PHD2 - Impact of Algorithms

[Stub Mandrel]

Look at this for an awful log, you can see what appears to be very obvious periodic error, and then DEC starts swinging uncontrollably from side to side. Changing the DEC max duration controlled this, but the program auto-increased it and the problem came back. This was with  Lowpass 2 on DEC and Lowpass on RA:

I then changed to Lowpass for both DEC and RA. The wild bit is due to the star being lost in cloud. The periodic error is still in evidence. The spikes are probably me stomping around:

After my early tweak and without the star loss episode, the RMS is about 0.7". So my advice for PHD2 users, is to try different algorithms and see how you get on!

My next task is to download thr latest version and try the PEC learning algorithm to see if I can flatten out that graph a bit more.

[Anthonyexmouth]

is the learning one the predictive pec? ive only used the RA hysteresis and resist DEC

[Stub Mandrel]

1 hour ago, Anthonyexmouth said:

is the learning one the predictive pec? ive only used the RA hysteresis and resist DEC

Yes, I've tried hysteris and resist before but Lowpass has given me the best results so far.