Strange idle problem...?

Right, we have a 97 S70 2.5 10v albeit a manual. I can tell you that your idle is not as it should be- starting from cold it immediately jumps up around 1300rpm and then instantly settles back to around 1100rpm which then slowly decays back to around 900rpm when fully hot.

Starting from hot it jumps up to about 1500rpm (some lag on the tacho) then instantly decays back to 900rpm. It does this because the ECU motors the ISCV to the fully open position about 10seconds after you switch off-ready for the next re-start.

Anyhow you have one of 3 problems, and possibly someone has been there before you to attempt fixing the underlying problem.

1) the inlet manifold gasket has sucked in, this is fairly common on the 2.5. Check with some soapy water sprayed around the manifold to head joint, and allow some to run around underneath. Any change in engine note and you have found the problem. Or one of the other vacuum hoses has fell off or split. There is one connection behind the PAS pump part of the PCV system which also likes to split and fall off. Its not easy to see and harder to get at without removing the manifold. 2) The TPS is faulty, or the wiring is. However this should set a fault code. A TPS which doesnt return to giving an "engine at idle" signal to the ecu will cause hunting and a stuck high idle. 3) Someone has messed with the butterfly base position to mask a stalling problem caused by 1) or 2)

It would help if could give details about mileage, starting performance, general drivability and mpg you're getting...According to the computer ours gives upto about 35mpg on a long steady run, about 30-31 around town, and after a recent 400mile fast motorway drive (90-100mph all the way) was showing 29mpg.

I would disregard headgasket problems, though you could do a leak down test for your own piece of mind...

On our car I would add it suffers from slight hesitancy when accelorating under load under 3000rpm with less than 1/2 throttle, immediately after cold start which it shouldnt do- feels like a lean condition, though I'm going to renew the leads /cap/rotor arm first. Its done 68k miles.

Hope this helps to begin with.

Tim..

I'd say that's how it used to behave before I broke the original IACV ;-) and put on the replacement. The idle speed, even then, seemed about right.

I wonder if it was more a fuelling thing for the idle / stumbling thing, than purely the idle speed set by the IACV.

Well I've had the car about 3.5 years, and it only started doing the idle stumbling thing late summer. In that time, the garage I use has looked at it twice, once during an oil change, where they couldn't find anything apparently wrong, and read the fault codes and the only thing logged was an intermittent vehicle speed sensor fault - reading my Haynes manual was confusing about this, because it said that there wasn't a specific VSS, the Haynes book said that it used the ABS sensors. The second time the garage looked at it was to check whether the head gasket had gone, and it's then that they found the leaking rad, explaining the water leak.

Wouldn't air leaks always show the problem, though?

As I said, it hasn't done the poor idle / stumbling thing since it's had the rad changed. And the idle speed is OK when it's under load, and it's been running a few minutes. It's only when cold for a minute or so after starting, or when it's truly warm when not under load, that the idle is either too low, or too high.

It doesn't seem to hunt right now - I'm inclined to think the incorrect idle speeds at the moment are a result of something not being right with the IACV I replaced.

The garage only got errors the first time (with the intermittent VSS code), the second time they said there were no codes stored.

The idle speed only went AWOL after I broke the original IACV (which I'm inclined to think was actually OK, now), and replaced it - that occurred a few weeks after the garage had first looked at the stumbling problem, and not found anything relevant, and I assume hadn't done anything, because nothing changed about how the car behaved.

I think I'm solely responsible for messing up the idle speed when replacing the IACV because only then did the idle change.

Sure - sorry. It's done about 168k miles, and for years has always started on the button. It's only recently on the days when it would stumble, that it would need cranking to start.

At the moment, it starts willingly, with just a little throttle, because as I said, it idles a little low (albeit it manages, but makes the initial start more demanding) when cold.

When warm it starts quickly - no cranking.

The engine seems fine - still quite willing, nothings really changed on that respect. It gets oil changes every 5k miles. Perhaps overkill, but I put quite a few miles on it, and would like to carry on for the foreseeable ;-)

I'd say it's a little under what I'd normally expect - perhaps by 1-2 mpgs. It does go up a little when it gets colder, but my gut feeling is that it's still a little higher than I would expect.

It's currently around 28-29mpg - and I'd probably expect it to be slightly higher. Not by much, but I'd say it's just marginally higher than I'd normally expect (I've had it for around 3.5 years and around 86k miles).

The garage checked that out - they were pretty sure it was OK.

And since it's had the rad changed it's not stumbled. The idle behaviour hasn't changed since then - and that only occurred after I broke the original IACV, and put in the replacement (albeit a used part). The only thing noticable since the garage changed the rad, is it hasn't stumbled, and there's no water leak.

Mine had more than that when I bought it ;-) (82k miles, 3.5 years ago).

Many thanks for your response, Tim - it's appreciated.

Hi again,

I think your rad change has something to do with the problems- I would check

*very* carefully the vacuum hose connection to the MAP sensor, which is under the bonnet slam panel, along with the cooling fan controller (and cruise control relay if fitted) Remove the clip on the intake manifold end and check the passage into the manifold, and also the rubber pipe ends for rubbing / splits etc which could open up, and the solid white plastic part. Beware the plastic stub on the map sensor is brittle.

This all has to be disturbed to R&R the radiator, so it may have temporarily solved the problem.. (radiator on our previous 850 leaked too- well rotted- its a common problem)

Our map hose was resting on the fan shroud and wearing a hole in it. Also check the 3 pin wiring connector for dirt or bad connection, do the same witht he TPS connector. These are located down low and often get wet / salty from road spray through the radiator.

I would also do a wiggle test on the hose and map sensor wiring / etc with the engine at tickover and see if you can replicate the symptoms at will.

I am surprised the ECU is not setting an "idle adpative limit exceeded" or "excessive idle compensation" code if indeed it is struggling to hold the engine at its programmed rpm...

I would re-check the ISCV and verify the throttle body is spotlessly clean then disconnect the battery and reset the ECU to make it relearn the idle characteristics. Make sure you allow it to start from cold without throttle, then drive normally for 6-8 miles, finishing off with allowing the engine to idle for about 5mins hot.

Abit to be going on with!

Tim..

Well as I said, it's not done the stumbling thing since it's been back from the garage for the rad change.

And as I had been chasing that problem for the best part of a couple of months, I'm currently going to follow the - if it ain't broke don't fix it ;-) It's just that from that perspective it's working fine right now, but if the symptoms re-surface I'll follow your helpful advice - it is appreciated.

Well I've replaced the already replaced IACV with another one. And the odd idle (too low when cold, too high when warm, and dropping under load) has gone away.

It now idles correctly when cold, doesn't dip when put in drive, and idles correctly when warm / hot.

I can still see some fleeting instances where it seems to think it's still got to behave as it did, for example when slowing to stop for traffic lights - the idle when cold can still go a little (but only a fraction) lower than I think it should, but as soon as it's got a few beats, it corrects itself.

I do wonder whether that's any learning / adapting by the ECU, and wonder whether it's worth pulling the ECU to clear anything, and let it re-learn? If that's feasible, do you know how long the ECU has to be removed for, for it to have lost anything like this stored?

Well it's only been a matter of weeks since I thoroughly did that, so hopefully should be OK for a while.

Many thanks, Tim - your help is appreciated.

Strange as it may sound, I find it easier just to remove the ECU for a while (do you know how long it should be?), rather than disconnecting the battery ('cos the terminals are all gungy, and I'd need to enter the radio code).

Since the replaced IACV, and the idle speed(s) now being spot on, it starts immediately, with no need to crank or use the throttle. And so far, I've not had any stumbling for a good while, so I'm ever hopeful that that's cured too.

Cheers.