90s Alfa 145 boxer - weird tickover problem - any sensible suggestions?

[Nomad]

Hi, do we have any classic Alfa experts in the house ?

Alfa Romeo 145, mid 90s classic, I've had it about 3 years and it's a fantastic car when it's working right
Engine is a 1.6 Boxer motor, 4 separate injectors, it normally works very well.

Odd tickover problems
It misbehaves when it thinks it's been getting ignored, now at tickover the revs drop then rise to 3k, then drop and ... repeat ... continuously.

It's happened before but we never worked out what we did that fixed it. I spent several days checking and tightening everything - no difference, called a mobile mechanic who redid my work - he gave up - it started working before he left the carpark
Last time it happened I arranged to put it over a pit at a mates garage, I got there, put the car over the pit, opened the bonnet (never touched anything!) and the problem went away

At anything above tickover it runs very nicely so the problem seems to be something that's specific to the low RPM / idle system.
There's no obvious inlet leaks, I blanked off the brake servo on case that was the cause but no difference, exhaust manifold & joints are tight, all wiring connections are fully home and feel ok.

If anyone has experience and knows how to fix it, I'd appreciate a message

Suggestions to sell it and buy something sensible like a Ford are not needed

 

[bricol]

You've been on the Alfa Forums? If its a common issue, it'll be mentioned.

Seen mention of a spring loaded valve in a connection from the cam cover to inlet manifold sticking open erratically, letting extra air in and increasing revs.

Mates golf of similar vintage, but back when it was newer, had a small hole worn in the servo vacuum hose - used to open and close, causing erratic revs.

 

[steveo3002]

sounds like an air leak , have you considered a smoke machine to rule it out ?

 

[Blobber]

Idle control valve I should think

 

[dave h 2]

The +ve crankcase vent valve fugged up or as Bobber above + the piping from the idle valve to the manifold might be a little oil affected inside or is there an anti stall that's vacuum powered

 

[dobbslc]

It does sound like an air leak that it's trying to correct. A smoke tester would find any holes I think.
I nearly bought a 145 Cloverleaf years ago, 2.0 engine in that one a bloody hell it shifted!
Ended up with a 155 Silverstone narrow body in the end.

 

[Nomad]

Good to know we have so many Alfa fans here - I knew there were a couple but more is always better
Thanks for all the suggestions so far, replies in no particular order :

I looked on the Alfa-owner forum but didn't see anything, I also asked the question but it's not very active. The other Alfa forum doesn't work for me - I've tried to register a couple of times in the past but never get/got any activation emails.

Idle control valve is free and cycles fine when ignition is turned on/off, it had a thin mist of oil so I cleaned it - no difference. The vents to/from the valve are clear, no blockages.

Spring loaded inlet valve (is this another name for the idle control valve??) - I looked but didn't see anything obvious. This is a boxer motor so the manifold is fairly simple, looking at the manual the other motors (Twinspark etc) are more complicated - maybe the valve is only on those??

+ve crankcase vent / anti-stall valve (another name for Idle control valve or a different thing?) not sure if they are fitted or where to find them but I did look for them.

Vacuum servo hose has been blocked off to test it, no difference.

I'd agree it sounds / feels like something it's trying to correct until it overshoots then it resets and does it again ...
I tried spraying easystart around the manifold joints thinking any leak would draw that in but it made no difference.

I'd heard smoke test mentioned before but could somebody explain it - how does a smoke test check for leaks? There's always a draft under the bonnet due to belts, pulleys, fan etc...

I've just spent an hour removing all intake related sensors (MAF, throttle position, idle control valve) and cleaning oil mist from them, unplugging & reconnecting all likely connectors including lambda, also looking for any vacuum lines that had been knocked loose or split. I'm happy the intake is now now clean and oil-mist free but the problem is still there.

 

[steveo3002]

you would introduce smoke to the inlet system with the engine off , block off any know outlets and see if smoke appears from a leaky gasket etc

 

[steveo3002]

does it have a switch on the tb for idle , some older cars do and they need to know the throttles closed to idle properly

 

[dave h 2]

Good to know we have so many Alfa fans here - I knew there were a couple but more is always better
Thanks for all the suggestions so far, replies in no particular order :

I looked on the Alfa-owner forum but didn't see anything, I also asked the question but it's not very active. The other Alfa forum doesn't work for me - I've tried to register a couple of times in the past but never get/got any activation emails.

Idle control valve is free and cycles fine when ignition is turned on/off, it had a thin mist of oil so I cleaned it - no difference. The vents to/from the valve are clear, no blockages.

Spring loaded inlet valve (is this another name for the idle control valve??) - I looked but didn't see anything obvious. This is a boxer motor so the manifold is fairly simple, looking at the manual the other motors (Twinspark etc) are more complicated - maybe the valve is only on those??

+ve crankcase vent / anti-stall valve (another name for Idle control valve or a different thing?) not sure if they are fitted or where to find them but I did look for them.

Vacuum servo hose has been blocked off to test it, no difference.

I'd agree it sounds / feels like something it's trying to correct until it overshoots then it resets and does it again ...
I tried spraying easystart around the manifold joints thinking any leak would draw that in but it made no difference.

I'd heard smoke test mentioned before but could somebody explain it - how does a smoke test check for leaks? There's always a draft under the bonnet due to belts, pulleys, fan etc...

I've just spent an hour removing all intake related sensors (MAF, throttle position, idle control valve) and cleaning oil mist from them, unplugging & reconnecting all likely connectors including lambda, also looking for any vacuum lines that had been knocked loose or split. I'm happy the intake is now now clean and oil-mist free but the problem is still there.

Vacuum lines need checking for blockages. Is there a small 15mm hose going into the air intake just past the filter box from the block?

 

[dobbslc]

Thread 'Vacuum leak smoke machine.' https://www.mig-welding.co.uk/forum/threads/vacuum-leak-smoke-machine.135696/

 

[Bullet2012]

Are you using super unleaded or standard fuel ?

 

[roofman]

Can normally find an air leak by spraying brake cleaner around the inlet area on idle, sucks it in and rpm rise.

 

[Nomad]

Hi. It has a throttle position sensor - I can’t remove it from the mani without major dismantling but turning it gives a very clear tickover / free revving engine so I’m saying it’s working.

Petrol is fresh 95 and has worked fine since I got the car a few years ago.

Will try to get time to look again at the breathers later.

Thanks for the info about the smoke test - I’ve been educated !

 

[pedrobedro]

Fluctuating idle like that is usually MAF sensor if it has one. Just try disconnecting it. I've never had any success trying to clean one so they get replaced.

 

[Bullet2012]

Petrol is fresh 95 and has worked fine since I got the car a few years ago

Understand,
However in a vehicle of that age I would suggest using super unleaded with only 5% ethanol.
Reason, the higher ethanol content in other fuels can gradually damage many components within the fuel system, from flexible pipes to pump internals and more. Especially things like diaphragms and vanes within the pumps.

 

[Nomad]

Petrol - good point. I've had the problem a couple of times in the past and it cleared so I was thinking it's mechanical or a loose connection. Something affected by the petrol that blocks then unblocks is a new idea

MAF etc, I tried unplugging each sensor in turn to see if it fixed things, it didn't (so either they're all ok or one has failed open circuit so result is the same as disconnecting it.

More investigations needed ..

 

[dobbslc]

I've had a very bad tank of petrol from Sainsbury's once and a few mates the same. I won't use them again.

 

[ajlelectronics]

Hi. It has a throttle position sensor - I can’t remove it from the mani without major dismantling but turning it gives a very clear tickover / free revving engine so I’m saying it’s working.

I chased a fault like that for months on my Cologne 2.9i. It turned out to be THREE faulty but otherwise brand new TPS. I sussed it in the end because disconnecting the TPS should put the engine into service mode, but there was no response to disconnection. A secondhand original proved the point.

 

[Nomad]

The Alfa is now happy again, a shame I'm not sure what fixed it

All the work I did was around the inlet manifold (removed the Idle control vale & refitted it, unplugged the throttle sensor & reconnected, also removed the intake pipe form the inlet manifold - when I started it with no airbox/filter it was idling as it should, airbox & intake refitted and it's still good.

My best guess is I've disturbed a gremlin that was hiding around the intake and it's gone away sulking - hopefully never to return

Thanks for all the ideas & help