996 not starting

Hi Guys, advice and wisdom needed. I have a RHD 2003 996 Carrera 2 Tiptronic, well maintained car, with 69,000 km but rarely driven the past couple of years. Its winter here in Australia but only about 13 degrees Celsius so nothing cold compared to over your way. Car has not been driven for a few months, battery is approx 2 1/2 to 3 years old, I do keep an eye on it and keep it on a modern style CPU controlled trickle charger for a few days every couple of months. All of a sudden Car will crank but not start. My first thought is maybe the battery although showing a good charge is weak? It measures 12.8v with a multimeter, and when the car cranks the Voltmeter in dash starts at 12V and drops to around 10v whilst cranking, then returns to 12v. It will turnover and crank for several attempts so battery appears strong but no start. Where do I start?

Check fuses C1 to C4 first. When you turn the key to the last position before cranking, do you hear the fuel pump (in the frunk) running for 2 seconds and then stops?

Quotes source post: 11535537

All Fuses c1 to c4 all check OK (checked with meter not just visual) I can't hear fuel pump running in the frunk then cutting out after 2 seconds (not 100% what it should sound like though) Can hear a relay clunk in the frunk when key first turned to that position and some Hi Frequency noise typical of 996 electronics

Is it clicking? Can it be the ignition switch?

When you crank, does the tacho needle bounce just a little bit? Is it possible the gas has gone stale? Also could rodent damaging wires a possibility since the car has not been driven for months? Next test would be to jump the fuel pump relay to see if the fuel pump works. If it works, you need to verify sparks.

Quotes source post: 11535705

Tacho goes to first bar (like 100rpm) when cranking No chance of rodent damage, is in a locked clinically clean garage with no signs of rodents Fuel would be 3-4 months old, but would that stop it from starting, wouldn't it more likely run rough?

Ok, lets jump the fuel pump relay and see if the pump works then. Next is checking sparks.

Btw, you have a scanner you can check for any pending error codes?

I found my old eBay durametric which I think is a knockofff buts works fine Ran a scan, motronic no faults logged, Alarm shows 2 central locking codes 59 and 60, and radio code though it's present and working fine ABS shows module failed to initiate, as do several others, presuming as it won't start Tried manually starting fuel pump relay with durametric and can hear relay click on in cabin, and minor fuel smell

Did you hear the fuel pump running though (a humming sound from the frunk)? The motor sound should be pretty obvious. If not, your pump is not getting power (wiring or relay issue) or it's gone. Fuel smell does not sound good. If you can't hear any pump running, the next steps will be to get to the relay and jump it there. If the pump runs, bench test the relay and also check if the relay gets power (both high current power and control power for its coil) during cranking. Next step is to get to the pump and check there. If you can hear the pump, you can jump the pump and try starting the car. If not, you need to check for sparks.

Quotes source post: 11536382

I still don't reckon I can hear the fuel pump running. If it is its mighty quiet. I have removed and checked the relay, applied a external power source, it is definitely switching as it should, and making and breaking contacts. check the power at the relay socket coil terminals and its getting 8volts with the ignition on and 12v when key turned to start position so i would assume its getting correct switching supply to the relay, I can here and feel the relay is definitely working Tried putting a jumper accross relay socket, still didn't start, but oddly jumper wire got stinking hot, wire was about 1sqmm so I would have thought thick enough for that current but maybe not, its a 15A relay I guess the next step is pull the battery out and check voltage is getting to the fuel pump. Bit miffed at this stage, hard to believe its the fuel pump at that low mileage.

The clicking of the relay is a good sign but the relay can still fail on the power contacts side so you will have to bench test it to be 100% sure (check for continuity of the power contacts when relay is activated). However, since you couldn't hear the fuel pump even when you jumped the relay, I would say it's likely your fuel pump is not spinning. If it does, trust me, you will hear it even though it's not loud. Kinda like the windshield washer pump running. 1mm sq wire should be able to take 12A without getting too hot and the fuel pump only takes 7A when it's running. If your multimeter can measure max 20A of DC current, you can put it inline with fuse C4 (25A) to measure the actual current. Note that it will blow the internal fuse of your multimeter (if there's one) if it goes beyond 20A though. If the current is much higher than 7A, that's yet another indication your pump is stuck. I agree the next step is to check if power gets to the pump at the pump. Once you're at the pump, you should be able to feel it and hear it spinning. If you need to hot wire the pump for testing, be careful not to create sparks near the pump in case there's any vapor leak. Also, you mentioned you smelt gas before...

If the smell of fuel is coming from the exhaust, check your spark. Yes put your nose in it and get someone else to crank. If not be very careful with bridging relays and find the leak first. Although having a sudden leak after trying to start it would indicate fuel presure.

Quotes source post: 11541678

I reckon I am getting fuel smell from exhaust when cranking. Silly question, but step me through procedure for checking spark, as each plug has its own coil pack, I'm unsure what to check; last car being an old 80's air cooled 911

My $$ is on bad fuel pump.

Quotes source post: 11543394

You can remove any one of the coils and use a spark plug tester similar to this http://www.autozone.com/autozone/acc...ier=10257_0_0_

OK. Here is where I am up to I have removed the battery and the access panel above the fuel pump. Put car battery on floor and powered car up with Heavy Duty 600 amp jump start cables. Removed power connector off fuel pump, put multimeter in 2 outside terminals (brown wire and green/white stripe) Turning the ignition key to ON position, no power to fuel pump connector Turning key to START position 12V supplied to connector I thought the pump was supposed to get a few seconds priming on the ON position first ? Should I try and apply 12V direct to pump before I plonk down $600 for a new pump? Confirming I have the correct terminals, 2 outside ones (2 inner ones look a very light wire guage, I assume these are the level sender?)

All you did was right and the pump is supposed to prime for about 2 seconds if you leave the key at the last position before cranking. Why don't you just jump the pump itself (bring the connections outside of the pump area and make sure brown wire is negative). You should hear the pump spinning if it's working. Btw, you can also measure the resistance of the pump and see if it's open circuit inside. It should have very low resistance.

Quotes source post: 11552010

Pump Measures 0.7 ohms across 2 outside terminals, which seems very close to a short,

What would be causing the 2 second prime not to occur?

0.7ohm may be ok. Can you try using Durametric to activate it again now that you can definitely hear the pump? I just tried it with key turned to last position then activate the pump using Durametric. I can hear it clearly at the driver's seat. No way you can't hear it when it's spinning

Quotes source post: 11552088

Hooked up the Durametric Tried starting fuel pump final stage with Durametric, can't hear pump even up close. Can definitely hear fuel relay switching on. Pulled fuel pump connector off, measured voltage with MultiMeter. and turned on again with Durametric, its definitely getting a switched 12V supply at the Fuel pump connector, so I think its pointing to a dead fuel pump. I did get a "fuel level sender faulty error drive to workshop" after plugging fuel pump connector back on, which I guess could be normal, but it doesn't show any errors logged in the Durametric fault codes still after refreshing it. Only bit miffing me is why I'm not getting 2 second priming voltage when ignition first turned on.

I have seen the 2s priming sometimes don't occur and I speculate that the DME may have some logic triggering it based on things such as last time engine was run, ignition was on, and what not. The fact that you get power to the connector and the pump still not running does point to a failed pump. To be 100% sure, you can plug in any car incandescent bulb (anything less than 80w) say a brake light bulb to substitute for the fuel pump and see if the bulb lights up when you activate the pump with Durametric. That will tell you the circuit can supply enough current. Then supply power directly to the pump and hear if it spins. That will tell you if the pump is bad. The pump sounds just like a windshield washer pump so you can't miss it if it's spinning.

Quotes source post: 11552197

I have tried supplying power direct to the pump, no sound or vibrations It does try to weld the wires onto the terminal, so I'm sure its drawing current, maybe a shorted turn in the motor windings I guess. I guess I should now remove pump from fuel tank and do a final test on bench before buying a new pump, in case its something weird like a failed connector between top connector and the internal one on pump itself

That's the logical way to proceed. However, I do not know a safe way to bench test the fuel pump unless you can ensure all the fuel inside the pump has been drained. Not a whole lot but still a hazard. Without being submerged in the fuel, the brushes inside the pump (being a simple dc brush motor) will creat sparks. Just be careful.

Any ideas on why there is no fault codes stored on a dead fuel pump? Maybe because last time car ran it started properly , drove fine, parked it in garage, and pump mysteriously died whilst parked? Spoke to a Indy Porsche Mechanic, his theory was moisture/condensation can build up and sit at bottom of tank as car wasn't driven for months and pump is at the bottom sitting in moisture and corrodes and seizes up. Car has only done 60,000km, so an early death for a fuel pump

There are no sensors monitoring the fuel pump (or fuel pressure) as strange as it may sound. Also the DME is not directly supplying power to it (it's via the pump relay), not like the fuel injectors that it can monitor the current drawn and flag the corresponding error codes. Btw, the pump is not cheap but it's less than $600 though http://www.ecstuning.com/ES1494742/ The moisture theory sounds plausible. You can do a postmortem on the pump and find out how it failed.

That does sound bizarre (no monitoring) considering how important and basic fuel is in a car !! I'm not sure if ECS ship outside USA, I'm in Australia, so $600 is a local price here and I shopped around, we are used to a bit of price gouging down under I'll contact ECS and find out, as that's even cheaper than Pelican (who do ship to Australia) Will keep you posted when I get the pump out and dissect it Thanks' for all the advice and help so far BTW

Hey, glad to help out here. Re monitoring, it's probably a cost issue since fuel pump failure is not that common and also it's straight forward to diagnose, not like those mysterious misfires. Oh yeah, I forgot you're in Australia. I must admit parts are cheaper in US compared to other countries. I do agree that the mileage is very low for the pump to fail. Look forward to what you find.

OK, Post Mortem in progress Pump now removed from Tank, no physical signs of damage, no cracked hoses or casing, water etc Setup pump up outdoors and connected some 3 metre leads, powered her up with battery, zippo, nothing, but did notice some vapour coming out of vent holes on top of pump motor Left it for a few minutes, went and checked it, and internal metal motor pump housing quite warm, so i would say my initial thoughts of a shorted turn in the windings is probably the cause Good news is, pump removal was really easy, had it out in 5-10 minutes after taking battery and cover off the other day, have left half a tank of fuel in tank, as couldn't drain with pump not working, as it came out so easy I don't think refitting new one will be too hard

Glad you were able to bench test it and confirmed it's seized. I love postmortem on failed parts I always do that. I think the motor is either seized or it has shorted winding like you said.

OK, pump arrived, sourced one locally for $500 (AUD) plus tax which wasn't too bad, free next day delivery Fitting was quite easy, would have been easier if I had taken a photo inside the tank the see exactly where the flat side was located before i pulled old pump, as you can easily be 45 degrees out, none the less, was all back together in about 20 minutes My recommended Hi tech method is stand inside the frunk so you can get your arm right in the bottom of the tank easily without contorting yourself too much, give pump a good wiggle to make sure all tabs are properly locked in. (as first couple of tries they weren't) Also managed not to blow myself up with half a tank of fuel still in her which was an added bonus She's alive again, fired straight up after a few cranks, seems to be running & driving fine (still didn't here pump running on first key on position though, maybe i will on next start now its fixed) Thanks for all the help, as usual power of the 996 forum is awesome

Glad you got it sorted!

I love a happy ending.

Nice to hear you finally solved your no start problem. Now go have fun driving your Porsche.

I would say the Fuel Pump replacement could be easily added to the list of 996 DIY jobs on the forum (I couldn't find one when I searched) Not a difficult job at all and not even vaguely mechanical by nature, all the fuel lines just clip on, not even a clamp i site anywhere, pretty much a plug N Play job

All this work for a relay? Its five bucks. Go buy one. I have a similiar problem, intermittant. So I will replace the relay as a PM measure.