Strange fault - backfires and odd timing light results.

david jenkins · Post by **david jenkins** » Tue Oct 23, 2007 3:35 pm

Just been out to take my car for its very first MOT (UK annual inspection after 3 years - is it really 3 years since I got on the road?) and it's started to behave strangely.

Story goes back to a few days ago - started the car from cold, and got a backfire through the carb before it fired up properly... didn't think anything of it and the car drove OK.

Today, when I stopped the car after an enthusiastic run (warming the car up ready for the MOT, officer!) the engine ran on, coughed, and fumes came out of the air filter after it stopped. This has now happened a couple of times.

My first thought was that the megajolt had packed up and I was running in limp-home mode - fixed 10 degrees BTDC. So I got the timing light out... and got some very odd results.

Put the sensor clip on plug lead 1 and the timing's where it should be, at 12 degrees. Revved the engine, the ignition advanced - and the timing light stopped flashing after about 2000 rpm. Tried the other plug leads and found that nos 2 and 4 worked as expected with the strobe flashing all through the rev range. Nos 1 and 3 will only go up to 1500 - 2000 rpm before the strobe blanks out.

My last test was to disconnect the MJ box - and I got sensible results on all leads at all rpm, albeit at a fixed advance.

The engine's running OK, if a little flat & unresponsive when pushed.

I'm really puzzled... can anyone suggest what may be going on? Or are my test methods suspect?

Background info:

Ford Kent engine (crossflow), 1660cc.
VR sensor firmly mounted at the front crank pulley.
Latest version of MJ firmware on a V3 board.

Post by **brentp** » Tue Oct 23, 2007 3:58 pm

David,

What does a datalogging session show? Any RPM spikes?

david jenkins · Post by **david jenkins** » Tue Oct 23, 2007 4:02 pm

Haven't got that far yet - I'm still trying to work things out.

DJ

david jenkins · Post by **david jenkins** » Tue Oct 23, 2007 10:24 pm

I'm actually starting to think that the strobe light issue is a red herring - the engine still revs freely to its limit, and pulls hard all the way, so I don't really think that I'm running out of spark on 2 cylinders on the way! I'm thinking that the strobe light's trigger is getting confused, as the ignition leads are very close together for a large part of their route - maybe it's getting false triggers from the lead next door (it's the type with 2 power leads and a sensor that you clip over the number one lead normally).

I think I need to look elsewhere for my problem...

aarc240 · Post by **aarc240** » Wed Oct 24, 2007 12:31 am

David,

You didn't happen to tidy things up for the MOT?, like securing the HT leads differently?
Just wondered because of the comment about the leads running close together and the timing light issue - any chance that you inadvertently introduced crossfire?

Art C

david jenkins · Post by **david jenkins** » Wed Oct 24, 2007 9:44 am

I'm starting to think the same way - not crossfire between the leads, but false triggering due to the close vicinity of other leads next to the one the strobe's clip was on.

As I said earlier, the engine is running reasonably well so it's very unlikely that I'm losing all sparks on 2 cylinders above 2K rpm!

I'll have to keep investigating...

MartinM · Post by **MartinM** » Thu Oct 25, 2007 8:02 am

It doesn't seem to explain everything, but remember that each cylinder/coil lead gets twice as many sparks as a mechanical ignition, so the timing light should be trying to flash twice as often as normal - it's illuminating the timing mark on the pulley every revolution rather than every other revolution in a conventional system.

Maybe that's why it stops working completely at a relatively low actual rpm - because to it, it's actually twice that rpm and beyond its capability.

Maybe....

DannyP · Post by **DannyP** » Sat Oct 27, 2007 3:27 pm

Martin, I understand what you are saying about double the rpm on the timing light, but surely a timing light could handle 6000 rpm, which would be 3000 in a single spark, non-waste spark setup.

I may be totally wrong, but even in a wasted spark system, the coil fires every 360 degrees for the two cylinders, it would fire at 180 for the other two cylinders. Four strokes: intake, compression, ignition, exhaust. Wasted spark fires between intake and exhaust cycles. So the timing light hooked up to the #1 wire will only display the timing at the looked for value, not 180 out.

MartinM · Post by **MartinM** » Sun Oct 28, 2007 7:26 pm

DannyP wrote:Martin, I understand what you are saying about double the rpm on the timing light, but surely a timing light could handle 6000 rpm, which would be 3000 in a single spark, non-waste spark setup.

A 6000rpm max (conventional) timing light would surely only work to 3000rpm in a wasted spark system?

I may be totally wrong, but even in a wasted spark system, the coil fires every 360 degrees for the two cylinders, it would fire at 180 for the other two cylinders. Four strokes: intake, compression, ignition, exhaust. Wasted spark fires between intake and exhaust cycles. So the timing light hooked up to the #1 wire will only display the timing at the looked for value, not 180 out.

Conventional ignition = 1 spark every 2 revolutions
Wasted spark ignition = 1 spark every 1 revolution
(see the wacky animated graphic at http://en.wikipedia.org/wiki/Four_stroke)

So you're completely correct and I've amended my post

david jenkins · Post by **david jenkins** » Sun Oct 28, 2007 7:32 pm

Interesting theories, but remember my original post - On 2 leads the strobe stopped flashing at around 2000 rpm, but on the other 2 it continued to flash through the whole rev range. My later post pointed out that despite this odd symptom, the car seemed to be running OK.

I haven't had time to do any more tests, but I have changed the EDIS unit as I had a spare one handy. Once I have both time and decent weather I'll push the car out onto the drive and try again! If I can find my ancient Heathkit oscilloscope I might even have a prod around and see what the VR, PIP & SAW signals look like (desperate measures!

).

cheers,
David

Post by **brentp** » Sun Oct 28, 2007 7:41 pm

David,

Do try to get the configuration software attached and capture some data while observing the problem. If you see RPM spikes these are attributed to false PIP signals- the extra SAW responses would likely confuse the EDIS module, resulting in misfires.

I too have seen timing lights 'blank out' at certain RPMs- but for me this happened in a particular RPM range. In some cases I was able to solve it by repositioning the pick-up along the HV line.

Brent

david jenkins · Post by **david jenkins** » Sun Oct 28, 2007 10:35 pm

Brent,

I will be creating some logs once the weather improves - it's no fun driving my car in the wet!

I still think that the strobe light problem is a red herring - as said several times - the engine is mostly running OK, and definitely isn't behaving as though 2 cylinders are getting no spark above 2000 rpm! In fact, the car is pulling hard right up the rev range, so I'm trying to get back to the basic problem - running on when the ignition is turned off (occasionally).

cheers,
David

david jenkins · Post by **david jenkins** » Mon Oct 29, 2007 11:55 am

OK - just did some more testing... was only able to connect up the config software and run the engine, but couldn't take it out on the road (but probably not necessary anyway).

First of all, to reinforce what I said earlier: when using the timing light, I get very strong and continuous flashing throughout the rev range, entirely as it should be, when clipped to leads 2 or 4 - in fact, with the double spark it's like shining a torch!

When connected to 1 or 3 the light is flickering badly even at low revs, and disappears when revved up - it just looks wrong, even before I point it at the timing mark and rev the engine. I changed the EDIS unit but it made no difference. I also swapped the ignition leads for 1 & 2 (at both ends!) and the results were exactly the same. So the EDIS and the leads are not the cause. I've also had all the plugs out for checking (all normal colour and condition) and they went back in random order, so it wasn't the plugs affecting the spark quality.

I had the config software running during these tests and made a log file - this looks OK to me, with no sudden changes in rpm values or similar anomalies.

Now I'm stuck...

aarc240 · Post by **aarc240** » Mon Oct 29, 2007 11:39 pm

Looks like the only thing you haven't tried is the coilpack which is about the only common thing to cylinder pair issues.
If you don't have a spare see if you can rewire the primaries to switch the coil halves to the other cylinder pair (even with temporary patch leads).
Do remember to switch the HT leads too! (I've forgotten that before now).

david jenkins · Post by **david jenkins** » Tue Oct 30, 2007 12:00 pm

aarc240 wrote:Looks like the only thing you haven't tried is the coilpack which is about the only common thing to cylinder pair issues.

That's what I'm starting to think... even though the pairs of good and faulty are strange! If 1 & 4 were bad and 2 & 3 were good (or vice-versa) then the chances of 1 side of the coil being faulty are quite high, or maybe the drive from the EDIS unit.

However, I have 1 & 3 bad, and 2 & 4 good! I can't see how the MJ unit could cause that problem consistently (it would be a clever trick). As I've swapped plugs, leads and the EDIS, there's only the coil left!

Ah well - a journey to the scrapyard tomorrow...

David