BTT Octopus v1.1BLtouch J43

[Jarewa]

Is it only me who has this problem of poor design? optoisolation LE357C. In tests it receives
probe accuracy results: maximum 0.046484, minimum -0.431641, range 0.478125, average -0.148594, median -0.144531, standard deviation 0.147996

And it's not a mechanical fault or a damaged sensor, but a poorly designed optocoupler. Because if I use a different pin (PB7 - PG12), the results are fantastic.

How to repair to obtain good optocoupler times?

[Jarewa]

I'll answer it myself, maybe it will be useful to someone. PB7 There isn't pullup, that's why it floats because the sensors are too small pullup BLtouch/CRtouch. The solution to the problem was simple, just make a jumper in J40 (set J38 5V) jumper DC-PB7. It causes pullup 3.3V R187 on PB7.

The results now are very satisfactory