Yet some tested and found 10K not to be sufficient. Raspberry Pi themselves say 8K2 or less in E9.With 1M resistor will result a quite slow slew rate (on any other digital input), that renders the signal quite long time passing the undefined region after pin release, besides less noise immunity. And it won't read 0 immediately, only after a hold-off time. I wouldn't call it perfectly sufficient. Far more usual are external 10K pull-ups (prevalent)/ pull-downs (which work even in this case).
It may be that higher than 8K2 can be used if a hold-off time is implemented. But E9 makes no reference to hold-off times.
It was not intended to be ad hominem. I was only referring to advice and cases where 'there can be no defect, only user error' defences which have been mounted.After all, I will try to ignore ad hominem arguments, let's not go that way, I rest my case
It appeared to me that you do not accept there is a defect at all and would thus fall into that category. My comment was directed at the position you appear to hold.
Statistics: Posted by hippy — Mon Sep 16, 2024 2:10 am