Unreliable shutdowns

[art0]

Hi again
One summer more, one more green pool :)

During the last weeks I am getting some erratic behavior. I get responses to the CMD_PROBE more often that the CHEMLINK commands and sometimes the unit totally stops responding. When there is no responses, it shutdowns and that is really a problem.

2020-07-15 12:57:48: -->
2020-07-15 12:57:51: <-- 1002 b0 11 64 37 1003
2020-07-15 12:57:52: --> 1002 00 16 0303220049374924 3d 1003
2020-07-15 12:57:52: <-- 1002 b0 00 c2 1003
2020-07-15 12:57:52: -->
2020-07-15 12:57:52: <-- 1002 b0 11 64 37 1003
2020-07-15 12:57:52: -->
2020-07-15 12:57:55: <-- 1002 b0 11 64 37 1003
2020-07-15 12:57:55: -->
2020-07-15 12:57:55: <-- 1002 b0 00 c2 1003
2020-07-15 12:57:55: -->
2020-07-15 12:57:55: <-- 1002 b0 11 64 37 1003
2020-07-15 12:57:55: --> 1002 00 16 0303220049374924 3d 1003
2020-07-15 12:57:58: <-- 1002 b0 11 64 37 1003
2020-07-15 12:57:58: -->
2020-07-15 12:57:58: <-- 1002 b0 00 c2 1003
2020-07-15 12:57:58: --> 1002 00 01 0000 13 1003
2020-07-15 12:57:58: <-- 1002 b0 11 64 37 1003
2020-07-15 12:57:58: -->
2020-07-15 12:58:01: <-- 1002 b0 11 64 37 1003

Have you experience the same problem? I am thinking, that maybe is heat or just far too many queries to the unit.
Do you know a way to revive the unit once it goes OFF?

It the communication stops, the gear shutdowns and eventually you get a green pool :)