Originally Posted by Mark Z
But this is with an "OE zener diode".
Keep in mind that my test was of only one device so there no way to know from it alone how wide the tolerance in Zener Voltages was for the components Lucas supplied at the time. Although it would be easy enough to test a dozen original Lucas Zeners to determine the answer to this question about variation it would take a few hundred dollars to acquire them on eBay. Or, wait a minute, maybe my friend has a couple at his shop...

I dropped by my friend's shop on the way to work this morning and asked him if he had any old Zeners on the shelf. He thought he remembered having seen one somewhere in the back room. I left the shop a little while later carrying 9 of them so even if a few turn out to be bad there will be enough to get a reasonable idea of the variation.

For reference on what is possible 40+ years later, I went to the Digi-Key site and filtered only for voltage tolerance, leaving all other options open. They show 21,207 different Zener diodes are available from them having +/-5% tolerance, but that drops to only 1,680 if I specify +/-1%, and they have just 3 at +/-0.5%.

If Lucas's tolerance wasn't better than +/-5% (a range of 14.25-15.75 V for a nominal 15 V) it would mean some people will have batteries with reasonably long lives, but others will have ones that cook fairly rapidly. Based on what is available today, I suspect a tolerance of +/-1% (14.85-15.15 V) was beyond what Lucas achieved at the time.

To do a reasonable test I'll park my good Zener at 1.0 Amps and see how long it takes for the voltage to stabilize when internally heated by the 15 Watts. Then it should be a quick matter to repeat the measurements on the other diodes. I probably won't get to this until the weekend but will post the results as soon as I have them.

Although the most significant outcome of this will be to satisfy my own curiosity a side benefit will be to give me enough information to select a diode for my Trident having the "optimum" characteristics within the intrinsic limitations of a Zener. That is, should I ever go to the trouble of measuring the one currently in it and then switching the diodes if necessary...