I assume the TRV calibration measures two values: 100% (where the actuator first hits the rod) and 0% (where the rod is at its maximum position, closing the valve).
When you calibrate the TRV while it is not connected, the calibration 'fails'. In what conditions does it fail? Only when it cannot determine the '100%' (open) value, or does it also make sure it could reliably determine the '0%' (closed) value?
Some forum threads around the radiator remaining warm even in 'closed' state recommend adding a 'spacer' helped for them - apparently the TRV physically couldn't completely close the valve. This suggests the calibration does not make sure it could reliably determine the '0%' (closed) value and just assumed 'all the way open'?
I think it would make sense for the calibration to fail when it can't determine the 0% (closed) value, signaling to the user they might need to add a spacer.
Alternatively, is there a way to get the calibration values through the API? In that case you could check if 0% was calibrated to 'all the way open' if your radiator remains warm and you suspect a spacer might be needed.
I see the logs (from /debug/log) do show some information:
Currentposition before move: 1 direction: open
Currentposition after move: 1690
Currentposition before move: 1690 direction: close
Currentposition after move: 5
...Total steps: 1645
Currentposition before move: 1 direction: open
Currentposition after move: 1666
Calibration passed
Last percent on boot: 0.00
Current position: 1666 - new:1645
Alles anzeigen
I'm not sure how to read this, though - does this look like a successful calibration detecting both 'sides'?