I'm not a super guru when it comes to ECU electronics. Normally all go well but this time I'm facing a strange problem. During mapping the S2000 2,4 ltr I'm facing misfires. I see the dwell time jumping very high (even red color on gauge) and you can see the "shadow" of the needle standing very high for a few seconds. Or dwell goes to almost zero. After some investigation I saw on the real time display the voltage got some spikes. When the engine is not running, no spikes, nice flat line. Once running I got spikes up the line and spikes under the line. Changing the earth points did not help at all. + I'm already taking 12 Volt supply direct from the battery. It can't be a "lost" of current because I also have spikes up the line, not only under (adding voltage). Every time I go a spike under the line I got a misfire (coil not fully charged). Coil is VW / Audi with build in amplifier, wasted spark. Toothed wheel is 30-2 and very clean signal. Changed ECU with an other one, same problem. Nothing I've done changed anything.
Some good Megasquirt electronic guys out here? I feel, must have something to do with electric interference. At least, I guess.
Are you using Tuner Studio / Megalog view to tune? If so, when using Megalog to look at a log you will be able to see whether Megasquirt is resetting from red markers in the time ribbon. I would think the low voltage spikes are causing resets hence the misfires. As you are seeing these in Megasquirt you must be getting an intermittent power supply to it. Looking at your voltage it is generally under 13v, which seems low to me. Normal voltage should be 13.8 - 14.5 volts when charging. So your high spikes are could actually be ‘return to normal’. The evidence seems to be pointing in the direction of the charging circuit. Dying Alternator? Faulty diode pack? If not, maybe a dry joint on the power supply to the Megasquirt. Worth running a soldering iron over a few joints.
Theres is no alternator on the engine (is running on the dyno). Pretty sure, after half a day of fighting all joints are 100% true.
Did some surfing on the net I found out low impedance injectors can cause this problem. Seems like they send a lot of rubbish back into the ECU. More as you can expect from a coil. It's the first time I'm using low impedance injectors. The symptoms I've been reading about are exact the same as I got right here. Tomorrow I will have it run on the stim in situation and see what effect the injectors have on the voltage reading.
The spikes on top of the voltage must come from a source higher as the battery voltage. Since there is no alternator and I did shut down the battery loader, the spikes must come from an induction back into the circuit. At least I hope I got it right. The spikes on top are not a direct problem, only the one under the line (very often serious low) cause misfires as like you do a "burn". All gauges on screen do a strange and quick "move".
Some people make mods so the spikes disappear but I prefer the switch to high impedance (at least IF .... )
Will keep you informed, if not working will try post a log.
Thanks
PS, I did had a faulty diode once, believe me, it takes a lot of time before you realize what's going on.
Yes, setup for low imp. Will test this evening on stim, with and without injectors to see what's happening. I did not noticed any problem when I've build the wiring an tested on stim but did not checked for other errors, injectors where opening and ignition was working.
OK, connected the stim and had it running. Since the engine is not real running you can hear all the noises and it's clear every time you see a spike (up or down) you here the injectors are not correct working (and possible ignition, could not see). Disconnected the injectors and voltage line stay perfect flat. Changed injector setup for high impedance, fitted 12 Ohm injectors and voltage line stayed flat as without injectors. Injectors perfect running.
Have not started the engine yet, will do tomorrow, but I'm sure the problem is solved.
I'm no Megasquirt authority but I thought they were only capable of supporting 1 low ohm injector per output or 4 high ohm - injector current was / in capped at 5A - maybe thats it and its over loading the output transistors? Your swop seems to bear this out?
Wel, don't know all about the models but this one should be able to fire the 4 low ohm injectors. Only, these can generate the noise. Seems like some make mods to the MS2 and clean up the spikes. I hardly understand what they exact modify and does not want to start soldering. Just replace the low ohm and will never fit them again. I've always been using high ohm but had 2 sets low ohm at the shelves and wanted to use them (saving money but loosing an other day!!!!!).
My son had all sorts of problems getting low ohm injectors on a Suzuki Swift engine working with his MS1, the drive circuits just didn't like them despite all sorts of recommended fixes. So when I fitted injection / MS2 to my Pinto I made sure I went high impedance.
I ran my pinto fuel only and had no dramas, once i connected the coil packs it misfired. The coils dirtied up the 12v .the fix was a diode across each relay to stop them dropping and a capacitor/condensor mounted near the coil pack.
Reply With Quote
well done, and very useful info. 
Bookmarks