Bias Voltage of Mercury Switchboxes

The following test was conducted by Mr. Bill Rogers regarding switch box bias voltage.

Bill Rogers, August 2005

The key piece of test equipment is an old HP tube-type oscillator that I use to simulate the trigger signal. It's the only signal generator I've found that can put out a high enough voltage to simulate a trigger.

I have two 332-7778A1 switchboxes on the bench with grounds and bias terminals connected together. I have 1 coil and plug connected to each switchbox. The HP is connected between switchboxes as a trigger input and fires each the coil/plugs once per cycle (or rev) 180 deg apart. The plug on one switchbox fires as the trigger goes positive and the plug on the other switchbox fires when the trigger goes negative. This represents 1 of the 3 trigger windings we find on our beloved Mercury V6 outboards, each firing 2 cylinders in opposite banks 180 deg apart.

For the stator voltage I'm using a variac with a 0-200 volt output and I'm running it through a full-wave rectifier to double the frequency to 120Hz. The variac voltage is set so that switchbox output is 150v peak to the coils (as measured with a Fluke multimeter and DVA).

The HP oscillator is set to produce a sine wave of about 35 volts peak and 100Hz. This is roughly what the trigger looks like at 6000 rpm. The real trigger is not a clean sine wave of-course (in fact a portion of the waveform is missing), but it has the same general shape near the zero crossings and that is where the action is.

If you look closely on the following scope screen captures, you will see some spikes riding on the waveform - this is when ignition occurs. NOTE: I added some red circles to the images to make them easier to see). I was planning to set up secondary probes on the plug wires but I didn't have to - the ignition events are easy to see on the trigger waveform (due to interference).


The above represents normal operation with no advance module connected. Bias voltage is about 25 volts dc. Notice 3 ignition spikes: 2 above zero and 1 below. In this case, ignition occurs approximately 1.1 ms after each zero crossing.


Here is what happens when I ground the white-black. Bias voltage is now zero. Two of the ignition spikes are hard to see because they line up close to vertical scale lines - and there is a fourth spike on the far right. Ignition now occurs just 0.2 ms after each zero crossing.
Please note: It's hard to tell by looking at the scope screen just exactly how far past zero the ignition spikes really are. I'm using the numbers from the scope's software running on a notebook PC.

So, with the white/black grounded, ignition happens 0.9 ms sooner. What is that in degrees? Each trigger cycle represents 1 revolution of the crank or 360 deg. At 6000 rpm, the cycle is 10 ms long. So if 10 ms is 360 deg, 0.9 ms is 0.9/10 * 360 or 32.4 degrees.

Grounding the white/black at 6000 RPM advances the timing well over 30 degrees!
I realize this is worst case, but that's what I wanted to know - just how much can varying the bias voltage change the timing. No wonder so many motors are fried from bad modules. If the module grounds the white/black, timing advances in a big way and kaboom!