Misfire problems
Discussion
The pics below show the plugs after a 50 mile run. The reason for posting this, is that I have a misfire which I can't pinpoint.
If you look closely you will see that plug 6 is really black and plug 7 is slightly black. The rest look as though they haven't been doing anything at all.
Plug 7 stopped firing when the car was in for an MOT and caused high HC readings. Substituted an old lead and it seemed to start firing again. (Passed the MOT, phew) Since then refitted the original lead and it's firing. The car is a TVR 390 with Megasquirt fuel and ignition(wasted spark, using Ford coil packs). Since going to this ignition system there has been a misfire somewhere, usually at higher RPM.
I have yet to find out the resistance of plug leads so that I can figure out if they're OK or not.
New Plug leads were fitted when converting to wasted spark
Plug gaps around 0.9mm
If you look closely you will see that plug 6 is really black and plug 7 is slightly black. The rest look as though they haven't been doing anything at all.
Plug 7 stopped firing when the car was in for an MOT and caused high HC readings. Substituted an old lead and it seemed to start firing again. (Passed the MOT, phew) Since then refitted the original lead and it's firing. The car is a TVR 390 with Megasquirt fuel and ignition(wasted spark, using Ford coil packs). Since going to this ignition system there has been a misfire somewhere, usually at higher RPM.
I have yet to find out the resistance of plug leads so that I can figure out if they're OK or not.
New Plug leads were fitted when converting to wasted spark
Plug gaps around 0.9mm
Edited by deetes on Saturday 9th August 14:41
Have you tried swapping the coil packs over to see if you can move the misfire?
Are you sure this is spark related and not an injector letting you down?
Have you had someone look over your MS tables to make sure there are no quirks in them? Less likely this one as it is unlikely to produce a single misfire but if it is producing a condition where a misfire is likely then a cylinder with a slightly different plug gap or a lazy injector would be the first to suffer.
Steve
Are you sure this is spark related and not an injector letting you down?
Have you had someone look over your MS tables to make sure there are no quirks in them? Less likely this one as it is unlikely to produce a single misfire but if it is producing a condition where a misfire is likely then a cylinder with a slightly different plug gap or a lazy injector would be the first to suffer.
Steve
Funny how one bank is running richer than the other - are you sure your'e not running dual tables or dual lambda's correction, which would run each bank differently.
On your datalogs, compare the pulsewidth (PW) for each bank.
And what megasquirt board version and code version.
On your datalogs, compare the pulsewidth (PW) for each bank.
And what megasquirt board version and code version.
Edited by eliot on Sunday 10th August 08:29
Steve
Haven't tried moving the coil packs, but I will.
Sent my my datalog files to the ExtraEFI man as he supplied the MS. He remarked that it looked realy rich all over.
I had been on the dyno to try and resolve all the issues, but it was way down on power from the previous year.
Elliot,
MS1-Extra and V3 pcb.
As far as I'm aware it's not running dual tables and it's definitely only got one lambda at the moment, unfortunately only a narrow band as I haven't got round to fitting my wide band.
I spoke to a local tuner who thought it could be something to do with the trigger angle and I've been trying to check timing etc against what MS is reading, but my timing light has broken. There was a lot of discussion on whether you could use a timing light, with a dial back facility, as some people reckoned it wouldn't work on wasted spark.
Haven't tried moving the coil packs, but I will.
Sent my my datalog files to the ExtraEFI man as he supplied the MS. He remarked that it looked realy rich all over.
I had been on the dyno to try and resolve all the issues, but it was way down on power from the previous year.
Elliot,
MS1-Extra and V3 pcb.
As far as I'm aware it's not running dual tables and it's definitely only got one lambda at the moment, unfortunately only a narrow band as I haven't got round to fitting my wide band.
I spoke to a local tuner who thought it could be something to do with the trigger angle and I've been trying to check timing etc against what MS is reading, but my timing light has broken. There was a lot of discussion on whether you could use a timing light, with a dial back facility, as some people reckoned it wouldn't work on wasted spark.
If you have any chance to fit a wideband sensor, I'd advise you to do that as soon as possible. I know they say it's possible to map it on the narrowband crossing events, and no doubt if you're an expert and everything is perfect before you start it probably is possible. But having worked through mine with the wideband I'd hate to have to do this job without it.
What plugs are you running? If it's running rich it will probably be a lot cooler than normal, and you might find the plugs don't get hot enough to stay clean. That can be a bit self-perpetuating since the dirty plugs will be more prone to a misfire which makes then even colder - and then gives the ECU a false 'lean' signal so it runs even richer and colder. You might consider going a grade hotter while you're getting the initial mapping sorted out.
What plugs are you running? If it's running rich it will probably be a lot cooler than normal, and you might find the plugs don't get hot enough to stay clean. That can be a bit self-perpetuating since the dirty plugs will be more prone to a misfire which makes then even colder - and then gives the ECU a false 'lean' signal so it runs even richer and colder. You might consider going a grade hotter while you're getting the initial mapping sorted out.
There was a very long thread on here about dialback timing lights. I think the conclusion we came to, is that if you set the dial back to zero, then what you see on the damper is the actual angle - if you are using a timing tape for example. However, if you use the dialback feature - the figures double. I.e 20' on the light is 10' actual.
Remember to zip up your log file, plus it would be handy if you post your MSQ also. Or email it to me and i will post it.
Still dont see how any form of ignition problem would cause an entire bank to run richer - the only thing common to an entire bank is the injectors - as they are paired together on the same output chanel.
Remember to zip up your log file, plus it would be handy if you post your MSQ also. Or email it to me and i will post it.
Still dont see how any form of ignition problem would cause an entire bank to run richer - the only thing common to an entire bank is the injectors - as they are paired together on the same output chanel.
Edited by eliot on Monday 11th August 08:22
Sent the files to Eliot as I couldn't figure out how to post them on here.
Took the plugs out again, after a run today, to see what they were like. Number 6 very black, the rest normal. I've tried changing coil packs and swapping leads. I moved that plug to another position so we'll see what happens now. Plugs are NGK BR7ES with 0.9mm gap
Took the plugs out again, after a run today, to see what they were like. Number 6 very black, the rest normal. I've tried changing coil packs and swapping leads. I moved that plug to another position so we'll see what happens now. Plugs are NGK BR7ES with 0.9mm gap
deetes said:
Sent the files to Eliot as I couldn't figure out how to post them on here.
Took the plugs out again, after a run today, to see what they were like. Number 6 very black, the rest normal. I've tried changing coil packs and swapping leads. I moved that plug to another position so we'll see what happens now. Plugs are NGK BR7ES with 0.9mm gap
If its just one plug that is black, my injector driver theory doesn't hold water.Took the plugs out again, after a run today, to see what they were like. Number 6 very black, the rest normal. I've tried changing coil packs and swapping leads. I moved that plug to another position so we'll see what happens now. Plugs are NGK BR7ES with 0.9mm gap
You could try moving the particular injector to another pot though, as you could have a "dribbly" one.
Also is the wiring the original stuff? - not uncommon to get corosion in the connector.
GreenV8S said:
eliot said:
Also is the wiring the original stuff? - not uncommon to get corosion in the connector.
Yes, I've had a couple go. Could also be a worn valve stem seal fouling that plug? CR check on that and a few other cyls..
New plug for no.6 and swap the lead to another cyl .. check the injector plug .. perhaps a contineuoty test..
I thought that with the Injection not being sequential that one injector couldn't affect one cylinder in particular, rather it would affect all the cylinders equally???
I've moved the plug to another cylinder to see if that changes anything.
The compressions were all checked the day after the disaster on the dyno. All were normal, can't remember the readings, and all were close to each other.
The engine ticks over quite nicely, but you can hear and feel somethings not right when you rev it. It just sounds rough at higher rpm.
I've moved the plug to another cylinder to see if that changes anything.
The compressions were all checked the day after the disaster on the dyno. All were normal, can't remember the readings, and all were close to each other.
The engine ticks over quite nicely, but you can hear and feel somethings not right when you rev it. It just sounds rough at higher rpm.
deetes said:
I thought that with the Injection not being sequential that one injector couldn't affect one cylinder in particular, rather it would affect all the cylinders equally???
Not really - if it was short circuit it would kill an entire bank (well actually the 5A injector fuse will blow). Open circuit, well it wouldn't inject. Easiest way to eliminate is to move it.Took it for a run this morning and checked the plugs again. Number 6 black again. So it's something on that cylinder. I checked the resistance of all the plug leads and they're all within the manufacturers specs.
Looks like I'll have to recheck the compressions again.
Regarding a duff injector, could I just pull the electrical connector off it?
Looks like I'll have to recheck the compressions again.
Regarding a duff injector, could I just pull the electrical connector off it?
deetes said:
Took it for a run this morning and checked the plugs again. Number 6 black again. So it's something on that cylinder. I checked the resistance of all the plug leads and they're all within the manufacturers specs.
Looks like I'll have to recheck the compressions again.
Regarding a duff injector, could I just pull the electrical connector off it?
If that injector is not sealing it would foul the plug. You could try moving the injector to another cylinder.Looks like I'll have to recheck the compressions again.
Regarding a duff injector, could I just pull the electrical connector off it?
Steve
You seem to have a lot of variation in plug colour. That could be inlet manifold mixture distribution problems or variations in injector output. It might at least pay to get the injectors cleaned and flow checked before you go back to a rolling road. Not much you can do about the inlet manifold whatever that is on your vehicle. The weakest plugs are the ones that look about right to me. FI engines don't generate anything like as much plug colour as old carbed ones, especially on unleaded fuel. All the even numbered bank looks rich.
The first thing I'd do to isolate the misfire problem is bring the plug gaps right down. Back in the days of mechanical dizzies engines managed quite happily on gaps of about 0.6mm. That's a big enough gap to produce a spark capable of firing 99% of fuel/air mixtures. As electronic ignition became standard on most cars gaps started to increase to obtain a bigger spark but you have to realise that a bigger spark only helps if you're trying to ignite a very lean, very rich, poorly mixed or otherwise hard to ignite mixture. The risk with a bigger gap and higher voltage spark is that the plug won't fire every time and each time it doesn't the voltage tries to go to earth somewhere else gradually breaking down the plug leads. A smaller gap hurts nothing until it gets so small the spark can't fire the mixture.
On highly tuned engines, and in particular turbocharged ones, a big gap can cause a failed spark when the cylinders are heavily filled or at high rpm. I had a customer who'd spent a year and god knows how much time on rolling roads trying to track down a misfire at high boost levels on his turbo engine which turned out to be nothing more complex than the plug gaps needing to come down by 0.2mm. The gaps were correct for the stock normally aspirated engine and its ignition system but the addition of a turbo and a bar of boost was too much for them. No one had thought to bring those factors into the equation.
So bring those gaps down to 0.6-0.7mm and you'll remove a lot of load on the coils and ignition leads. You can raise them again later 0.1mm at a time once you've solved any other problems.
On the very rich cylinder you need to do a compression test, check the tappet clearances and ideally put a dial gauge on the rockers to check cam lift to eliminate the possibility of worn lobes.
Finally make sure the ignition leads are clean and well separated from each other. If necessary make up something better than the OE lead clips. If you get any "crosstalk" between the leads, which is not uncommon on a V8, it can cause all sorts of misfire problems.
Again, dredging through the memory banks I recall a case of a Rover V8 that misfired like a pig but only on right hand bends. The owner had assumed, quite reasonably, that it was a fuel surge issue and had spent ages trying to track it down. It was actually G forces moving one of the ignition leads until it touched another one. So also look at the free lengths of ignition lead and where those might swing to when the car's in motion. Add some more clips to shorten the free lengths.
Dave Baker
Puma Race Engines
The first thing I'd do to isolate the misfire problem is bring the plug gaps right down. Back in the days of mechanical dizzies engines managed quite happily on gaps of about 0.6mm. That's a big enough gap to produce a spark capable of firing 99% of fuel/air mixtures. As electronic ignition became standard on most cars gaps started to increase to obtain a bigger spark but you have to realise that a bigger spark only helps if you're trying to ignite a very lean, very rich, poorly mixed or otherwise hard to ignite mixture. The risk with a bigger gap and higher voltage spark is that the plug won't fire every time and each time it doesn't the voltage tries to go to earth somewhere else gradually breaking down the plug leads. A smaller gap hurts nothing until it gets so small the spark can't fire the mixture.
On highly tuned engines, and in particular turbocharged ones, a big gap can cause a failed spark when the cylinders are heavily filled or at high rpm. I had a customer who'd spent a year and god knows how much time on rolling roads trying to track down a misfire at high boost levels on his turbo engine which turned out to be nothing more complex than the plug gaps needing to come down by 0.2mm. The gaps were correct for the stock normally aspirated engine and its ignition system but the addition of a turbo and a bar of boost was too much for them. No one had thought to bring those factors into the equation.
So bring those gaps down to 0.6-0.7mm and you'll remove a lot of load on the coils and ignition leads. You can raise them again later 0.1mm at a time once you've solved any other problems.
On the very rich cylinder you need to do a compression test, check the tappet clearances and ideally put a dial gauge on the rockers to check cam lift to eliminate the possibility of worn lobes.
Finally make sure the ignition leads are clean and well separated from each other. If necessary make up something better than the OE lead clips. If you get any "crosstalk" between the leads, which is not uncommon on a V8, it can cause all sorts of misfire problems.
Again, dredging through the memory banks I recall a case of a Rover V8 that misfired like a pig but only on right hand bends. The owner had assumed, quite reasonably, that it was a fuel surge issue and had spent ages trying to track it down. It was actually G forces moving one of the ignition leads until it touched another one. So also look at the free lengths of ignition lead and where those might swing to when the car's in motion. Add some more clips to shorten the free lengths.
Dave Baker
Puma Race Engines
Edited by Daveuk9xx on Wednesday 13th August 06:23
Dave
Nice to see you on here.
Those sound like excellent suggestions.
Also - got to say your website is one of those that I have really loved reading - I've learnt a lot from it :
http://www.pumaracing.co.uk/
Look forward to that book - how's it coming on ?
Nice to see you on here.
Those sound like excellent suggestions.
Also - got to say your website is one of those that I have really loved reading - I've learnt a lot from it :
http://www.pumaracing.co.uk/
Look forward to that book - how's it coming on ?
Gassing Station | Engines & Drivetrain | Top of Page | What's New | My Stuff