Turbo Minivan - weird misfire from cold on cylinders 1 & 2.
Turbo Mini Forums

Hi guys.

Could really do with some advice about a misfire I’m suffering from on the van since rebuilding the engine over the winter.

I’m running Megasquirt MS1 direct fired wasted spark Ford coilpack, VR crank sensor and 36-1 toothed wheel. Up to when cylinder 4 popped on the RR last April, this had been running well since finishing the resto in 2008.



I rebuilt the engine over the winter with a lower CR (8.15:1) achieved using new cast 1293 Omegas which I have had machined to give 14cc bowls and I’ve fitted a NA MG Metro Cam which replaces the AP2, along with an RTS clutch. These are the only changes I’ve made. I’ve just got it all back together again and running.

The misfire I have is weird as it only occurs from a cold start and only on cylinders 1 & 2 and the plugs are noticeably wetter with fuel than 3 & 4 as shown in the shot below (Plug No 4 is far left, No 1 is far right).





No 2 plug, second from left is very wet but is also sooty showing that it’s trying to fire cleanly but is not being allowed to for some reason whereas 3 & 4 are perfectly dry and blackened from the use of a little choke to start.

This is a little weird as the coilpack fires 1&4 together and then 2&3 together so you’d expect any fault with cylinders 1 & 2 to also affect the other good cylinders which it isn’t. The only common theme with 1&2 is that it uses the same inlet tract and I suppose it could be a manifold gasket leak, but read on as the ignition is doing something weird. Thing is, a manifold leak would lean the mixture off in cylinders 1 & 2 rather than cause the obvious over wetting so I’m thinking this is an ignition issue rather than one of fuelling.

With help from Almichie, I’ve changed the coilpack (and run a new earth to the body), 20mF capacitor, put in new plugs (BPR7ES) gapped to 0.028in, tried new Mondeo leads (and swapped over 1 with 4 then 2 with 3), changed the crank sensor (today), all with no change.

A compression test from cold with a closed throttle shows the cylinder pressures within the 5% tolerance;

Cylinder 1 – 152 psi
Cylinder 2 – 152 psi
Cylinder 3 – 155 psi
Cylinder 4 – 160 psi.

The map value at idle is set to 11 degrees and this exactly matches the figure on the crank pulley with a strobe so there’s good correlation there.

However, with a strobe connected, the flash rate on cylinders 1 & 2 is noticeably slower than 3 & 4, and on cylinders 1 & 2 there appears to be a small gap in the flash indicating the misfire.

But why the hell is the strobe rate slower than on 3 & 4 when these are from different coils and by the nature of the coilpack, 1&4 are the same and 2&3 are the same ?

Could it be something to do with the dwell and spark durations in Megasquirt as I’m sure Rod S posted something up a few years back about the Ford coilpack being sensitive when directly driven by igniters in the ECU ?

This stopped me from getting to Bingley today and is driving me a bit mad.

Sorry for the length of the post, just wanted to get a few points over.

Any help much appreciated.

Cheers.

Richard.

Minivanless, but reluctantly happy living with the decision. There'll be another one day.
http://www.youtube.com/profile?user=minimadmotorman#p/u

hummm, weird,

try swapping the plugs round? 1&2 with 3&4?

shut the plug gaps up a bit?

Edited by turbominivanman on 30th Jan, 2011.

Minivanless, but reluctantly happy living with the decision. There'll be another one day.
http://www.youtube.com/profile?user=minimadmotorman#p/u

Edited by Rod S on 30th Jan, 2011.

Schrödinger's cat - so which one am I ???

Going off on a tangent....could this be more mechanical issues....sticking valves etc? Got a level compression readings, but was this done warm or cold?

Edited by Jason G on 30th Jan, 2011.

Minivanless, but reluctantly happy living with the decision. There'll be another one day.
http://www.youtube.com/profile?user=minimadmotorman#p/u

Can you rev through the miss or is it all the time? Does megasquirt have any trim functions related to coolant temp that could have altered. Seems a strange one. You say the earths are good these gave me some issues with my Megajolt.

Minivanless, but reluctantly happy living with the decision. There'll be another one day.
http://www.youtube.com/profile?user=minimadmotorman#p/u

The only other thing I can think of is your settings in MS aren't ideal for your new lower compression ratio, especially if it only happens when cold.

What settings do you have at the moment ?

Schrödinger's cat - so which one am I ???

Minivanless, but reluctantly happy living with the decision. There'll be another one day.
http://www.youtube.com/profile?user=minimadmotorman#p/u

Is the crank sensor close enough to the trigger and does the trigger have any run out?

are the settings for the 36-1 trigger set correctly, and is the dwell for the coil set for 3ms? with battery compensation turned on? usually 4.5ms

Rod/Sprocket.

In terms of crank sensor - I've set this up at 0.8 mm and checked at 4 points radially around the trigger wheel. Not really a sniff of a difference in the gap so runout tis pretty good but it's a bastard to get to.

Apart from a trigger angle of 56 degrees to cure a slightly misaligned sensor, I'm using a 5 degree crank advance angle and I'm using a 'fixed angle of -10' ie, following the map, with no trim angle.

These are my trigger settings:

Trig Pos A: 3
Trig return Pos A: 8
Trig Pos B: 21
Trig return Pos B: 26

These are my Dwell settings :

Cranking Dwell: 6.0 ms
Running Dwell: 3.6 ms
Min discharge period: 0.1 ms
Battery compensation: ON.

These settings have been with me since 2008. What are your opinions, are they ok ? Difficult to know what's optimal from the info on the MS forum.

Richard.

Minivanless, but reluctantly happy living with the decision. There'll be another one day.
http://www.youtube.com/profile?user=minimadmotorman#p/u

The settings are quoted slightly differently in MS2, ie, we don't have to set trigger and trigger return - but as the A and B are 18 apart, that shouldn't be the problem.

I do think the (running) dwell is a bit high though. I'm at the maximum recommended of 3.5, Paul dropped his to 2.8 (I think from memory) after misfire problems and possible coil damage and Sprocket suggests 3.0.

So although I'm closest, we are all less than you. So it's certainly worth dropping it to see what happens.

The selective changes affecting only 1 and 2 - if you swapped the leads at the coilpack end only (or the plug end only but not both at the same time) to reverse the current flow through the plugs and it made no difference, it does point to a mechanical/fuel issue such as you have already considered like an air leak. But it could be an air leak on 3 and 4 which you have masked with a slightly rich mixture so 3 and 4 run OK despite the leak and 1 and 2 foul because of the mixture.

Schrödinger's cat - so which one am I ???

Thanks Rod.

I'm away at present but will do as you anf Sprox suggest on Tuesday evening to reset the running dwell to 3.0 mS. I'll also check the coilpack resistance and if that doesn't sort it I'll probably start taking the manifolds off and check the gasket.

Interestingly, just thought about your comments regarding the mixture setting and for the first time I've had to screw the mixture screw right in to richen it enough to get a decent idle mixture. That's probably a significant lapse on my part to remember and point out.

Shit, was hoping to avoid taking the bloody manifolds off ! Will reset the dwell first.

Cheers guys. Will be back with results on Tuesday evening.

Much appreciated.

Richard.

Minivanless, but reluctantly happy living with the decision. There'll be another one day.
http://www.youtube.com/profile?user=minimadmotorman#p/u

Could you try your old dizzy? This would eliminate MS issues and tell you if your mixture etc is wrong.

If something is worth doing, it's worth doing half of.

Do try to keep up John, Richard told us earlier that he had sold the dizzy !

Me no ready very good sometimes.

If something is worth doing, it's worth doing half of.

Maybe so, but you must read bloody quick to have got back so fast !!

richard , although you say that a leak on the inlet to 1/2 would not make it rich ,a leak on 3/4 would make it rich on 1/2 if you adjusted the carb to be right for 3/4 .maybe thats why you have the screw in so far ?

Medusa + injection = too much torque for the dyno ..https://youtu.be/qg5o0_tJxYM

Minivanless, but reluctantly happy living with the decision. There'll be another one day.
http://www.youtube.com/profile?user=minimadmotorman#p/u

possibly the spark is running down the insulator which is wet from the rich mix ? so the kv doesnt build up to a high enough level to trigger the strobe ?

Medusa + injection = too much torque for the dyno ..https://youtu.be/qg5o0_tJxYM

Minivanless, but reluctantly happy living with the decision. There'll be another one day.
http://www.youtube.com/profile?user=minimadmotorman#p/u

Right guys.

Managed a sneaky 20 minutes on the van tonight and reduced the running dwell from 3.6 ms to 3.0 ms.

Didnt make an ounce of difference. The car started and ran with the misfire still present, as equally as bad as before.

Didnt check the HT coil resistance which I'll do tomorrow but before removing the manifold gasket, I think I'd like to see what would happen if I chuck a dizzy in.

Anyone local near Chippenham, Bath or Bristol got an A+ dizzy and a coil I could borrow ?

Richard.

Minivanless, but reluctantly happy living with the decision. There'll be another one day.
http://www.youtube.com/profile?user=minimadmotorman#p/u

Try spraying some carb cleaner around the manifold face to prove a leak in the gasket. My money is on fueling rather than ignition mate.

I reckon its fueling/air leak too.

If you get stuck I should have a A+ Dizzy lying around somewhere, but I'm not too local

Edited by dischub on 1st Feb, 2011.