R1 should probably be a higher resistance #2
Description
The transistors are being used in a very-not-recommended way (I did say the design was 'spicy'). Someone pointed out that the avalanche current will probably degrade them with time, and this could be somewhat mitigated by increasing the value of R1, probably to 100k or more. I'll test it before updating the schematics.
The current design was stress-tested for about a year, at which point I found no evidence of any problem. Still, I can't think of any good reason not to try and mitigate the issue -- it's not much work for me to just swap R1 with 100k or whatever and make sure the signal is still good. I'll find time to do so eventually, and then I'll close this issue.