If you now understand that there are two different ESD tests, one with the piezo grill ignitor and one with the new gun, what are you considering the "sparker test"?
Either will do, but I think you should keep doing with the grill igniter simply because it's become a de-facto yardstick. It's difficult to compare new/old results if we switch to something else.
Do those igniters wear out with time? It might be good to measure the sparks and make sure it's holding up, producing sparks just like the first day.
There's a fair difference between the two PCBs you show.
They've moved the buzzer and a few components around, yes.
When you really zapped the AN8002 the major trace damage happened around the area of the buzzer. Maybe they moved the buzzer away to improve that area of the PCB.
Regardless, the thing that failed first on that meter was the transistor clamp - nothing to do with PCB spacing or layout. If the transistors in the AN8008 are the same then I'd guess it will fail in the same way.
Of my sample of two grill starters, there is a WIDE variance!! And yes I believe the output changes with time. I have a starter that looks very similar to this one that is new and it puts out a much larger spark. The one I use for testing actually came from a old grill I scrapped. The spring tension is a fair amount less than the new one.
It only really gets used for the meters and I have checked the output a couple of times during the course of all this testing and it appeared to be the same.
I have changed the test methods before as I find better ways to run them.
When I "zapped" the AN8002, it was the higher energy generator that caused the damage you mention. I don't record any of this testing in the spreadsheet. It's more an FYI than anything.
Because the PTC and clamp are located pretty much in the worst possible location setting up a loop that runs more than half the length of the board, it's really hard to say how changing the layout will effect it. The part placement and layout are VERY important when you start pumping several amps through the board in sub ns.
You may feel the clamp would take the initial hit and I would tend to agree with that for the slower rise time transients. The grill starter transient is pretty much over in 10ns. The rise time on the other generator is around 1us. It matters!
I would really have no way of knowing what failed first on the 8002. You can guess it's the clamps but I can't say that for sure. Again, many thing will play into this including the layout.
I would say if I had been able to replace the clamps or even just remove them and get the meter functional again, you would be correct. That was not the case.
Where are you sending the $10 if the 8008 fails the gun? I think if it fails you should have to send Dave $20 to cover his cost of the 8008.