Dec backlash won,t clear

[PATRICK FORSTER]

Phd2.05a dec backlash won, t clear at all. Dec baclaxh clears just fine in phd1.14.2.
Pat Forster

[Andy Galasso]

Pat,

Are you using the same calibration step size in PHD2 as you were using in PHD1? PHD2 will not inherit any of your settings from PHD1, you have to enter them again in PHD2. Not sure if that is your problem, just a thought...

Andy

[PATRICK FORSTER]

Andy, thanks for your reply. Yes I checked the settings in PHD 1.14.2 and PHD2, same settings. I even increased the mounts RA and DEC auto guide rate to 99%,hoping for a bigger response. I even set the Calibrate  rate to 6000, 3000 seems to work on a 8x50 guide scope.  I use a 8x50 guide scope and a SSAG with main scope a Canon  400mm l lens. I have removed and re-installed couple of times the betq PHD. Still  the steps to clear backlash keep accumulating...43 steps and back lashnot cleared. I even bumped the mount NORTH button to clear baacklash and then calibrate, since PHD clears N first in backlash. As soon as I go back to PHD 1.14.2 it takes 2 steps to clear BL and N cal begins and finish with S cal. So this is the second night I have tried the beta PHD and BL clearing fails.One thing, about  5 days ago I used my OAG, SSAG,  on my AT 430mm F5.6 with PHD 2.5.cbeta, and I recall it calibrated ok in N and S. I wonder a FL of 430mm is over twice  the FL of the 150mm FL guide scope, bigger steps??

Pat Forster

[PATRICK FORSTER]

Here is PHD log on beta PHD2.

Pat

[Christopher Madson]

Hey Patrick,

    My Mi-250 won't clear declination backlash if I'm using pulseguiding.  The fix, as suggested by many guides and faqs from Mr. Stark and others, is to manually move your mount in declination either N or S depending on how it's oriented (mine is a "N" move) when it says clearing backlash.  This will put the star back into the correct position and get you going without error.

This is a known problem with PHD too.

Good luck,
Chris

[PATRICK FORSTER]

Hi Chris,

              Just before I started PHD calibrate, I nudged the North button on my CGEM hand controller. Star moved North and then I began calibrate. Last night I let the 'clearing backlash' go on for 50 steps, finally cleared backlash, but going North took 60 steps and failed. I revert back to PHD 1.14.2 and cal after two steps clearing backlash worked fine and Dec cal was done in 8 steps, S and N. Funny with PHD 2.05.a, W and E calibrate in 6 steps, great moves too. DEC is super tiny movements.

I use ST-4. cable is just fine for ST-4 port guiding.

Pat

[PATRICK FORSTER]

Chris, I did't manually move it when starting backlash!! I did the manual  N move before starting calibrate!!  Will try this.

Pat

[PATRICK FORSTER]

Used the cgem hc to move star North when clear dec backlash came up, star did move nicely 3 pix north and N calibrate started and 45 steps later star did not move. reverted back t0 PHD 1.14.2 and very nice cal, north did 6 steps and south finished. Seems phd 2.05a just sends tiny pulses and mount won't move at all in Dec cal.

Pat

[Bret McKee]

I have responded to Patrick privately, but wanted to update here too.  The log file that Patrick provided shows that the Dec calibration moves are generating 0 second guides. More as I discover it, but I wanted to let people know that this appears to be a code issue of some sort and not operator error.  

Bret

[PATRICK FORSTER]

Thanks Bret!

Pat

[James]

 Hello, Patrick...

 

 I would suggest some...

 

 1. Calibration times: 45 steps are too big. calibration steps should be someting about 20... you can adjust calibration steps(ms) in brain button to get some ideal calibration times. Look at the site http://www.wilmslowastro.com/software/formulae.htm  you can calculater calibration step(ms) for your system. For example, autoguider rate = 2 pixels/sec => 500ms is needed to move 1 pixel. Of course, the shorter your guidescope, the larger guidestar Dec, the more calibration steps is needed.

 

 2. I think this is maily because of motor or something problem. This is mostly because of balancing problem of DEC. I recommand you balance your scope exactly(I use a piece of wood for both arm's balancing on bed in room)

 

 = James =

[John Wainwright]

Sorry I didn't get back to you with a log, Bret, I was at GSSP, so flipped to using ST-4 guiding to avoid losing precious imaging time (see one result here - http://www.lyric.com/astro/images/M8-3-PS-2-sml.jpg, tracked with PHD2 driving a G-11, SXAO and Lodestar guidecam).

As mentioned above, the DEC backlash clear works fine with ST-4 guiding, and the driving state-machine is clearly just using the move-North calibration code, which works fine I think.  It's likely an ASCOM driver issue, though the code seems to be passing in the right duration value to the IDispatch call.   Very odd.   

I hope to set up again tonight, will try to repro on a machine with the source tree and poke about some more.

John.

[PATRICK FORSTER]

Hi Bret here is the debug file from last night. Still no DEC BL clearing on first try with PHD 2.05 Cal set to 800. Funny, but next try, I set Cal to 2000 and Dec cleared in many steps but North did not move much at all and then at the 59th N step, South started to move 3-4 pixels and off the screen the guide box went. Seems N cal sees the mount moving , vibration, wind etc and thinks its calibrating. 

Pat

[Andy Galasso]

Hi Pat,

I just looked at the log and the new logging Bret added shows that North calibration moves are being set to zero because your Dec Guide Mode setting is set to "South".

As a workaround you should set your dec guide mode to Auto before calibration, then set it to back to South after you are calibrated (assuming you really want it on South; I find Auto works quite well, but that's a separate discussion.)

Now that we understand the issue, a fix should be quite straightforward.

Andy

[PATRICK FORSTER]

Ahhh, will set DEC mode to Auto and then try it again tonight! Thanks!

Pat

[PATRICK FORSTER]

Set DEC to AUTO and by golly DEC b/l cleared and N and S calibrated in 7 steps!  Thanks. Further guiding with PHD 2.05.a was not good. Imported all of PHD 1.14.2 settings in to the beta and graph was all over off the charts, and had to stop it and revert back to 1.14.2 which was nice and smooth.

Pat

[Bret McKee]

Patrick:

Glad to hear we have made some progress.  If you still had logging on last night, please share the log files so we can see if there is anything obvious causing the bad guiding.

Thanks again for all your help,

Bret

[Andy Galasso]

I recently committed a fix for this issue, so with the latest builds the workaround of changing Dec Guide Mode to Auto before calibrating is no longer required.

If you want to try the latest build, you can copy the PHD2 r416 executable to your PHD2 install directory and run it from there. (Windows only; for Mac you'll need to wait for the next alpha or beta release.)

Andy