Oscilloscope Pictures From Aermotor/Webster

Magneto Setup

In December 2003, I got the 'ol Tek scope out and had some fun.  There's been a lot of argument about just what happens when a Webster plugoscillator magneto does it's stuff so I figured I'd try to learn something about it. 

What I wanted to find out was:

  1. How high the voltage spike is when the ignitor points open.

  2. What the electrical oscillation frequency is.

  3. What the mechanical oscillation frequency is and how many times the ignitor strikes after tripping.


When the caption says that "tripped off engine", it means that I turned the engine by hand and the magneto tripped off the trip rod.



When the caption says "tripped off mag. bar", it means that I had the engine trip disabled and used the Webster manual trip lever to fire the magneto.



Since this magneto is freshly rebushed, it is not as 'limber' as it would be had it been run for several hours.  Although this mechanical tightness may affect the mechanical resonant frequency, the electrical properties will be the same as those of a well 'broken-in' mag.

Here are the pictures along with descriptions of what you are seeing:

This is the setup I used.  The high impedance oscilloscope probe was connected to either the ignitor terminal or the unconnected magneto wire depending on the test being done (check the captions) and was grounded to the engine frame.  The scope used was a Tektronix Model 468 100 megahertz digital storage scope.


Here is the voltage on the magneto wire with the ignitor disconnected.  

The mag is tripped off the regular trip bar of the engine.

(The scope is in normal storage mode, sine response and the zero voltage line is at the vertical center)


The sensitivity is 5V per division (vertical) and 5ms per division (horizontal) showing the peak open circuit voltage to be approximately 18V positive.  The time for one oscillation of the magneto after it is tripped is about 12.5 milliseconds, making the mechanical oscillation frequency around 80 cycles per second.


Just to see the difference, the mag is still disconnected from the ignitor but the magneto is tripped with the Webster manual tripping bar.

(ZeroVolts is vertically centered)


Sensitivity is 10V per division (horizontal) and 5ms (5/1000th of a second) per division (vertical).  The maximum open circuit positive voltage is about 29V and, I think because the mag is pre-loaded more with the Webster bar, the mechanical resonant frequency is a little higher, about 100 cycles per second.


All the remaining pictures except as noted are with the ignitor connected and tripped off the regular trip rod of the engine in the normal fashion.

The zero voltage line for this picture is the next-to bottom marker.


This picture shows that there are several sparks generated within one fiftieth of a second after the magneto trips.  The scope is set for 50V per vertical division and 10ms per horizontal division and the zero voltage line is one division from the bottom of the screen.  The total amount of elapsed time between the left side of the screen and the right side is 1/10th of a second.  The second tall spike shown is over 400 Volts!  Now you can see why you get 'bit' when you're touching the ignitor lead when it trips.


This picture better shows that the voltage spikes go both ways as the magneto oscillates mechanically. 

 The polarity of the pulses is random probably because of the interaction between the magneto's electrical oscillation and the mechanical bounce of the ignitor points.

50V/vertical division, 5ms/horizontal division (1/20th of a second), with zero Volts at the vertical center.


Here's another picture with zero vertically centered showing the bi-polarity of the magneto output.

50V/5ms (Normal mode, sine response)

You can see that the voltage spikes go off scale for both polarities.


50V/5ms (5/1000th of a second) zero voltage is at the bottom.


As you can see, the voltage spikes that occur when the ignitor points open are pretty steep, over 400 Volts (the negative spikes are cut-off by the scope).  This is caused when the energy generated by the magneto is stored while the magneto armature is turning and the ignitor points are closed.  As soon as the ignitor points open, the magneto field rapidly collapses and makes a short high voltage pulse.  This is how the spark is generated.



The reason there is more than one spike (and, presumably more than one spark) generated could be due to one of two things.  One reason is that when the ignitor points open, there is a high frequency damped oscillation generated by the inductance of the magneto and the stray capacitance in the circuit.  The voltage rises until the spark jumps across the ignitor points as they open.  A spark has a very low resistance so when the spark jumps, the magneto is effectively shorted out for a very short time until the spark disappears.  After the spark is gone, the voltage again builds up from remaining energy stored in the magneto.  Once the voltage rises enough, another spark is generated and so on.  The other reason (not my choice) is that the ignitor points bounce shut and open during the first few thousandths of a second and this causes the multiple spikes.  You decide.


Speeding up the horizontal sweep of the scope we see that smaller spikes are buried in the larger spikes and are starting to appear as we speed up the sweep of the scope.  This gives credence to my damped oscillation theory.



Here, the entire width of the screen is only 1 millisecond (1/1000th of a second).  The 'spikes' are really short!



The scope is DC coupled in all of these pictures so, differentiation couldn't be the cause of the relatively gradual rise in the baseline voltage.  This phenomenon-could- be caused by localized heating of the ignitor points during an ignition sequence which micro-carbonizes oil on the points and makes the short-circuit resistance slightly higher.  At least, that's MY theory.  I could be wrong!  You can see this effect in the previous scope picture, too.  What's your theory?


Here, the sweep is only 0.5 milliseconds (1/2000th of a second), and we still can't measure the width of the spikes.



Now, we're cranked-up the sweep pretty good to a full-screen 100 microseconds (1/10,000th of a second).  

We can sorta see some width of the spikes, although they're still too skinny to measure.



Now, we're starting to get there.  

With the entire horizontal sweep occuring in 1/100,000th of a second, something is starting to resolve itself out of the spike.



Now, we're getting somewhere!  

This picture shows the electrical resonant frequency of the magneto at the instant the ignitor points open.

50V/0.1us (1/1,000,000th of a second from left to right).


Here is explained why we get "ignition noise" on the radio.  What we see is a very short radio signal that is generated whenever the ignitor fires..  The interference signal or 'click' you hear on the radio every time this engine fired would be strongest if you tuned a short wave receiver to somewhere around ten Megahertz.  Although it would be strongest around 10 megahertz, it is a "dirty" signal with lots of harmonics, and can be heard all over the bands. 



Now, don't get what you see wrong.  We're now seeing the ELECTRICAL resonance which occurs each time a spark occurs.  The MECHANICAL resonance is many thousand times slower!



Well - what do you think?  Have I got it right?  If you have any additions, questions or comments, please email me at:


[email protected]