Reverse engineering an electrical spark igniter

Article By : John Dunn

When the electrical spark igniter in the kitchen stove stopped working, this engineer dug in to see how it was designed.

The electrical spark igniter on our kitchen stove stopped working. I bought a new one, replaced the old one and everything went back to normal. However, I decided to look into that old igniter to see how it was designed. Here’s what I found.

igniter circuit boardigniter schematicFigure 1 Here is the kitchen stove igniter circuit board.

There was this one little circuit board whose schematic I traced out. I had to assume that the neon lamp was an NE-2 because there was no nomenclature. In any case, it sure looked like an NE-2.

There was a step-up transformer encased in a brown tinted, and quite impenetrable, but transparent epoxy.

igniter transformerFigure 2 The igniter transformer was encased in epoxy.

The transformer seemed to be all alone inside the epoxy so when I measured a DC resistance on the secondary winding as 540 ohms, I figured that there was a very large number of secondary turns, meaning a very high turns ratio indeed.

The basic operation seemed to be a relaxation oscillator delivering pulses to the NPN base. Each time the NPN would pulse the transformer primary, a high voltage spark would be generated from the transformer secondary. How the designers assured themselves that the NPN would always be kept within its safe operating area (SOA) limits, I don’t know because there were no transient protective measures taken vis- à -vis the transformer’s primary winding so far as I could find.

To confirm my understanding of this module, I made the following SPICE simulation.

igniter simulationFigure 3 The result of this simulation match the actual operation of the igniter.

Resistor R6 is shown where the transformer would go just as a convenience, something to let the model demonstrate that the NPN transistor was being switched.

My biggest simulation problem was modeling the neon lamp. That part was not included in the SPICE software. However, knowing that an NE-2 type bulb will “strike” or be turned on with an applied voltage of 90V and that it will extinguish or be turned off at pretty close to 60V, I made the model highlighted in yellow in the above sketch. I didn’t bother with simulating the on-resistance of the ionized gas because lamps like this must always be used with a ballast resistance, which would be very much higher than the electrode-to-electrode resistance of the lamp itself.

The approximately three-pulses-per-second of this simulation match quite well to the actual operation in my kitchen stove as I set about to cook up some scrambled eggs.

John Dunn is an electronics consultant, and a graduate of The Polytechnic Institute of Brooklyn (BSEE) and of New York University (MSEE).

Related articles:

Leave a comment