Fuel tank breather valve / slack tappet simulator

Special Circumstances

Any tips on testing one of these to see if it is working ok? Or temporarily disabling it?

Seems like the car (mini r50 2003 1.6 petrol) will throw a code if it is electrically disconnected. Probably raise an error anyway if I do the following:

Is it ok to pull the vacuum line between the valve and the engine, blank it off to prevent vacuum leak and drive it a little to see if a funny idling/stalling problem goes away?

Assume I'm an idiot who knows what a socket set is but not much more in terms of engines. For example... what would be a suitable way to blank off the vacuum line. Never done it before and would rather not see whatever I use being sucked into engine...



Delphi fuel tank breather valve, common to several GM cars including saab. Rover / MG enthusiasts christened it the "slack tappet simulator" as the solenoid pulses in time with revs sometimes and can make quite a worrying noise. Mine seems to be at this quite a lot of the time, at least that's what I hope the noise is!
Tbh it can rattle all it wants if I can track down this funny engine stumble.

shietpilot

Special Circumstances wrote: »

Assume I'm an idiot who knows what a socket set is but not much more in terms of engines. For example... what would be a suitable way to blank off the vacuum line. Never done it before and would rather not see whatever I use being sucked into engine...

A bolt with a big head into the vacuum line/hose. Then a cable tie or worm gear clamp on the house to squash the bolt.

Special Circumstances

Anyone else got some pointers for me on this pleeeeease :-)

The fact that I can hear that valve tack-tack-tacking means that the solenoid is working to some degree I suppose.
So... maybe
A) not closing right when you clutch and brake and come to a stop. ( I think it should close during fuel cutoff for deceleration)
Or maybe thats fine but when it starts pulsing again at idle there is a vacuum leak somewhere in those evap lines.

Anyone have general pointers for what goes wrong in these evap systems?

I have 3 things I can change here -
1)unplug electrical plug. If car runs ok I guess it's ECU software problem pulsing that valve at the wrong time. Or it is not closing properly when commanded.
2)block vacuum line to engine
3) leave valve connected to line to engine, disconnect line to charcoal canister.

I'm not fully sure what will happen if the valve is good/bad in those different experiments, only one way to find out unless someone here knows the answers! Will have a bash at it over the next few days, see what I find.

Waterson

I gather from your post the issue you have is an occasional power loss or drop in revs perhaps at idle? It is normal for the tank evap solenoid to be actuated at idle. I cant recall the layout of that mini engine, but i assume it has a hard plastic line going from the charcoal cannister to the inlet manifold. I use the plastic protective caps you get on new fuel filters to plug those lines, though i have used blobs of bluetack (larger than the hose diameter) before. I thnky you may be barking up the wrong tree though, can you give us an ide aof your symptons?

Special Circumstances

Yes, I think I am unfortunately barking up the wrong tree.
I was hoping you'd see this actually - you have made some very interesting debug related posts in other people's threads. I'd be very grateful if you have any insights for me.

The slight tapping I'm hearing is possibly from a worn chain / tensioner. I heard unplugged the valve and could still hear the noise unfortunately. Wishful thinking, the valve just happens to be next to the timing chain.
Pulled the evap-to-purge-valve line and got a P0444 purge valve circuit open code the minute I started the car.

Cold start issue only. Variable in it's severity.
Start car, drive a few meters to approx 200metres. Brake and clutch, coming to a junction etc - rpm will drop way down 400? 500? Won't stall but will shudder and stutter. Sometimes shoots back up on its own to approx 1800rpm.
Will continue to do this if you are shuffling around inside a multi-storey or whatever.
Will clear as soon as you get a chance to drive it on. So sometimes it does it once, max twice in the housing estate in the morning, but never as often or severe as in the multi-storey in the evening.

Idle was bad on it for a while too - don't know how long that has been an issue - factory idle speed 800 was increased to 900 years and years ago and bad idle only came back when someone reset it to 800 recently. Don't know if that's related or separate.
Online forums suggest many struggle with the 800 idle, and that the funny "almost stall" thing is not rare. Some people suggest a dodgy map, dodgy throttle body controller, dodgy evap circuit.

Plugs, coil(DIY), leads(DIY) done (quality brands)
Fuel filter done. (DIY)
Throttle body cleaned (DIY)
K&N cleaned and re-oiled (DIY)
Inlet manifold gaskets replaced, and injectors seals sealed up a bit better (not replaced) by a decent mechanic.
New battery.
New chain tensioner - chain was making a little noise on startup until this was done.

Most diagnostics will see no codes - one specialist was able to see a "fuel formation"??? code and advised cleaning the K&N.
Hard to debug - I'm rarely there when it is at its most vexing, and most mechanics have failed to reproduce it, either due to it being not cold enough. Even if they did the window you have to figure it out is quite short.

Looks like it's either learn to live with it or start replacing expensive things, unless a mechanic can come up with a dodgy looking signal from something.

Special Circumstances

Wait now.... progress!
No stuttering or stumbling idle this evening!

Ok. Was worse than usual this morning. Has been doing its trick every evening for a long time.

Disconnected canister to valve hose. Plugged the hose to avoid P0444. No almost stalling at idle. Accelerating was a little off. I guess I've pretty much introduced a vacuum leak which is compensating for something at idle but causing problems when driving.


So...

Charcoal canister blockage?
Throttle body doing something weird at first few idles?
Tmap sensor over estimating the amount of air at idle for the first while?

Seems like option 1 to me but it's such an obscure one.. I've never heard of anyone in ireland having problems due to charcoal canister!!

Waterson

Ok, maybe get your mechanic to do another plug in. Check coolant temp sensor readings and also fuel trim values when idling. These might give an indicator of what the ecu is doing. It sounds like a fuelling issue, but I do believe these suffered from bogey fuel maps on the earlier models. There is a software update available iirc.

Special Circumstances

Waterson wrote: »

Ok, maybe get your mechanic to do another plug in. Check coolant temp sensor readings and also fuel trim values when idling. These might give an indicator of what the ecu is doing. It sounds like a fuelling issue, but I do believe these suffered from bogey fuel maps on the earlier models. There is a software update available iirc.

Your are correct about the maps being a work in progress over the first few years.

I have a ds150e or something claiming to be hahaha and delphi software. Driving around the estate with a big donkey of a laptop showed nothing glaringly wrong in throttle position, coolant temp or short term fuel trim the few times I managed to reproduce it.

Now, this isn't my area of debug expertise, so it would have to be a gross error, I doubt I'd pick up on anything subtle.

Waterson

I have a ds150e or something claiming to be hahaha and delphi software. Driving around the estate with a big donkey of a laptop.

Ha, Ive been there. Not familiar with the ds150e, but delphi software should be able to give you livedata.
You're looking for the coolant temp sensor signal voltage to change proportionally with the engine warming up.
In regards to fuel trim, anything + or - 10% on the short term (adaptive FT its probably called within the BMW world) at idle might give you a clue as to whats going on.

Special Circumstances

I find the delphi a bit frustrating. Update rate seems poor too. (could be the unit I guess)
I had a graph mode but that seems to have disappeared. So all I have now is a list view. off topic, but if I had payed my own good money for it I'd be disappointed at the quality of the software.

As far as I can see coolant temp (which I get in degC rather than voltage) is plausible.
I haven't seen short term FT go beyond +-5. The , errr, copilot was looking at the values on the day it did its worst example of almost stalling so maybe didn't know what she was supposed to be looking at. From the glance I gave at the numbers there was no mad spike in them.

Might see if I can pull out that charcoal canister at the weekend, see if there is anything obvious wrong with it.