The Val

[VL400]

We have the non programmable MSD 6AL-2 (the new digital bullshit one), its using the original coil output from the camira module to the wire labelled "FROM POINTS OR ELECTRONIC IGNITION AMPLIFIER(ORIGINAL COIL - WIRE)" in the MSD instructions. So dwell should still be controlled via the ECU and should be the correct polarity - I hope! The instructions show a diagram for how to wire it up on a delco managed chev using the 7 pin HEI module so have gone by that. Not sure if we can run just the dizzy as we need the reference pulse for fuel, maybe the tacho out of the MSD could drive the ECU reference? But worth a shot, would have fixed timing but might help narrow it down.


The multispark stops at 3k RPM and you just get one spark. Your meant to get 20 degrees of crank rotation worth of multi sparks up to 3k RPM, the scope pics look close to 20 degrees worth but was not something I measured actually. Whats wrong with the scope pics frequency? They were taken at different RPMs so wont have the same widths or frequency, only the second and third pics were close in RPM as I just moved the scope lead to a different point.

[vn5000]

Using the first shot as an example,youve got a 15ms wavelength.1000/15=66.Therefore you should have a frequency of 66Hz.
Also have you checked it does stop multisparking at 3000rpm

[VL400]

Oh yeah that freq counter on the display was screwed up due to it measuring the reluctor freq - counts the positive and negative pulse as the cycle, we just went with the tacho and TP for RPM.

[vn5000]

Have you got any scope screen shots running at 6000 - 7000 rpm.

[vn5000]

An easy way to test if the msd unit is causing the problems would be to get a good bosch coil and justwire it to the camira ign module.

[VL400]

None captured as the crank trigger was a little dangerous at those RPMs so only ran it up a couple of times and sat in the car in case the trigger went flying off. It looked identical on the scope to the low RPM just didnt have the multi sparks - TP RPM trace was nice and flat, tacho was stable and shift light was glowing. Spark sounded fine and as it went past 7k the MSD rev limit kicked in and could hear it drop spark.

EDIT: With no MSD it runs fine but falls over under full load from just over 4k RPM - made right on 400RWHP at that point.

[Lawso]

If you pickup the trigger for the MSD before the module, it should be ok. Like you said, it will run at fixed timing, but hey, it's a drag car, a lot do. It's either idling or flat out.
The only other thing to consider is start retard. MSD do a module for that.

I would think that MSD would not have changed the trigger edge for the new digital box, as it would need to be retrofitted in place af a 6AL.

You will have no control over dwell using a MSD 6, they are a 'smart' module. They need a constant duty cycle to trigger correctly (ie 30/70, 50/50, etc). If you are trying to control dwell, the duty cycle will vary accourding to the required coil charge time. This change usually happens on the rising edge, which is why the MSD can get confused.
Your scope should be able to show this as a pulsewidth variation across rpm on the output of the delco. Maybe compare the pulsewidth of the ref output from the camira module to the trigger output from the delco. I don't think that they will be the same.

Haltech have some pretty good explainations of the different types of modules and the triggers that they require. Try taking a look at one of the older model manuals (e6X) as they do go into some detail. The manuals can be downloaded from the website.
Regardless of the brand of ECU, the logic is the same. You can however change the ignition output with a Haltech, and you can't with a delco.
I geuss it may be possible to make a converter box to change the ignition trigger from falling to rising by inverting the signal?
You guys are the smart ones when it comes to electronics! I did look into it once and I believe that it just requires a transistor circuit and a timer?

Try putting a timing light on the engine when it is running and see if the timing is stable across all RPM. It will most likeky not retard when you want it to either.

[VL400]

Hmmmm, wil certainly be checking all that out further.

My understanding of the MSD is dwell is not really used as its CDI so haltech and the likes just set it to 50% so there is a definite square wave trigger. The spark output from the camira module is falling for dwell to start and rising to fire the plug which is correct according to the 6AL instructions - the ref out from the camira module to ECU is opposite so would be out by heaps and would vary but were not using that to trigger the MSD as we wanted ECU spark advance control. There is certainly dwell control in the ECU and ref pulse width and EST out are different widths and happening at different times, although I cant see much advance happening in the scope caps so will need to check that one out.

Timing is pretty stable free revving - commanded the ECU to fixed 20 deg and revved it up, couldnt see it change much on the balancer. When letting the ECU do its thing TP and the timing light agree with each other - although idle was hard to get a stable reading as I was using proportional spark at that time.

[Lawso]

I think you have it backwards, but correct me if I'm wrong.

The waveform for the ignition module output that you have on the previous page shows that the coil is triggered on the falling edge as the curcuit is switching high to low.

For the coil to charge, voltage must be passed through it. The module closes the circuit ( High) then releases (Low) to fire the coil. The amount of time the module is high would determine the dwell. So the rising edge must vary on a falling edge trigger as it is the start of the dwell period.

If you are relying on the rising edge (which the MSD 6 does) on a system that controls dwell, your trigger edge will not be consistant as it will be moving according to the amount to dwell that the ecu is commanding. This will lead to some pretty unstable stuff happening with the timing.

Inverting the waveform so that whilst the coil is charging, the MSD will see 0v, will allow it to trigger on the rising edge, which will bring it into line with the module falling. This should take the variable of the module's rising point out of the equation.

It works on a points system connected to the white 'points' wire on the MSD as the dwell does not change, it is set by the points gap which gives the MSD a fixed duty cycle.
The points system will prove a similar waveform to the module, but the gap between rising and falling will remain constant.

I would like to know how the ecu determines dwell. Is it purley an RPM requirement or does it take load into account?

[VL400]

I am thinking the other way

The yellow trace is the white wire of the MSD so is the old coil - connection and the blue is the ref out of the camira module going to the ECU. Yellow trace should be normally high, then gets pulled low to charge up the coil and goes high again to fire the plug...

Dwell.jpg (27.33 KiB) Viewed 4244 times

But yes the dwell time will change unlike a points sytem. The ECU has a few different modes of dwell, it uses a static and dynamic calc. During AE it chucks in up to 3.96mS of dynamic dwell and then decays out 1/8th of it on each ref pulse after AE fnishes. The static dwell time is RPM dependant so will vary through the rev range. There is also battery voltage compensation, less volts more dwell time.