So I spent most of the day doing basic maintenance, adjusted the valves, bled the brakes, checked a few things.. I'm going to Pocono for a DE event next weekend (april 6-7). In the meantime, I also installed an American International S-25A power line noise filter inline with the 12V signal coming into my relay box (power everything: fuel pump, megasquirt, wideband, etc).

Took it for a ride, felt really good except i was experiencing tach spikes down to 0 and up to 9000rpm. The tach signal would cut out. I checked the data logs, there are no voltage spike issues, just tach. When the spike occurs, I can see the tach actually spike. So I came back, removed the noise filter, went for another drive.. Same problem!! Tach signal drops to zero for a split second. I checked all the connections, made sure the coil is tight, checked the pertonix (gapped properly).


I have no idea what's wrong... At this point i'm pretty frustrated and just want to resolve this pesky issue and have fun next weekend. I might have to take a day off work to get this resolved.

Any suggestions as to what I should look for? Any parts I should order just in case?

When adjusting the valves, I did pop the dizzy cover off. I checked the rotor and all that. It;s not a spark distribution problem, it's a tach pickup problem.

For you megasquirt gurus out there, please help! I attached my log files and .msq file. Thanks!!

Info on my system: Stock frozen dizzy, petronix, vb921 injector driver, low Z injectors (PWM). MS II V. 3.0.

plot of datalog. after removing the noise filter, it appears i only experience tach spikes to zero, not up. This is really frustrating..

I also logged this several times. RPM sticks at one particular value for exactly 1/2 second, then drops to zero for the next data point. Something is funky..

Time SecL RPM MAP TP vBatt
2044.122 95 4256 98 63 14.3
2044.283 95 4250 80 25 14.5
2044.444 95 1993 34 3 14.4
2044.616 95 1993 20 0 14.4
2044.702 95 1993 18 0 14.3
2044.787 95 1993 17 0 14.3
2044.829 95 1993 17 0 14.3
2044.88 95 1993 17 0 14.3
2044.93 95 1993 18 2 14.3
2045.015 95 0 20 6 14.3
2045.167 95 3047 41 18 14.2

Its been a long time since I had problems with my tach signal, but I remember going to a shielded and grounded tach signal wire from the coil to the relay board.

I was having problems with the tach signal dropping out, but I was driving it with more voltage than you are. However it still helped clean the signal up.

I used a shielded wire with a twisted pair. One lead was used for the tach signal from coil, the other was unused. I then tied the shield to the ground at the back of the engine block.

Good luck and have fun at the track,

When I first built the system I shielded the pickup wire. This issue just crept up today. I must have bumped something. I HOPE I make it to the track.

I did some thinking and may have narrowed the hunt down to a few areas.

Here's my thinking:

MS gets the tach signal for the Pertronix (Hall Sensor).
Coil is powered from relay box, along with wideband O2
MS controls spark by sending signal to (-) coil.
Dashboard tach picks up from (-) coil.


So, my wideband O2 doesn't appear to be resetting. Meaning that the coil isn't losing power. Since the dash tach also dies when the ignition cuts out, means the coil isn't triggered since there is no tach signal.

Last night I rerouted some wires, upgraded to latest MS code, regapped the pertronix pickup, testing the wiring, replaced a few crimps.. I also found that my tach pickup was NOT shielded. Initially I shielded the tach pickup when I was relying on the dizzy to trigger the coil. Then I moved over to have MS control spark, and moved the shielded wire to the (-) coil. So, I connected the shield on my tach pickup.

I'll give it a shot tonight and see what happens. Hope for the best!

EDIT: Has anyone performed this mod? http://www.megamanual.com/ms2/inputHEI.htm

I'm not sure if it will help unless you use an HEI or other ignition module. Any thoughts?

Put some more into this headache...

Changed the coil and pertonix.

NO difference.. something is fishy.

Tach signal drops to 0 for a split second, the back to normal. Runs very rough at high rpm, everything was perfect 2 weeks ago. No other signals spike, batt voltage is fine. Really scratching my head on this one.. at this point there is no way in hell i'm going to pocono this weekend. Really bites the big one!!

If anyone has any other suggestions, please let me know.

Just posting an update in case anyone has a similar issue in the future and runs a a search.

There is a known issue that a lot of people experience related to the hall sensor input. See http://www.megamanual.com/ms2/inputHEI.htm.

Last night I completed mod #2, I can't tell if it worked or not because I can't get the car started. It runs for 2-3 seconds and dies. I posted logs and all the details on msefi.com. http://www.msefi.com/viewtopic.php?p=193381

Essentially things went from bad to worse. For now i'll keep my mouth shut regarding the whole concept of MS being experimental and a true engineer-it-yourself solution (Well... it is).

I'll post updates, hope i can get this sorted out by this weekend. I'd hate to lose a $450 non-refundable Drivers Ed deposit due to this stupid issue.

Quick thought, remove the tack lead and see if this clears up the problem. The
tack uses battery (12 volts), this may cause proplems with the optical pick up
which uses 5 volts.

The tach is isolated from the circuit. It picks up off the (-) coil. MS gets its signal directly from the Pertronix (hall sensor).

I'm going to take tomorrow off of work and essentially start from scratch. Clean slate, see what happens. My fuel figures might not work with the new ignition config. My logs indicate Pulse Widths might run too rich and kill the engine. Will keep everyone posted...

Since I resolved this issue about a month or two ago, here are the solutions to series of problems:

1) After rewiring the hall input circuitry (mod #2), the engine would no longer run for more than 2-3 seconds. Turns out I wired the Opto IN to +12V instead of +5V, ooops. Rewired and tach signal issue cleared up, no more rpms dropouts / spikes. 904SVO hit the nail on the head.

2) Went for a test drive, car ran like crap over 3k rpm, hesitated reving, felt like the ignition was cutting out. Started playing around with fuel, spark settings, etc etc.. all wrong. Turns out the plugs were fouled from running too rich for a year or two. The strangest part is that the fouled plugs just so happened to cause ignition problems the very same weekend my ignition signal started developing issues. Found this problem using the old timing light trick. Clip the pickup to each spark plug, check for consistent flashing at idle and while driving.

3) While double checking everything I could possibly think of, I decided to measure fuel injector output. As suspected, one bank produced 12% less fuel when collected in a 100ml container triggering one injector at a time using injector test mode. At high pulse widths, they balance out. But at low pulse widths, 5ms, there is rather large different in the injected fuel. (I documented all the results if anyone is interested) This is left over from the original issues I had when I first built my MS board and had flux on one of the injector output transistors. I guess there is still an imbalance in the circuitry. Solved this problem by moving all the injectors to one bank and running in simultaneous mode. Finally my Wideband O2s are stable thanks to consistent fueling and fresh plugs

Learned my lesson... again.. ALWAYS go back to the basics. MS II is a complex system, more often than not the problems i've experienced are as simple as can be. After all of this I missed a drivers ed event, money down the drain... but at least I figured everything out and the car is running better than before. In addition, my head temps are noticeably lower (previously could reach 400F, now under same load/scenario max at 350-375. PROBLEM SOLVED. THE END

Since this headache I got a few autoxs in, car feels really good (this time with confidence).. FINALLY!

Thanks for posting the solutions. Sounds like quite the brain teaser.

Happy Driving!!!