Cold hard start Rich erratic idle shenanigans

[peanut]

I've been chasing a hard cold [style="background-color: #ffffff;"]start[/style] and rich erratic idle problem with my car ever since the current cold spell set in.

Two weeks ago the car started instantaneously every time , as always, with an immediate stable idle and perfect running.

One morning I went out to take the car to the garage for its MOT and it wouldn't start until I had given it full throttle and pumped the pedal like crazy. (It subsequently failed the Emissions with a CO of 10.031 and an HC of 1239)
For the past week I have reaquainted myself with the Bosch manual and trawled the net looking for similar cases in the hope of narrowing down and pinpointing the cause of the problem.

Not wanting to fall into the usual trap of fitting a whole raft of new components to the car until it hopefully starts working again , I tried all the obvious checks first .
TPS stripped and conacts cleaned click reset voltage and resistance OK
AFM stripped... wiper arm cleaned adjusted and solder bridge to bridge arm etc Voltage and resistance OK
Engine Temperature Sensor Voltage & resistance checked OK
Idle Control valve voltage & resistance OK checked operation with 12v battery OK
Vacuum hoses spray checked with carb Cleaner. no leaks found.
FPR and FPD pulled vacuum hoses whilst running. No fuel let by, OK
Pulled the ECU opened up check for damp /condensation. Cleaned connector terminals .

Getting desperate now .....I test and clean the entire Ignition system and change spark plugs...
Still no change ....so I'm thinking I can't put it off any longer I'm going to have to check Fuel pump flow and pressure even though the car is running like a dream and pulling like a train right up to WOT.

I pull off the fuel rail pressure test nut and am surprised to see absolutely zero fuel ,,,nothing , nada zilch.
Connect length of garden hose and bottle to the fuel rail to check the flowrate ....switch on the ignition expecting a momentary squirt of fuel as the fuel pump fires and shuts down but see no fuel at all ? Ah ha ! thats not right surely. Beginning to wish that I'd trusted my gut instinct and gone straight for the fuel pump and injectors from the off ..

I'll have a read of the bible tonight and draw up a list of ECU tests all the sensor and fuel system inputs and outputs for tomorrow but at the moment I'm expecting to find that the ECU is the culpret in all this , probably due to all the damp and condensation in the car this past 2x weeks . I hope not,.. as that is going to be really expensive to replace. I will also check the flow rate and performance of of the injectors as one or more may be leaking which would cause a rich idle and loss of fuel pressure

UPDATE 1st March
Well there has been some development so i thought I'd add the outcome and current status to the thread here so that it is easier to follow progress if any.

I managed to start the car and get it to the garage last week and eventually got the emmisions right down to get an MOT pass however the rich idle mixture issue has not gone away or been cured yet so I'm still looking for the cause and a permanent solution. Heres what happened at the MOT station.

Quick recap car suffers with a rich idle mixture. It would hard start on full throttle and pumping, then cold or hot the idle would hunt from 600rpm up to 1200rpm and back continuously until eventually it died from the plugs sooting up. The car revved normally and performed faultlessly with perfect throttle response up to WOT.

At the testing station hooked up to the exhaust emissions analyser for the MOT retest the levels were still too high . I tried removing the connector for the ICV (idle control valve ) and within a few seconds the 02 and HC levels had dropped significantly .I believe the default condition of the ICV valve when disconnected is fully open as per cold start condition ?

With the ICV disconnected I also removed one of the vacuum pipes and the 02 then immediately dropped further from 10.031 right down to 1.783 ! within 8-10 seconds. (pass is 3.500)
At the same time the HC also dropped from 1239 down to 138 ! (pass 1200 ) Amazing !

. So presumably the ECU had registered the extra air supply and adjusted the pulse rate to the injectors to lean the idle mixture back to normal .?

Now I've just got to figure out why the engine is getting too much fuel at idle or insufficient air or whether the ECU is either faulty or not receiving the data from the AFM, ICV, or engine temperature sensor etc,

New pump ,strainer, filter and engine temp sensor have now been fitted and the fuel pressure is stable at 3bar .Leak down test and flow rate are fine too so I'll start with cleaning and testing the injectors as the car has been stood unused for long periods over the past few years.

Following that I'll have to test all the sensor voltages in the ECU connector to make sure that the ECU is seeing what it should.

 

[bmnelsc]

My understanding of fuel pump operation on startup is as you described (pressurizing with ignition on prior to startup). You don't mention that you have checked your O2 sensor and wiring. I have seen posts on various Porsches where a faulty O2 sensor has produced similar problems - most recently a later model 911 suffering a sudden 20% increase in fuel consumption along with rough running/idle issues particularly when cold.

Back to the fuel pump...my '91 S2 NEVER starts immediately - the engine always spins for 1 to 2 seconds before it fires (not sure how many revolutions that is). I've tried turning the ignition on and waiting a few seconds for fuel pressure to build before cranking but it seems to make no difference. The local OPC did the last tune up and has serviced the car a couple of times since and never mentioned anything about how my car behaves so I assume it is "normal".

Keen on seeing your resolution to this problem - good luck.

 

[924Srr27l]

The Bosch Fuel pump should have a Check valve screwed into the end of it which is designed to hold the fuel pressure when the engine is off.

If the spring has weakened or it has some other malfunction the fuel pressure will be lost and will have to built up when starting (More cranking than normal)

before it's enough for the Injectors to spray at the correct pressure and amount etc... ?

R

 

[bmnelsc]

Thank you - that sounds like it might be the solution to my starting issues.

 

[peanut]

My understanding of fuel pump operation on startup is as you described (pressurizing with ignition on prior to startup). You don't mention that you have checked your O2 sensor and wiring. I have seen posts on various Porsches where a faulty O2 sensor has produced similar problems - most recently a later model 911 suffering a sudden 20% increase in fuel consumption along with rough running/idle issues particularly when cold.

Back to the fuel pump...my '91 S2 NEVER starts immediately - the engine always spins for 1 to 2 seconds before it fires (not sure how many revolutions that is). I've tried turning the ignition on and waiting a few seconds for fuel pressure to build before cranking but it seems to make no difference. The local OPC did the last tune up and has serviced the car a couple of times since and never mentioned anything about how my car behaves so I assume it is "normal".

Keen on seeing your resolution to this problem - good luck.

thanks for the suggestion, I looked everywhere for the 02 sensor until another PCGB member said that the 944S2 doesn't have one.

Its bypassed by a wire loop connector which sits alongside the reference sensor connector.This makes the 944S2 run 'open loop'

I've never timed my S2 start but I guess its about 1-2 seconds like yours which is pretty quick.

I am beginning to suspect my fuel pump non return valve is faulty or I have a leaky injector as the fuel system is not holding any residual pressure. I'll do a timed leakdown test tomorrow to check it out.

 

[peanut]

The Bosch Fuel pump should have a Check valve screwed into the end of it which is designed to hold the fuel pressure when the engine is off.

If the spring has weakened or it has some other malfunction the fuel pressure will be lost and will have to built up when starting (More cranking than normal)

before it's enough for the Injectors to spray at the correct pressure and amount etc... ?

R

yes Its begining to look like this is the problem although I have not eliminated a leaky injector yet. I'll do a leakdown test tomorrow .

I've never seen the fuel pump on mine.... I presume you access it behind the rear wheel arch ?

 

[peanut]

Thank you - that sounds like it might be the solution to my starting issues.

what starting issue does yours have ?

 

[bmnelsc]

Thank you - that sounds like it might be the solution to my starting issues.

what starting issue does yours have ?

Just having to spin the engine for a few seconds before ignition (even when warm). Virtually everything else (non Porsche) I've owned since fuel injection started faster than my 944.

Re O2 sensor - sorry - I remember reading that post. Porsche started using O2 sensors on the 944 from the beginning in the North American market although I see ROW cars could use the wire resistor loop plug instead. If indeed GB did not require lambda sensors and you do have a resistor wire - it could be your resistor wire has gone bad - I would expect similar behavior as a bad O2 sensor as the resistor wire loop is pretending to be a functioning O2 sensor.

 

[924Srr27l]

yes Its begining to look like this is the problem although I have not eliminated a leaky injector yet. I'll do a leakdown test tomorrow .

I've never seen the fuel pump on mine.... I presume you access it behind the rear wheel arch ?

Yes, I think that's where it is.

All 8 (2.5 / 2.7 / 3.0) of these car's I've owned have always turned over slow and not fired immediately like pretty much every other car does!

A 928 I had also had a starting issue which cranked (faster then the 944's) but the ECU cuts the fuel pump if it doesn't start within 3 seconds or something, I don't think the 944 has this?

My 2.7 944 turns over so slow it sometimes stops, I release the key and then try again after a few seconds. I' don't know why (High compression?) but it always does fire sometimes after two or three goes so I've got used to it ! even when it does fire it's not on 4 cylinders for a few seconds ! and hence sounds like a Harley V twin!

R

 

[IanK]

I have a slightly different issue:

When cold, even after not being run for a couple of weeks the engine starts almost instantly. However after driving it for a while, then stopping for 20-30 mins, it takes maybe 5-7 seconds to restart - this seems like an eternity.

I have purchased a new non return valve, but not yet fitted as it seems it should be harder to start after being left for a week if this was the issue.

Would welcome your views.

Ian

944 S2 Cab

 

[peanut]

Yes, I think that's where it is.

All 8 (2.5 / 2.7 / 3.0) of these car's I've owned have always turned over slow and not fired immediately like pretty much every other car does!

My 2.7 944 turns over so slow it sometimes stops, I release the key and then try again after a few seconds. I' don't know why (High compression?) but it always does fire sometimes after two or three goes so I've got used to it ! even when it does fire it's not on 4 cylinders for a few seconds ! and hence sounds like a Harley V twin!

R

thats definitely not right R ....

My S2 spins the engine over like a demented hampster in a wheel !...

You either have a duff battery,

or your battery and engine ground connections are fubar,

or your starter motor brushes are worn out

or you have bad connections on the solenoid

or the solenoid is fubar

You might try running a jump lead from your battery negative terminal direct to the engine block and see if there is any improvement in cranking speed .It would only take a few seconds and might pinpoint a bad engine/chassis /battery earth.

It wouldn't hurt to clean up the positive lead connection on the solenoid .

If you were to connect the black jump lead from battery negative to engine and then connect your positive (red) jump lead between your battery positive and the positive connection on your starter motor (the connection with the big red wire on the solenoid)

Your engine should then crank over on the key like a Terrier with a rat ......pinpointing bad connections . if it doesn't then your starter motor or solenoid is clapped out.

 

[924Srr27l]

thats definitely not right R ....

My S2 spins the engine over like a demented hampster in a wheel !...

You either have a duff battery,

or your battery and engine ground connections are fubar,

or your starter motor brushes are worn out

or you have bad connections on the solenoid

or the solenoid is fubar

You might try running a jump lead from your battery negative terminal direct to the engine block and see if there is any improvement in cranking speed .It would only take a few seconds and might pinpoint a bad engine/chassis /battery earth.

It wouldn't hurt to clean up the positive lead connection on the solenoid .

If you were to connect the black jump lead from battery negative to engine and then connect your positive (red) jump lead between your battery positive and the positive connection on your starter motor (the connection with the big red wire on the solenoid)

Your engine should then crank over on the key like a Terrier with a rat ......pinpointing bad connections . if it doesn't then your starter motor or solenoid is clapped out.

Yes i do need to try the Direct battery route, but this engine or all the Big 4's are infamous for being Slow to crank over compared

to other engines. Even my 968 was the same.

I tried a WOSP Reduction gear starter motor and it was worst! so I took if off and it's sitting here Brand new apart from the one fitting exercise!

I think the compression is almost 11:1 and it cranks straight away reasonable speed for the first second, then starts to slow up dramatically, then it virually stops to a whuurpp whuurp..!

Even when it fires it's so Pulsey and noisy , there's no silencing restrictions at all.

The Gas has a Straight flow all the way out from the two (Turbo Manifold) 2 into 1 pieces and only (1) straight through rear silencer

if you could call it that. The pipe size is also smaller than the original which was (50.8mm) ID bore , this one made in Finland is about 47mm.

R

 

[peanut]

The Bosch Fuel pump should have a Check valve screwed into the end of it which is designed to hold the fuel pressure when the engine is off.

If the spring has weakened or it has some other malfunction the fuel pressure will be lost and will have to built up when starting (More cranking than normal)

before it's enough for the Injectors to spray at the correct pressure and amount etc... ?

R

Well its beginning to look like it is the check valve causing the hard starting[]

and the ICV valve causing the rich idle and poor emission problems

Fuel system is definitely not holding any pressure for even a few minutes.

i'll connect up the fuel pressure gauge later today to do a leak down test.

Heres an update so far .

I took the car for a a forth retest this morning and eventually managed to get the emissions down to pass level by adjusting the engine air /fuel mix using the vacuum system.

Initial readings were CO- 12.000 HC - 1239

First I tried adjusting the idle richness by turning the idle mixture adjust screw on the base of the AFM but there was no noticable improvement at all.

After the MOT Technician had reset the exhaust analyser I pulled the ETS (engine temperature sensor ) connector off .....no change.

Next I pulled the vacuum pipe off the FPR (fuel pressure relief valve ) to check that the FPR wasn't letting fuel into the vacuum system and straight away the 02 and HC readings started to creep down 02- 8.20 HC 830 . ! well at least its in the right direction and suggests that air supply is the crux of this whole starting and idle issue.

A further reset of the Exhaust analyser ....this time I pulled the ICV (idle control valve ) connector off to see what difference the ICV was making and immediately the 02 and HC readings plummeted to CO - 2.40 & HC - 265 ! Wow

Further fiddling and a forth reading and the CO was now 1.783 the lowest it has been in over 8 years ...

HC was now a meer 138 She had passed with flying colours .

So that would suggest that there is still either an over supply of fuel or an under supply of air when the engine is at idle.

I've ordered a new ICV valve to complement the new ETS sensor which should correct the idle air bypass levels .

In addition I have ordered a new fuel filter and once I've checked the fuel pressure and leakdown I'll probably need to order a new fuel pump and non return valve too.

i'll continue to update the thread and hopefully it might just give someone a couple of ideas to try when they have a hard start fluctualting idle and rich mixture problems

 

[Mike_Pollock]

That's a great diagnosis path!

Mine failed on emisions , but I forgot I'd not put the thermostat back in after fiddling ......

 

[peanut]

oops!..[] another double post apologies

 

[peanut]

oops!..[]

these can be the most frustrating cars when they don't start or run correctly.

I've been pulling my hair out for the past 2x weeks .

I've got to remove the fuel pump and tank strainer and give them a clean. I'll fit a new filter whilst I'm there. I've still got to locate the non return valve yet

 

[peanut]

oops!..[]

these can be the most frustrating cars when they don't start or run correctly.

I've been pulling my hair out for the past 2x weeks .

I've got to remove the fuel pump and tank strainer and give them a clean. I'll fit a new filter whilst I'm there. I've still got to locate the non return valve yet

 

[bmnelsc]

I was reading a Rennlist forum post about running rich last night and in this case the trouble was found to be a poor ground on the MAF. The owner had cleaned and tested his MAF (static tests) but had not tested over the range of operation (voltage output from the MAF changes over the range of travel from idle to WOT). After checking all other grounds (including DME ground) he temporarily moved the MAF ground to another location and found his problem solved. Another point raised here was the measure of vacuum at the FPR - too little vacuum can lead to rich condition at idle/part throttle. Good Luck!

 

[bmnelsc]

Any further news?

 

[Waylander]

The check valve is only for when the engine is hot

it should hold pressure for about 15-20 mins

this is to stop the fuel boiling in the injectors when they are very hot, if it boils you get a vapour lock and it becomes an ass to start.

there is an American company that sells electronic DME relays that prime the fuel system at ignition on.

i also seem to remember that the fuel pump will not be switched on by the ECU until 200 rpm is reached at cranking