francesco BUT I see only the RA graph and NOT the DEC graph.
The RA and DEC RMS numbers are the root mean square deviations from their mean values.
If the mean value is not at zero, even if the RMS error is small, the curve can still be outside the the visible graph area. I suspect that is why you don't see the declination graph.
When this happens, it usually indicates that the feedback loop does not have enough gain to compensate for some constant bias -- i.e., the feedback loop has barely enough gain to try and pull the error in to a constant value which is non zero.
In your case, since it is in the declination axis, the first thing to suspect is bad polar alignment. If you had used the new All Sky Polar alignment in ASIAIR, the large error from ASIAIR's algorithm can also be causing this type of error (see my posting on the All Sky Polar Alignment subject in this forum).
It is also possible that "third axis balance" (check Google for what that is, if you are not aware of it) has caused your balance to be off for a different part of the sky (many beginners only check their balance with the OTA parallel to the ground).
You can try to increase the aggressiveness (loop gain) of the declination channel to see if it can pull the declination error in. The problem with doing that is too much aggressiveness will cause oscillations, so it is better if you get to the root of the problem, instead of trying to let the autoguider compensate for something it is not designed to do.
FWIW, asking the autoguider to compensate for poor Polar Alignment is also asking the autoguider to perform a task it is not designed for.
Chen