Wierd Pulse From Megsaquirt

I've got a question. This is in relation to coil firing and Dwell signal settings while configuring my Megasquirt to run my mid-engined 6 cylinder Triumph Spitfire (Spit6) project.

Conditions:

	* MSnS-E running Megatune 2.25b627, with 026a.
	* Setup for crank fired ignition 36-1 wheel, being simulated on the bench top with the Stimulator.
	* GM Coil pack with the three VB921's mounted in the bottom of the coil pack where the stock electronics would have been. The 'drive' from the ms box is via the current instruction... 330ohm base resistor feeding he VB921s from the tops of the resistors per the instructions in the current MSnS-E docs.

The docs all say...5mS cranking, around 3mS firing, and .1mS minimum discharge times

(sorry for the photo quality.... the pics were taken with
my new camera phone about 20 minutes ago.)

benchtop1.jpg (90799 bytes)
benchtop setup

To the left, the MS box, top middle, an electronic tach being fed from pin 37, currently reading just shy of 6KRPM. On the right, the modified GM triple coil pack (6 cyl setup) Under the bench is a 12V battery. The MS is being powered from a bench top power supply.

I found a 'break-out' board on EBay. That is what's attached to the end of the MS box. I added a short ribbon cable to feed the stimulator. This allows me to tap in drives like the injectors, the tachometer, the coil drives and etc while still feeding the MS box, all the sensor signals from the Stimulator

scopetrace1.jpg (64998 bytes)
traces on the scope

OK..to the strange part. The top trace is the signal across a single 0.01 ohm resistor, wired into the common ground lead feeding the coil pack box (with VB921s) from the 12V car battery under the bench. As viewed, you're looking at Coil A, Coil B, Coil C, then Coil A again. The bottom trace (in sync with the high spike, coil B signal) is the signal directly from the MS box via the break-out board tap that feeds the base of coil B's VB921 transistor.

coilpack1.jpg (60062 bytes)
sparks in the dark

Although dark, (on purpose) show the three spark outputs at 6krpm coming from the coils. I've simply got a small clip on one post, and the handle is near the other wasted spark post. The gap is about 050 to 060 in this picture. (simulating two sparkplug gaps each)

So... why document all this? Well a question comes up here. If I set the 'Running Dwell' to 3mS the bottom plate of the coil pack, which is the heat sink for the three VB921s, gets hotter than hell! So I added the 0.01 ohm resistor in, and started playing with the dwell settings. Right now, when these pics were taken, the 'running dwell', at 6krpm, is set to roughly 1.7mS. This is where coil B's (highest spike) levels off.

QUESTION???
OK...WHY...would one coil's current signal (coil B) spike so high compared to the other two flattoping at a much lower level?
   1) Could it be that the other two coils are saturating
        that much earlier than B's coil?
   2) Could the drive level going to the VB921s from the
        MS box be current limited, and limiting full turn-on
        off the VB921's?
    3) Other ideas?

If I 'up' the Running Dwell time to around 2-2.5 mS I start getting very small oscillations on the trailing edge of the signals upper 'flattop' trace. It doesn't degrade the signal, but it looks like some other digital signal bleeding through. (clear as mud right?)

Any ideas would be GREATLY appreciated.

...UPDATE 11/2005....

After a lot of head scratching, I finally figured out what was going on here. Even though the VB921s were all firing and the sparks from the three coils were all even and steady, coil A and Coil B's VB921 were apparently statically damaged. An ohm meter reading across the bases to the emitters, showed A and C's VB921 only having about 980 ohms. B's VB921 was reading around 190k ohms. So I ohmed out the three spare 921's I had here, and found they too all had a base to emitter resistance of around 190K ohms. So after replacing the two in my box, all 3 trace pulse all look the same now.

The threasd on this is at http://www.msefi.com/viewtopic.php?t=13174&highlight=

©1987-2008 All material copyright© Teglerizer 1996-2008	last edited 09/14/08	hits since last reset

A Very Strange Setup and Pulsing Problem