Lumpy and sluggish when cold

Hey, have had a search but cant fond the answer…

I have a 1990 1.8 8v MK1 Cabrio GTi
When starting from cold, the idle is a bit lumpy…when you push the throttle, it sounds like its missing.  Acceleration is also poor for about 5minutes until it has warmed up!

Any ideas what could be the problem?? Have changed the dizzy, rotor arm, leads, plugs and checked the timing. Have also checked for air leaked which i cant see and noticable ones.  Not familiar with these engines tho so need your help.

Cheers,

Andy

What's the mixture like through the warmup?

seven..

what lol??  please explain, not v techincal when it comes to cars

Cheers,

Andy

Mixture is how rich or lean the engine is running, ie air-fuel ratio is lower or higher than stoichiometric. Its typically measured as AFR or lambda, but CO is also measured because it gives a correlation when rich.

Measure it just after startup and keep measuring it all the way as the car warms up to normal temperature.

You're going to need a wideband lambda or gas analyser to measure it - but it will take you straight to the problem, if there is one here.

ahh ok thanks for the advice…so i just really need to run it from cold and findout whether its running rich or lean and how this varies as the engine temp increases.

Il try and get a hold of something to measure that or take it to a garage and get it done!

Cheers,

Andy

right…still not managed to check the CO2 but checked a few other things.

WUR - Resistance 22Ohms so OK
Wiring to WUR when cranking - 10V (too low??)

AAV - when cold, squeeze and revs drop, when warm no change presume thats OK.

5th injector wouldnt play a part i presume as it only operates for 10secs?

Only other thing i could think of is the temp sensor (located on pipe from head to rad with brown plug) Resistance when coolant around 30degrees is 65ohms….i dont no if this is right though??  When unplugged running gets worse!!

Any other ideas pleaseeeee…will try and get the CO2 measure ASAP

Thanks

Andy

CO2 is really hard to measure, it involves inflating an expandable gas chamber over a period of time, then analysing the contents…

CO is much easier to measure, and this is what you want to do.




But ideally, wideband lambda (any MoT place will have both, and it would be sensible to put both probes into the exhaust, so long as there's enough venturi effect to get a reliable idle reading on the lambda.

ok ye thats what i meant…il get that done ASAP.  Anyone know the reading for that sensor tho?? Just really too start eliminating things

Cheers,

Andy

The K-Jet fuel injection doesn't use any inputs from any coolant temp sensors (the KE-Jet does…) so any difference in running plugged or unplugged is  :dontknow:

hmm there is a DEFINATE difference.  With it plugged in, and warm, it revs fine, unplug it at same temp and it doesnt rev!!  Why do they have that sensor if its not used anyhow??

Cheers,

Andy

Its possible yours is a later model with KE jet, or a variation of it (well technically a variation of K Jet). Without thoroughly looking at the car I can't tell.

The same advice still applies though - get the mixture looked at.

AFAIK VW coolant sensors are NTC. Very few people on here run an EFI so unlikely to find a table with temperatures and resistances on here, but on other forums and a wider search, I bet you could find a little table supplying those values no problem.

Right, finally got my hands on a Gunstons test and tune number.  Calibrated it to read 2% in the air as per instructions then tested the car.

Initially - shot up to +5.3
After 1minute    -0.9
After 5min         -0.4
After 10min       -0.4
After 15min      +0.2
After 20min       +0.3

Thats when the engine was up to temp and just before fans cut in.  

Soooo…any ideas whats wrong then? Does it just need adjusted or does this show that somethings faulty?

Cheers,

Andy

Ok that's showing its lean when its up to temperature. You need to let the car cool down again (but warm up the gas analyser - leave it connected to your battery for at least half an hour) then repeat the tests and observe the mixture during warmup.

that not what iv done??  The analysier was warmed up, as per instructions for about half an hour, 40mins and calibrated then i took readings from when i started the engine (once calibrated) to when it was warm!

Andy

Aaaah right, rereading it, yes. So it looks like its lean all through warmup and into normal temperature - (although the gas analyser obviously lost its calibration, it can't be a -ve CO value unless its calibration is off - but this isn't your fault, they're just rubbish!) So I'd be tempted to recalibrate it, then adjust to approx 2.0% CO once its warm, and then try it all again (or simply evaluate how the car behaves when warming up).

Hi Andy, did you ever get this issue with poor starting when cold resolved. I have exactly the same problems you described with my 1989 cab GTI.