigger the light. I think that particular EVAP code has to occur for two consecutive drive cycles to set off the light. The valve can have an intermittent failure but lets check for codes as this is all guess work.

[deilenberger]

DuraMetric told me - no codes anywhere on the truck. I guess it's one of the few times I've been disappointed by no codes. It thinks everything is working just fine.

Update on this - sorta a cross-post to rennlist..

It happened last night right in front of a friend who happens to be a very good high-end car mechanic.

When I saw him today (I was checking on a parts delivery for my wife's BMW 5-Touring's brakes..) he immediately said "check valve" - as in the one in the fuel system that keeps it pressurized when the pumps aren't running so it can start quickly. He said it sounded like no fuel.

So.. gotta get a fuel pressure gage and hook it up and see what the bleed-down rate is on the fuel-rail. Unfortunately the check valve appears to be part of the fuel pressure regulator which is part of the passenger's side fuel pump access panel assembly - the part that fastens into the opening in the tank.

Part isn't silly awful money (around $125 or so) - but replacement looks like a smelly PITA job. And to do it right requires buying (or renting/borrowing) a special tool used by Porsche, Audi and VW for the locking ring for the assembly into the fuel tank. Or - a hammer and an old screwdriver.. (BTDT in the past, trying to avoid McGuiver's as I get older..) Best price I've found on the tool is a bit over $40.

The DuraMetric showed up tonight. It very happily found absolutely no codes at all in any module it could read in the truck. It only managed to read some of them after an instrument cluster reset (accomplished by pulling the cluster fuse - left-door/dash fusebox.) Apparently the cluster also serves as the computer buss "gateway" for the diagnostics, and if it gets confused (which apparently it can rather easily) it won't let any codes sneak out to the OBD-II port.

Interesting stuff this. Next - find a fuel gage.

So what next? Fuel gage. My friend said a loss of fuel pressure won't trigger the OBD-II system, and likely won't even cause any codes to be stored. I guess he's right because there was nothing to find.

I did learn how to reboot the instrument cluster/gateway - a fuse in the left side dash fuse box. 10 seconds out and it's rebooted.. Makes a nice whirring noise when the fuse is plugged back in.

[deilenberger]

A bit of an update... haven't gotten a fuel pressure test gauge yet - but that's my next step. My WAG is the check valve in the fuel pump (probably in the pressure regulator) that's responsible for keeping the system pressurized between starts is allowing pressure bleed-down. Harbor Freight has one for $25 - probably go pick it up tomorrow.

I stopped at the local dealer this AM, and they'd sold the car new, CPO'd it and serviced it. They had offered me a service history printout.

It was taken to them back in 05/10 for EXACTLY the same problem. The printout said the DME had no stored codes and the mechanic couldn't replicate the problem, so.. it got the sunshine treatment. There were other incidents of the truck being in for the same repair multiple times. I can understand how Porsche gets a poor reliability reputation when dealers don't actually fix things. The dealer service manager asked if I wanted to schedule an appointment after I commented on the repeat problems. They could give me one in 2 days - which to me means not many people are bringing their cars in for service, and it explains why the several Porsche independent speciality shops in the area are doing so well.

[mudman2]

When the check valve goes it does throw a code readable via Durametric. but mine was intermittent for a few weeks prior to the code appearing. Mostly had the problem after fill up with the engine running

[deilenberger]

Finally got to Harbor Freight today.. and of course the $25 (-20% coupon) fuel pressure gauge I was buying turned into a $150 bill (this just seems to happen at HF..)

Got home, put it together (nice kit - comes with all sorts of adaptors..) Pulled fuses 13 & 14 (do NOT drop a fuse in the underhood fuse box - it's a PITA to diassemble to get to the bottom, I could have left it there - nothing to short, but just couldn't bring myself to.. but I digress..) Pulled the fuel pump fuses then cranked the car to relieve any residual pressure, and hooked up the gauge.

I actually needn't have bothered pulling the fuses. I replaced the fuses, and tried starting the truck. On starting - it did it's usual bumbling start - and the fuel pressure gauge read a nice solid 4-Bar. I tried pulling fuse 13 - no change - solid 4 bar. Put it back in, pulled fuse 14 - engine died as fuel pressure dropped to 0. Tried restarting - another bumbling start - and then I had 4-Bar again - so the fuel pump switchover that others have described worked as expected, and both pumps were capable of 4 bar.

I replaced both fuses - then shut the truck off. Pressure immediately started dropping - reaching 0 Bar in about 15 seconds. Tried this several times, start it (starts lean) check pressure - 4 bar, shut it off - pressure immediately drops.

So - I'm guessing the check valve isn't checking. Or I have a really badly leaking injector. I really feel it's probably the check valve since once the truck does start, it smooths out quite quickly - no indication of a rich mixture, and when I replaced the plugs - they all looked the same. If one injector was dribbling fuel, I'd expect to see it in the plug.

Anyone know where the check valve is? The spec in the manual is the system must hold 3-Bar after shutoff for at least 10 minutes.

I'm going to approach the dealer with the info - since they hadn't bothered doing a pressure test (or ANY testing) when it was in before for this problem.

Oh - what else did I buy at Harbor Freight? They have a neat boroscope tool - usually $130 - on sale for $79 - couldn't pass it up. Has a flexible probe about 2' long with a camera on the end with two LEDs for illumination (the brightness of these can be varied.) Color LCD - about 3"x2.5" - battery operated. I figured I could try to see if the heater pipes had been replaced in my truck. I can now authoratively say they haven't. By wiggling it around behind the engine, with the tip bent to point forward a bit - I was able to see the pipes where they pass through the clamp at the rear of the block - and they're black plastic. I also picked up a set of storage bins, and some small toys (aluminum valve caps, 1/2" deburring tool - free with a coupon!) It's not easy getting out of HF for less then $100. I've proven that many times.

Edited April 21, 2011 by deilenberger

[bigbuzuki]

Great and logical troubleshooting Don.

You should show your dealer technicians how its suppose to be done. :rolleyes:

That fuel pump check valve must be getting nervous by now.

Good luck.

Edited April 21, 2011 by bigbuzuki

[deilenberger]

Thought I'd update this.

The truck now starts correctly. :notworthy:

It spent the past week at Princeton Porsche in Lawrenceville NJ puzzling their techs and Porsche-Germany.

The initial thought (and diagnostics via the manuals) was fuel pumps having a bad check valve. So - it got two new pumps. No joy. Next thought was the fuel-pressure-regulator, since it's supposed to work as a check valve. New FPR - no joy. Finally Porsche Germany had them looking at the spaghetti nest of plumbing inside the tank for a leak in a hose. In order to check it out, they removed everything from the tank, setup a test with the pumps in gasoline, and wired it up. No leaks. Finally - since the only component they hadn't tried was the fuel filter - they borrowed one and put it in. JOY! They ordered a new one and installed it this AM - and the truck now starts perfectly. Apparently there is an undocumented check valve inside the fuel pump that even Porsche Germany is unaware of.

I give Princeton Porsche's service department 5 stars on what is an ugly job well done (mucking about inside a fuel tank with gas in it is not a pleasant job at all.. BTDT and don't want to do it again.) They persisted, when other dealers had given up. They kept me well informed on what they were doing, and I had use of a loaner Audi for the duration (2.0T - it isn't a Porsche for sure.. but at one point they felt bad it was taking that long and they offered me a Panamera loaner. Not wanting to get spoiled, I stuck with the Audi..)

If you need GOOD service in NJ - I can recommend them without reservations. Good friendly people, who didn't mind me bothering them with my ideas on what it could be (and I was wrong - but no one had thought of a check valve in the fuel filter.)

Edited May 10, 2011 by deilenberger

[PKN]

That is excellent info and a relatively cheap fix for what could have been something much worse. I might just get one in just in case. Thanks for the persistance and excellent trouble shooting

PKN

[wvicary]

Any chance you could post a pic of this filter check valve? I can't seem to recall one being in the filter housing.

[deilenberger]

Any chance you could post a pic of this filter check valve? I can't seem to recall one being in the filter housing.

Sorry - nope. I don't have parts that were removed - apparently Porsche wants them back for warranty claims. If there is one (and I'm going on the tech's diagnosis, which seems to have worked - just tried starting it after it sat for 4 hours - started right up :) ) I suspect it's internal to the plastic housing. Can't see how the housing might be disassembled - but I guess with a big enough hammer anything is possible. Have you had that flange/filter apart?

[wvicary]

Any chance you could post a pic of this filter check valve? I can't seem to recall one being in the filter housing.

Sorry - nope. I don't have parts that were removed - apparently Porsche wants them back for warranty claims. If there is one (and I'm going on the tech's diagnosis, which seems to have worked - just tried starting it after it sat for 4 hours - started right up :) ) I suspect it's internal to the plastic housing. Can't see how the housing might be disassembled - but I guess with a big enough hammer anything is possible. Have you had that flange/filter apart?

I've had the filter housing apart once on a Cayenne and a few times on the Touaregs but for some reason don't remember any type of valve. Glad you got it sorted though.

[deilenberger]

I'd love to have a fuel filter and FPR in hand so I could do some checking on what valves are where.. it seems there are multiple places the valves COULD be - but no real documentation that I could find on where they're supposed to be. The fuel system diagram sort of shows some in the fuel pumps, but that's about it. I uploaded the fuel system diagram.. still not at all clear where the check valves really are..

ATI CAY2 0604 In Tank Fuel Delivery.pdf

[deilenberger]

Quick update - the P!G has gone > 10,000 mile since the filter was changed, and started flawlessly each and every time :) The filter assembly was the problem.

While I was in LA (drove there in it from NJ) - a brand new Porsche boxed filter showed up on ebay - by a seller in LA. I won it for $20 (!!) and went to his house to pick it up. It is still in the truck since I'm still on the road driving back from LA.

The trip is almost 10,000 miles so far.. I took the scenic route. ;) The truck so far has given me awful trouble - a front sidemarker bulb burned out. Once I figured out how - it took a few minutes to change. It has used NO oil at all, and no coolant (I had changed the pipes to aluminum ones before I left.) It also has run across several deserts at temps well over 100F, AC keeping me fat and happy - and the temperature gauge has stayed rock steady at 180F.

I've been taking photos of the P!G in various places along the trip - so I'll post a travelog of sorts when I'm done.. just shows you CAN leave home in a Cayenne.. (seems some people are afraid to..)

Edited August 15, 2011 by deilenberger

[ALEV8]

Don that is wonderful news. How hard is the fuel filter to replace labor-wise? I know you guys saw my video and said my start up is normal, but it sure seems sluggish.....

[deilenberger]

Don that is wonderful news. How hard is the fuel filter to replace labor-wise? I know you guys saw my video and said my start up is normal, but it sure seems sluggish.....

It's an ugly job since both sides have to come out and you're working inside the fuel tank. Probably not that "hard" - but it probably is that unpleasant to do.