More Rb87 lamp measurements

I previously did some measurements to find out how supply voltage to the lamp affected stability but because I had taken it apart for repairs, I thought I would measure a bit more on it:


The temperature is measured on the internal baseplate of the lamp where the transistor is heat-sinked, but that temperature obviously has little to do with the amount of power spent on the lamp.

The lamp driver almost acts like a resitive load, but not quite, and the best correlation seems to be that light depends on power:


The straight line fit is 1.38 [V/W].

Those data were measured with the lamp already lit.

Another question entirely is how much voltage is required to light it in the first place.

This plot is made by turning the lamp supply off for five seconds and then applying the voltage plotted on the X-axis and measuring when A7TP2 reacts to the light:


Below 16.7 volts, the lamp fails to reignite in less than three minutes, and given the shape of the curve, there is little reason to think that it ever would even if we waited much longer.

A cold start would obviously not be as fast.


It seems that sufficient power to light the lamp sets the lower limit on lamp power, whereas the upper limit is probably component stress on the A12A1 assembly.