missfire between 3800 and 5000 rpm?
Turbo Mini Forums

Hi all,

Finally got the MS2 box fitted in the mini, running ignition only. I've got a fairly conservative map for now until i feel more confident with mapping.

However, I have noticed now that the coilpack is being driven off MS2, there is a missfire high up in the rev range, which is seemingly random and has no pattern. It also did the same with the previous map which was very racey (not conservative at all). As i pushed it higher in the revs, the MS2 reset itself suddenly?

I have since adjusted the mixture to more near stoich on idle but I dont think this was the cause, seeing as the MS2 reset. I currently dont have a cap fitted to the supply feed on the coilpack, could this be the cause? I've also adjusted the maximum dwell down to 2.5ms from 3.1ms which has aided it a little bit, and adjusted the maximum spark time from 0.7ms to 0.6ms.

At a bit of a loss really, any ideas guys?

You probably need to adjust the VR pots on the MS mainboard.

Also make sure that you have a clean supply and earth to the MS unit.

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

OK, thanks Paul. That reminds me that I did wildly adjust both of them when I encountered another problem with the tacho out settings, but that's been solved now.

Your map that you gave me was fantastic by the way, but I want the engine to last a few miles more yet

Edit - Do you have any advice on how to adjust the pots? Like number of turns or anything similar?

Edited by Carl S on 16th Jun, 2010.

Glad the map was of use.

I've just been searching through old threads on here to find the recommended pot adjustment but can't find it.

Rod S knows. I think it's something like one of them fully anticlockwise and the other just a few clicks from anti-clockwise. I suggest you search the MS build manual.

The other potential problem could be noise on the VR sensor cable.

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

http://www.megamanual.com/ms2/vradjust.htm

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

Cheers Paul, just found it as well.

I'll have a fiddle when its back from MOT.

I don't have an actual answer for the best VR pot positions, only to say mine needs about 7 turns different setting on one of the pots on the engine compared to on the JimStim.

I'll get my notes out later to see which it is. I only set it up on the engine by trial and error, watching the RPM signal on the laptop to find when it became unstable.

In addition to Paul's suggestions, I would also check the VR sensor gap, not just at one point but several, to make sure the trigger wheel is concentric, and also your VR polarity - easy to get wrong.

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

A couple of additional points. You need to make sure you have resistor plugs and/or resistor wires.

Also, some VR sensors require an additional series resistor on the input. A 10k value is often used.

And to elaborate on Rod's point on the sensor polarity, you need to check the timing with a strobe light and if there is a timing difference between what you have in your MS and what you see on the engine that increases with RPM then this is a good indication that you have the wrong polarity. In that case, swap the wires or change the input capture direction.

Jean

http://www.jbperf.com/

It's sounding more like its the polarity thats the problem as I did have to set the BTDC degrees for the trigger wheel very low (its at 62 degrees as apposed to 90) to get the timing anywhere near correct.

You mention changing the input direction Jean, is that a software setting? I'd much rather do that then rip the cable apart again.

http://www.jbperf.com/

Excellent, thanks Jean, saves me a bit of work.

do you have the suppressor wired between the coils middle pin and ground?

turbo 16v k-series 11.9@118.9 :)

Denis O'Brien.

please fit it too. once you start loading up the engine you will get resets / missfires around peak torque. :) as found on any ford with a coil pack, not just the EDIS cars :) usually light grey or black with a bosch 2 pin connector, but only one wire..

turbo 16v k-series 11.9@118.9 :)

Denis O'Brien.

I'll pick up a 20uf cap from maplin tomorrow. I need to get some bits to sort out the tacho feed anyway.

My coil never came with a supressor for some reason.

Edited by Carl S on 16th Jun, 2010.

you ideally want it as close to the coil as possible and the ford item is nicely weatherproofed, just incase your passing a scrappy!

turbo 16v k-series 11.9@118.9 :)

Denis O'Brien.

There's only one scrappy near to me and they're rip off bastards so i'll just wrap the cap up well I think.

Well, had a bit of a play this morning. Couldn't even get it to run on 'falling edge' so i'm guessing the polarity is correct. I tried altering the BTDC degrees value to several different values(60, 90, 110 and a few inbetween) for the crank wheel but it would just either not even attempt to start or backfire out the carb. Using a timing light, it does appear to advance and retard properly so I think I can rule polarity out.

I now need to load TunerStudio on to the tuning laptop as it's current version is out of date, but it has no networking capabilities and so I have to transfer over stuff via usb flash drive. I also need to get MegalogViewer on there so I can see what the VR signal is doing and whether it is noisey. I think this is my chief suspect and so I will carry on looking at this in detail untill I can find some solid evidence that noise is the issue.

If noise is not the issue, I think it's a case of adjusting the dwell timings for the coilpack as that seemed to make a difference before, it did not however, prevent MS from shutting down when the missfires got out of control.

Here's a log I took a few minutes ago. I blipped the thottle quite high occasionaly, and held it high at one point. I will log rpm along side next time.

Attachments:

275.72kb

You've got a tooth in the wrong place. About 4 or 5 teeth before the missing tooth.

Turn on noise filtering.

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

I think I may know what that is as well, there is a tooth that has a very tiny step in it a few teeth before the missing one, but it's a horizontal step, not a vertical one, and it's about 2mm I think.

I will try the noise filtering Paul.

Just to add, should I be worried about the MS completely resetting itself? Or does this issue go hand in hand with resets?

You won't know for certain until you fix the signal issue, but I would say quite likely. I never had any issues with resets when I was doing my testing but that was on the garage floor rather than the more noisy environment of an engine bay (my alternator, for example, wasn't connected to anything).

I've done some experimening with the software noise filtering on the JimStim and it is very good - but if you have a physical problem on your wheel that can be directly attributed to what is happening, why not fix it at source ???

ie, if it is a damaged tooth, put a blob of weld on it and file it back to the correct shape ???

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

If I had the time and resources Rod (I'm busy all next week untill I leave for Worcester for 12 months and I don't have a welder), I would fix it, but noise filtering will have to aid me for now.

I will see about getting all the teeth milled a little deeper (horizontally) by my friend who machined it for me to begin with, so that all the teeth are even in width. This probably won't be for some time however.

Well, i've been out working on it most of the morning and have so far had very little success.

I have tried each method of noise filtering, including combinations like polarity and interupt mask, and the rogue tooth still appears on the tooth logger. I've also tried very harsh filtering but this just makes the car run like a bag of nails.

The only thing I have not tried so far is to turn on the regular noise filter, followed by adjusting the 'noise curve' map. I have tried just turning on the regular noise filter without adjusting the curve and it was no different.