One direction DEC after meridian flip

[Astroswell]

Hello everyone! I have a EQ6-r pro mount installed in remote observatory. I use GS Server and I have my Reverse Dec after flip activated. My guiding in DEC was horrible most of the time and I later I've discovered it was due to the enormous backlash in that axis. I could feel the mount head wiggling when I tried to move it. Unfortunately I had no option on trying to fix it mechanically, and backlash compensation didn't work for me either, so I decided to give a shot to guiding in one direction. My guiding improved dramatically after that! 

BUT only until the meridian flip. After it, the DEC would start running away and since the direction was set to NORTH, after the meridian flip the PHD would just stop reacting to the movement of the star. You can see it in the log called "before". After that I tried to switch the direction to SOUTH manually after the flip and it started guiding well again(the log called "after", though the seeing sucked that night). 

So my question is: does PHD provide the means of doing it automatically? Maybe via some adjustments in the switch resistance or something? 

Otherwise I'll have to try to implement that fix it in the code myself.

Logs are in the attachment. My resolution is ~1.8 arcsec/pixel so any guiding below 1" is fine.

[Bruce Waddington]

I've spent quite a bit of time looking at this problem, and it is considerably more complicated than you think.  The point of reversal in Dec drift is a function of the azimuth and altitude errors in polar alignment.  If you have a permanently installed system,  you can know ahead of time at what sky location this reversal point occurs.  This is not the general case, by any means, and deducing that sky location in real-time by watching the Dec activity is difficult and typically unreliable - especially if the mount exhibits other problems relating to its Dec axis.  Your system, for example, shows these kinds of behaviors (Dec in green):

This is from side-of-pier west where you need to use north-only guiding.  These are pretty clearly caused by something shifting around in the payload, probably the guiding assembly, but they exemplify the reality that setups with severe Dec backlash or stiction usually come with a basket load of other problems.

That said, it could make sense for you to force an automatic switch in Dec direction at the sky location where it is needed - since you are permanently set up and presumably don't change the polar alignment.  Rather than flailing around in the PHD2 source code, you would be better advised to write a simple app that uses the PHD2 server interface to accomplish the job.  It can monitor the hour angle position of the scope and send the appropriate command to PHD2 to reverse the Dec direction when your particular "trigger point" is reached.

Regards,

Bruce

[Astroswell]

Thanks for your help, I've decided to use my modified version of PHD2 for the time being, until I sort out the mechanics of the mount.

For those who are interested, here's the modification I've done.