Mks5 however, I'm still having the Dec trace jump off the graph.
Did it (1) suddenly jump, or did it (2) slowly drift away linearly (straight lie slope), or did it (3) drift rapidly away (faster than linear drift)?
If it is the (2), that is caused by polar aligmnet error (i.e., the typical declination drift that you use to polar align using the drift method). It should not be allowed to drift more than 7.5 arc second per second of time, if you are using 0.5x sidereal rate guiding -- since you will not be able to keep up correcting for it. This problem is also possible if tracking was turned off.
If it is (3), it is due to some bug -- i.e., the correction pulses are issued with the wrong sign! If that occurs only after an ASIAIR auto Meridian flip, there is a bug that ZWO claims to have fixed in v2.1 Beta for the Onstep and Celestron mounts).
If it is (1), it could be almost anything. I assume it had worked earlier in v1.9 with no cable drag, or belt slippage?
I have the other max Dec and max RA both still at 2000.
The max declination and RA step durations are there to limit to amount of correction in case the centroid measuremets has some large error.
If you had set it for 2000 milliseconds, and you are guiding at 0.5x sidereal, it means that each correction pulse can move the mount by a whopping 15 arc seconds! That is way more than a good mount needs. The typical problem with the max pulse duration is set to such a high number is that it causes wild sudden changes in the guiding when seeing is poorer.
Here is the Dec randomly jumping off the graph:
If that red (declination) graph then keeps running off the chart, the decimation pulse correction has the wrong sign. Bug somewhere.
If the red curve settles back down, reduce the declination max pulse. As I mentioned earlier, your max pulse setting is allowing huge corrections when they are not needed.
See if you can adjust the gain of the guide camera to get more guide stars too, if that sudden declination move is caused by a centroid estimation error. Try using the guide camera dark frame subtration to see if it helps give you more guide stars.
Since ASIAIR's multi-star guiding is weighted by the signal to noise ratio of the stars, you can try to remove that very bright star since it is predominating the centroid estimation -- i.e., you are pretty much using just one star guiding when one of the stars has that much better SNR than the rest. See if you can increase to gain of the camera to the point that bright star saturates and thus get excluded from the set of multiple stars. Notice that your primary guide star (with the green square) is showing just peaking 58 ADU and except for that one bright star, the rest could be even dimmer and noisier.
Chen
