P38 Gems rough idle slight misfire

This site contains affiliate links for which LandyZone may be compensated if you make a purchase.

spannersNK

New Member
Posts
5
Hi All, hoping somebody can shed some light on this it's driving me nuts!

I have a '96 4.6 p38 with BRC LPG and a Tornado chip. The liners went at 92,000 and so I rebuilt the engine with a Turner Engineering short engine with a H180 Cam, it started first turn of the key and ran like a sewing machine for about 8,000 miles. One day I noticed the idle was very slightly rough, causing a bit of vibration when sitting in traffic, now whether it came on suddenly or gradually I'm not 100% sure but I'm pretty good at noticing that sort of thing so I think it was most likely suddenly.

I've been trying to sort it for around 5000 miles now with no real success. It's barely noticeable and as I say more annoying than anything else as you really feel it when stationery, seats and steering wheel vibrate and because I know it was and should be super smooth it really irritates me!

It's got no worse or better over the 5000 miles and as soon as the revs go above about 800/850 you don't feel it but I can still just hear the miss through the exhaust. It's very much a regular miss/cough ever half second or so at idle and less frequent as the revs rise. Power and general running is otherwise perfect.

It's the same on petrol or LPG but you do notice it a bit more on LPG I guess because it's more sensitive than petrol.

I started with the basics, plugs and leads were a changed, plugs good colour bit worn, one Magnecor lead a bit cracked and they were 35,000 miles old so changed them as a precaution.

Compression 175-180 on all cylinders

No difference.

So far:

1. Went to a local LPG chap for some diagnostics, he tried a new MAF and suggested changing O2 sensors in case of intermittent fault, no change

2. Faultmate shows no error codes and I think all readings are within range as far as I can ascertain.(screenshot attached) Reset adaptive values 3 times after component changes.

3. Changed the idle air control motor as it was sticking in one direction.

4. Checked for vacuum leaks, thought I'd found slight one at the drivers rear end of the manifold, brake cleaner made idle rise very slightly and when checking the coils one rear manifold bolt was loose so rebuilt the top end with new valley gasket etc as a precaution.

Again no change, vacuum shows 17in, bit low on the gauge but is this a normal value? I have a H180 cam in it.

5. One coil pack showed very slight low resistance so changed it to rule it out.

6. checked crank sensor and reluctor ring, one tooth slightly bent, fixed it, no change. (am I right in thinking this either works or doesn't) Once it wouldn't start for about 20 mins after having the ignition on for a while while running fault mate and sometimes the idle will drop to around 450/500, can the CPS cause a rough idle?

7. Cleaned my MAF have tried a working secondhand one as well, car runs without MAF after a bit of revving but still same miss at idle without it so figure it can't be that and don't want to just fork out for a genuine one without good reason (cheepos are not advised from what I read). Airflow varies between 25 and 28 at idle (760rpm) bit above spec of 22 but voltage is 1.67 at idle rising to around 2.8 at 3000 so think this indicates it's functioning?

8. Have disconnected injectors one at a time and each results in a different constant miss,plus as is the same with LPG can't be that?

All of this has improved the mileage a bit but not the misfire.

Have pretty much run out of ideas now, don't think it can be any major internal problems as it so new and has no bad noises, plus it came on suddenly and has not got worse in 5000 miles with off idle running still 100%



Any ideas? Anybody know a Rover genius in the SE/Kent area as it's driving me mad!

Thanks in advance.

Will.

results.jpg
 
Last edited:
One for Saint, I think!

The battery voltage looks low but I'm not sure if that would affect the idle?

You've checked coils and plugs.
You've checked vacuum.
You've checked breathers ?

Once you've fiddled it can take a while for the trims to get back in line.

Has the GEMS engine got EGR?
 
One for Saint, I think!

The battery voltage looks low but I'm not sure if that would affect the idle?

You've checked coils and plugs.
You've checked vacuum.
You've checked breathers ?

Once you've fiddled it can take a while for the trims to get back in line.

Has the GEMS engine got EGR?
Battery voltage is normal for Faultmate.
 
Yep checked of those numerous times, and had a really close check of the drive plate when I checked the reluctor ring.

Took it for a spin this morning with the MAF and o2 sensors disconnected and did a reset, bit down on power as you might expect (assume it runs on default map with all disconnected?) and still the same miss at idle.

It's so even regardless of what I change I'm thinking it has to be timing related, could the crank sensor have gone out of range somehow but not log a fault?
 
I doubt it would run if the CPS wasn't working. The CPS started failing on my Mig when it was hot and it didn't muck about: 1 minute all was fine; the next moment all was dead and I was in a high-speed brick with massively heavy brakes and steering. Good job it was me and not the missus driving.
 
Is it possible any of the replacement parts were faulty?

If you're near Bucks I know the whereabouts of a 4.0 GEMS you can swap parts over to test with - as long as you promise to put them back afterwards!
 
I doubt it, as nothing changed anything so they would have to be faulty in exactly the same way, unlikely with 3 different MAF's, plus disconnecting the maf and O2's made no difference to the rough running.

Thanks for the kind offer but I'm in Kent so a bit far over, as eightinavee said I think it requires a scope to get to the bottom of what's happening with the ignition. It can't be fuel related as it's the same regardless of Petrol, LPG and with and without the O2's and MAF.

I've connected those transparent plugs to each bank and they all show even consistent sparks, hard to tell though being wasted spark but it feels as if the timing is just a bit out, is 15 degrees advance right for idle?
 
Yes slightly, but nowhere near the level that would give a lumpy idle, plus as it ran perfectly for 8000 miles with it fitted before the misfire started that can't be the problem. If it had always done it it might be different, it's not a lumpy cam idle anyway it's actually more like what you get with a worn cam at about 100,000.

But again as it came on suddenly and hasn't got any worse or better over 5000 miles it's not that either.

Just wish I could diagnose the cause!

What I need is a Rover guru in the South East with the equipment and magic ears to see exactly what it's doing at idle.

Anybody know of such a person?
 
Yes slightly, but nowhere near the level that would give a lumpy idle, plus as it ran perfectly for 8000 miles with it fitted before the misfire started that can't be the problem. If it had always done it it might be different, it's not a lumpy cam idle anyway it's actually more like what you get with a worn cam at about 100,000.

But again as it came on suddenly and hasn't got any worse or better over 5000 miles it's not that either.

Just wish I could diagnose the cause!

What I need is a Rover guru in the South East with the equipment and magic ears to see exactly what it's doing at idle.

Anybody know of such a person?
Yup, but we're all up north :D
 
Umm.. just looking at your faultmate printout...

timing advance 15 deg ? i'm Thor not Gems so dont know if that is where it should be ?? my 3.9 classic is at 6 deg +/- 1 deg

Also your Bank 1 and Bank 2 O2 sensors seem rather disparate..

Pre Cat O2 sensor - Bank 1 2.12 / Bank 2 6.22
Fuel Trim short term - Bank 1 7.5 / Bank 2 11.2

They should be 5v max, so if the O2 sensor is out of range that would affect the fuel trim on that bank too would it not ?

Maybe you have a faulty O2 sensor ? if they are recently changed (not on like limpets) you could swap them over and see if the readings shift
 
Last edited:
Just got to this thread, but I am along with Adam on this, the fuel trims and Lambda readings seem to far off to me!

Esp. considering he has reset fuel trims a few times.....

(Thanks for the nomination....touched - no puns Alan!! :D:D:D but I am far from expert!)
 
Is the H180 cam "hotter" than the standard? I found when fitting half race cam's in engines that the tickover was always a bit lumpier. A higher tickover speed was required:)

I was thinking this - my RR has possibly a non-standard cam in it (there's a mention in the history file) and has never idled so you can balance a 50p on the plenum but it really starts to dig in over about 2900 rpm.

How about a good old fashioned comp test, injector cleaner and a good hard caning?
 
A multimeter would be a good start, the Titania oxygen sensors are fed 5v from the Gems ecu. So quite how you get 6.25v from your diagnostic tool has me guessing. The short term trims are fine,for a multifuel engine with non standard cam, no matter how hard you try to trim the LPG ecu - you are unlikely to get it to emulate what GEMS would do.
If it were in my workshop I'd be attacking it with a proper diagnostic tool,(Testbook) closely followed by my scope.
I cant be bothered to read through all the posts again, sorry - very tired this evening, but has a compression test been done ?
 
Back
Top