Ticking over like a bag o' spanners - Fixed

I had it fully up to temp with oil pressure just under 1 bar the other day, ragged all over the place, still ran fine. Last night gave it some when fully up to temp and still fine.
You can tell now within a 2 minute drive from cold whether it's right or not just by giving it a rev and watching what happens when it drops.
Good - drops straight down to 1k then slowly down to tickover.
Bad - straight down to about 400rpm, nearly conk out then up to tickover and vary.

:frustrated:

I think I've fixed my car :grin:

:boogie: :boogie: :boogie: :boogie: :boogie:


Not saying too much yet until I'm confident the lumpy tickover has been resolved.

Car feels totally different though after 1st couple of rides. Will update in the next day or so.

13 months???

would it not be better to get some rent $$ out and go to a decent garage to get checked over properly in case its doing more damage.

I just could not live with that for 13 months.

Yeah!! Although for a long time, the fault was intermittent which made it more difficult to diagnose, but at the same time ruled out (IMHO) most non electronic issues.

Still not 100% convinced I've sorted and also not 100% convinced I know how I sorted it but will probably report back tonight after I've driven it a couple more times. :?

NLW73 said:

would it not be better to get some rent $$ out and go to a decent garage to get checked over properly in case its doing more damage.
I just could not live with that for 13 months.

Click to expand...

Would've been an option if I wasn't currently living below the rent line, although I did take it to a garage within that period (last year) and no codes were showing on the ECU. Read the codes since and nothing showing. If the ECU is controlling the running of the engine within the programmed parameters, then nothing will show, even though this may show slight idle issues. Most members wouldn't even have noticed it, especially if they're the ones who constantly running their HVAC on 'Auto' as it shifts the revs up on tickover (due to running the compressor).

My car had some of the symptoms yours had, albeit not the same degree of rev fluctuations, but it had a lumpy idle when warm. Durametric did not highlight any errors, but after checking the car thoroughly my Indy suspected it was lambda related and recommended changing them.

Based on this I changed both lambas as part of my exhaust swap and I'm fairly certain it has resolved the issue as the idle is now stable whatever the operating temperature.

alex yates said:

I had it fully up to temp with oil pressure just under 1 bar the other day, ragged all over the place, still ran fine. Last night gave it some when fully up to temp and still fine.
You can tell now within a 2 minute drive from cold whether it's right or not just by giving it a rev and watching what happens when it drops.
Good - drops straight down to 1k then slowly down to tickover.
Bad - straight down to about 400rpm, nearly conk out then up to tickover and vary.

:frustrated:

Click to expand...

Out of interest mate, have you checked what voltage is showing when it runs like crap? I noticed a few weeks back, I had been on a long journey, stuck in traffic a lot and with two phones charging from the car. Battery dropped down to like 13.5 while running and I started getting some idle issues like you describe. Unplugged the two phones draw and it went away?

Been an interesting read spanning you over the course of a year Alex. Do you now have a sweet running car?

Itching to know what the problem was in the end.

I had a similar situation on my race car a few years back, sadly it cant be related as I think that was a vac leak. so a mechanical issue which cant be scanned.

If its still being an issue - has it actually been to an OPC? dig deep I know, but a persistent problem is just a ball ache

THAT WAS IT!!!

My post cat lambda sensor was chuffed, it didn't bring a fault up so I replaced it, then still a bit rough so I put a 10mm spacer (from memory) and it completely solved the problem. Or the spacer was for a different issue - either way. rough idle was lambda sensor

Niggling issues are just the worst hope its sorted :judge:

Normal service has been resumed :sad:


So - this is the script (any advice welcome):

Yesterday I had some spare time so I thought I'd have a look at where the knock sensors are, with the intention to change them if they were the guilty parts.

Here's the sequence of operations to help me understand what's going on:

1. Car up on ramps (just so I'm not bent over all the time).
2. MAF disconnected & airbox removed.
3. Throttle body disconnected and removed.
4. Vac pipes disconnected from rubber plenum sleeves, plenum and each rubber sleeve removed.
5. Knock sensors located and disconnected.
6. Oil sensor disconnected and electrical connector to vac pipe block above the alternator also disconnected to enable me to move wires around and help with access to knock sensors.
7. Both knock sensors removed, inspected for any cracks, etc. and mating faces cleaned up with some emery paper.
8. Knock sensors replaced the opposite way round (L on R & R on L).
9. Everything else reconnected, plenum cleaned out and refit, throttle body refit and airbox. Reconnect MAF, throttle body, etc.
10. Turned over and fired up perfectly.
11. Went for a 5 mile spin, engine ticking over beautifully - plugged code reader in and watched live data - all stable figures and left/right banks identical.
12. Had my Tea, went for a 12 mile spin, got home, car still ticking over great.
13. Went to work this morning, tickover on arrival degrading.
14. Came home this afternoon and running lumpy again. Plugged code reader in again - live data show fluctuating fuel trims and left bank in minus figures whilst right bank in positive figures.

So - what's the verdict?

Here's a few ideas from me:

(ps. car was running beautifully last night when I went for a spin)


1. One of the (or both) knock sensors is playing up and causing the problem. Taking them off, giving them a shake and refitting them temporarily sprung them back in action.

2. Disconnecting one of the other components for a couple of hours and reconnecting it has reset something and after a few miles of driving has tuned it back to how it was.

3. Oil getting in through the vac pipes into the plenum and cleaning it out yesterday made it run fine, until more oil got back in again. Evidence of oil in the plenum when removed was only light coating on the lower inside faces of it .

4. The plenum was fit incorrectly and causing a vac leak. refitting it stopped the leak until it's gone through the motion of getting hot/cold/hot/cold.

:dont know:


I'm in the position of:

1. Buy two new knock sensors, fit them and see if it cures it.

2. Remover the airbox, throttle body & plenum AGAIN, clean out and see if the car temporarily runs ok.

3. Torch the F*!£ker!!!!!!! :gunfire:


:frustrated:


Any ideas guys??????????


Oh - and to top all this off, my right back box sounds like it's started blowing like crazy, even though it's a 2 year old Dansk and no signs of any leaks in the hole system.

This car is starting to get to me :nooo:

Some info on knock sensors ..

http://vems.hu/vt/help/v3/v3_knock_control.html

I think its unlikely the knock sensors are faulty .. possible but very unusual and with out a fault code ? hmm ...

Only way to prove is to scope them im afraid.

I take it you tightened them up to 20nm , they are pressure sensitive so over tightening them will break them , and you can see from the link what they do and how they work so ill not repeat that.

i didnt read all the posts so can you confirm for me this is an idle problem ?

Can you post the adaption figures you have .. which must be a few days driveing since you cleared any fault codes as this will reset the adaption figures.

Atm it sounds like 1 bank is rich and the other is weak .. thats into the lambda sensor / maf sensor fault area or exhaust fault .. blocked cat etc , air leak near lambda sensor .. that sort of thing.

If we are talking idle then can you see the throttle pedal values and throttle position values ?

If so post them and also watch them as you depress the throttle slowly .. tell me at what figure the revs increase and also with the air con on if you can maintain 1500 rpm .. im hopeing you can as if you have basically 1k or 2k with a/c on and cant maintain inbetween with engine hot then its similar to a fault i had recently .. that was a DME .. but that did show random throttle valve fault codes.

Ill look at the data and check my fuel sheets at work ( different ones for different models and i havent memorised them ) i can also chat with the guys and see if we can think of anything ..

Please bear in mind .. its a hell of a lot easier to fault find when i can work on the car than here but ill do my best.

Post here or PM .. either is good.

I can't save data on my code reader. I've read a couple of other posts/threads that suggest a faulty knock sensor can play havoc with the tickover.

What I can't understand is I take them off, have 20 miles of perfect driving (with identical readings off both banks sensors and stable timing reading) and then it starts to deteriorate back to where it was.

I could really do with plugging a duramteric in, but I'm not local to any specialists - nearest is Colne at 16 miles away + the fact I'm broke at the moment.

With your code reader can you see the above values and write them down to post or is it just basic fault codes ?

Knock sensors are possible .. its just that its a pretty rare thing and for 2 to go at the same time .. only times ive ever changed them is due to rodent damage .

Im thinking cost atm .. replaceing something without knowing if its the fault kinda goes against what i do.

trouble is you need to scope them to prove and even then the signal pattern looks pretty much like junk .. its just an ac wave pattern.

Im interested in the adaption values .. they tend to tell us a fair bit about long term running .. depending on how far out they are gives a clue as to what might be wrong.

Also the trouble with intermitant faults is they are just that .. always a bitch to find .. you have to gather every clue you can as the the fault is never there when you test it.

I was thinking of tank venting as well .. car has to be hot for that to start .. but really any fault on that would affect both banks.

Have you replaced the MAF sensor in the last few years ? i wont go into details but that is a possible.


Btw intermitant faults have a habbit of not comeing back if you have touched something ..always for me at least its never related when i finally find whats wrong .. so .. touching something is no gauarentee that that is the problem im afraid.

Im not sure if my link said how knock sensors work but they are basically a crystal thats gives off a small electric charge when under pressure .. ( same as your lighter if you smoke ) .. so a pressure wave ( a knock ) would generate a voltage that the dme can respond to and alter the timing accordingly , as such i would expect them to work or not work .. hence im not sure thats the fault .. and hence i dont want you spending money without more tests .. if you follow me