It's back! Expert advice please

[puffin]

Nearly two years ago now I wrote this post:

http://www.scottishelises.com/phpbb/vie ... &sk=t&sd=a

The problem's now back with it dying twice in a week (once in the Clyde Tunnel and last night in the outside lane of the M8 in heavy traffic ).

Ironically I was driving it back from a C service and MoT pass that required no extra work (first time ever!) and had told the garage about it but they couldn't reproduce it.

So I have to take it back on Tuesday to be fixed and it would be great to go armed with a theory or two of what it is. Suggestions so far: dodgy fuel pump or leaky inlet gasket.

Full symptoms:

Drives fine normally and will sit in traffic no problem when the engine is cold.
Fuel pump does the usual noises when the key turns (although has on occasion made very strange sounds at the same time).
Sitting in heavy traffic after a long motorway run (so everything is up to temperature) it will just stall without warning (usually when applying some throttle as compensating by pressing on the throttle is how I notice that the pedal is not, actually, doing anything and the engine is dead).
Fails to restart (although the starter turns, it just doesn't catch)
After waiting about 5mins it will catch but if it's not allowed to cool down sufficiently it will die fairly quickly.

Help!

[campbell]

My Dad had a Scirocco that did this. Turns out the fuel pump was overheating. On the Scirocco, though, the pump sat outside the fuel tank so got cooled by passing airflow...so on m-way etc, it ran fine. Once in traffic, though...doomed.

Not saying this is deffo the prob, not nearly qualified to do so, but just interesting comparison I suppose.

With the Elise pump inside the tank, not sure what this means re: overheating and cooling as presumably the petrol still acts as a reasonable heatsink? OTOH, if the pump is degrading, and running hotter than it should, then maybe the natural tank heatsink is no longer sufficient. Hmmmm.

Out of interest...does it make any difference whether the tank is full or near-empty?

[Gourlay83]

My S1 did the same thing. When the car got hot it would cut out, leave it to cool and it starts fine.

The problem with mine was the Idle Air Control Valve IACV. It was sticking when hot and cutting the engine out.

I read through your last post and noticed there was a breather hose split, is it the one that goes to IACV. If so, its probably let loads of crap into the Valve.

Cheers
Alan

[ABZ-Elise]

I'd say it was the fuel pump. When I first got my Elise I had a similar problem. At first I though it might be an air leak but it turned out to be lack of pressure from the fuel pump. It first only happened every now and again then it started happening every few miles on a trip. The car would cut out while driving and you'd have to wait a while before it would start again. The pump would prime every time I turned the key but not with its normal sound...more of a grinding sound. I swapped my pump out and its never happened since.

But before you decide to go and get a new fuel pump fitted, make sure you've checked everything else first because a new pump will set you back about £300 and it would be a shame to spend that if it doesn't solve the problem.

[robin]

I agree that in this instance the IACV is the most likely cause (normally I would have said fuel pump wiring - BUT you say the fuel pump primes immediately after stalling each time).

If you can identify the IACV and disconnect it's electrical plug a couple of minutes after a cold start (i.e. when the idle speed is still higher than usual).

On the S1, the IACV has a 5-wire connector. It's near the throttle body and has a 3/4 inch diameter black rubbery air hose connecting it back to the throttle body.

If you stand behind the car, boot lid up, looking at the engine if you can identify the oil filler cap and then move backwards in a straight line to the big metal inlet plenum you'll pass across the top of the IACV in the process.

With that disconnected, you may find cold start is a little trickier, and the idle will probably remain a bit quicker than you want, but crucially you can see if the car ever stalls. It may also pop and bang a little more when you lift off the throttle at higher engine speeds - probably won't though.

Cheers,
Robin

[puffin]

So are you saying that one way to diagnose it would be to reproduce the problem then disconnect the IACV - if it starts then it's the IACV that needs sorting?

[robin]

No, I'm saying that if you disconnect the IACV at the next cold start (i.e. as soon as the engine has been started, then disconnect it) and leave it disconnected, then drive around for long enough so that you would typically expect to suffer a couple of stalls. If you get no stalls, chances are it's the IACV sticking when hot. If it still stalls, chances are that it's nothing to do with the IACV.

The IACV is like a water tap, but for air, obviously - it allows a small amount of air to bypass the normal throttle and enter into the engine anyway - the engine management computer (known as the ECU) controls how open or closed this tap is. When cold idling it's open quite a lot to allow for an increased idle speed. When warm idling it's open less, because less air is needed to idle a warm engine at lower speeds than a cold engine at higher speeds. When you open the throttle, the ECU will mostly close the tap completely (though in fact it will sometimes open it for other reasons, I believe). So normally when you come off the throttle, both the throttle and IACV "tap" are closed and the engine speed starts to drop. When you hit around 1,600 RPM in this condition, the ECU will start to feed fuel back into the engine, and it also opens the IACV to prepare for idling. If you continue to keep the throttle closed and the revs reach around 1,000 RPM, the ECU goes into idling mode and opens/closes the IACV as required to maintain a steady idle speed of 800 RPM or so. How much air is required depends on what loads are on the engine (is the alternator doing more or less work, are you trying to pull away using just the clutch and no throttle, etc., etc.).

If the engine and IACV are hot when the IACV is closed, it may be that when the ECU commands the IACV to open up, it just doesn't, perhaps due to crap in the IACV combined with the effects of expansion. However, if we disconnect it while the IACV is clearly open and cold, then it will never get closed (because we've disconnected it from the ECU) and we don't care whether or not it might stick (it will just stick open) and so the engine will never stall (we hope!).

Hope that makes sense!

Cheers,
Robin

[puffin]

That makes a whole lot of sense, thanks Robin.

So much in fact that I just did what you said and took it out for a test.

Started the engine, removed the connector from the IACV and went out for a 20 mins lively drive to get everything up to temperature. Pretty much immediately I can feel a difference in power (which I wasn't expecting) and notice that the engine is now cruising at 90C instead of the normal 96C (also unexpected). I get home and stop start around the car park for 10 mins with the engine noisily idling at 1400 revs without incident. The engine gets up to 110C at one point but the fan kicks in and keeps that under control. Success - it's the IACV. Except just as I reached my parking space and turned in the engine stalled. And wouldn't restart. And the fuel pump isn't priming - I can hear the relay clicking but the pump isn't doing anything.

So there you go - it's the fuel pump? but the IACV probably needs a clean too?

Anyone with enough knowledge to say definitively from that description?

[robin]

IACV is probably a red herring. You should reconnect it.

When you say it won't start, does it even crank?

The fuel pump doesn't prime on stall unless you have turned the ignition off for about 30 seconds or more.

So, a proper test would be:

Ignition off for at least 30 seconds.

Disarm the alarm/immobiliser as usual.

Turn ignition to run position (but not start).

Listen for the fuel pump.

Turn key to start position and listen for it cranking/turning over.

Tell us what you hear.

Cheers,
Robin

[puffin]

Test 2 complete.

Luckily the engine hadn't cooled down much so I was able to induce the stall fairly easily - I did it twice to make sure it did the same thing.

When it stalls the buzzing noise that I presume is the fuel pump goes totally quiet and revs drop to about 500rpm for a couple of secs before the engine shakes itself dead.

Waiting 30 secs with the ignition off and then turning the key results in a very feeble priming noise from the pump - about half the duration of normal and tailing off in volume as if it's struggling. However, the engine did restart after a couple of seconds of cranking (only to stall again a minute or so later).

[robin]

OK, so it is the fuel pressure I would say ... most probably caused either by the fuel pump, or an issue with the wiring. I'm just going to have my dinner - will post up procedure for inspecting the fuel pump wiring later (you could try searching for fuel pump wiring in technical - sure I've posted it before ).

Cheers,
Robin

[robin]

I knew I had posted this all not so long ago:

another fuel pump stops when heated up thread

Cheers,
Robin

[robin]

BTW, in your case, the only thing that is (probably) relevant is taking the connectors apart and looking for signs of corrosion or overheating - there will be +12 present at all points (I'm guessing) as your pump "sort of" runs - so there's voltage, just not enough ability to supply current to make the pump run properly.

Cheers,
Robin

[puffin]

Oh well that's waaaaaaaaay too technical for this girl, but I'll pass the contents of this thread onto the garage.

Thank you for your help, robin.

[campbell]

Oh well that's waaaaaaaaay too technical for this girl, but I'll pass the contents of this thread onto the garage.

We can only hope that they can understand it...Robin has been known to be drafted into garages to help debug auto electrics when their skills have run out