Jump to content

  •  

CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.

Photo

PHD2 backlash clearing ARGHHH!

  • Please log in to reply
9 replies to this topic

#1 Ricepow

Ricepow

    Lift Off

  • -----
  • topic starter
  • Posts: 10
  • Joined: 28 Nov 2017

Posted 23 February 2020 - 05:12 PM

Have a "new to me" LX200R 16", and up until about a month ago PHD2 was doing great with calibration and guiding. I've got it set up in ASCOM with the meade LX200GPS/R driver, and I've deleted and reloaded both and it and the PHD2 program multiple times in the last month, so they're both up to date. I still had a residual 1.5 arcsecond error in RA that was driving me bonkers so I went in and cleaned/regreased both RA and DEC drive worms, beefed the springs up, and also downloaded PEMPRO trial version and tried (with no success) to get it to dump a better correction curve onto the PEC. After all of this, I tried to set up and run PHD2 again and found to my horror that it REFUSES to complete the calibration. It will perform the RA (or west/east) calibration, but when it then goes to "clear the backlash" it will send clear commands for hundreds of cycles w/ no tangible result, and subsequently the DEC calibration never completes. BUT WAIT! It gets better! "It's your cable!" I hear you say, but i can put it in manual guiding and it moves quickly and w/ very little backlash to any command either north or south sent to the mount via the phd2 manual guiding. Which tells me the cable, and serial to usb adapter is working fine. But it refuses to send declination commands either in calibration or in normal automatic guiding. I have looked at the guide logs and they mirror that. Also of interesting report is that the calibration graph shows that the RA/DEC axis are colinear. And by that I mean they exhibit 88.8 degrees of orthogonality error, so essentially colinear. When you watch the guiding graph, the declination error and the RA error or ON TOP of each other, ie the same. It's like somehow the declination axis has vanished from the telescope. I'm wondering if pempro somehow made a deep dark firmware change to my scope's control system?

 

Anyway, I'm about ready to pull my hair out, and I'm open to suggestions. I've attached one of the dozens of log files which thankfully don't record audio or you'd hear a stream of astronomy focused profanity. 

 

Thanks,

Frustrated as Hell in Texas

Attached Files



#2 John Miele

John Miele

    Soyuz

  • *****
  • Posts: 3,827
  • Joined: 29 May 2005
  • Loc: North Alabama

Posted 23 February 2020 - 05:24 PM

PHD2 can get pretty squirrely at times...this is a long shot...but you might just try to remove and reinstall your PHD2 and see if that helps. I had a bizarre error with PHD2  where it randomly changed my calibration step size no matter what I set it to. The error went away after said procedure...you never know maybe PEMPRO wigged out a setting in PHD2?


Edited by John Miele, 23 February 2020 - 05:25 PM.


#3 Jim Waters

Jim Waters

    Skylab

  • *****
  • Posts: 4,051
  • Joined: 21 Oct 2007
  • Loc: Phoenix, AZ USA

Posted 23 February 2020 - 05:41 PM

Have you posted for help here?

 

https://openphdguidi...g/getting-help/



#4 Ricepow

Ricepow

    Lift Off

  • -----
  • topic starter
  • Posts: 10
  • Joined: 28 Nov 2017

Posted 23 February 2020 - 05:59 PM

Jim, no but I will.
I will also uninstall and reinstall PhD to with pempro already uninstalled

It literally acts like I have turned the declination to off. I did this once before early on and I had the declination turned off by mistake, but this time I have tried it in North South and
auto and nothing changes

#5 Jim Waters

Jim Waters

    Skylab

  • *****
  • Posts: 4,051
  • Joined: 21 Oct 2007
  • Loc: Phoenix, AZ USA

Posted 23 February 2020 - 06:02 PM

Make sure you delete the PHD2 directors too.  There are 2.  One for the program and one for logs and settings.



#6 BobE102330

BobE102330

    Sputnik

  • -----
  • Posts: 40
  • Joined: 26 Sep 2019

Posted 23 February 2020 - 06:58 PM

Guide speed in the ASCOM driver? It defaults to 0.1 x sidereal on install. I was bitten by that.

The other one that got me was step size in PHD2 - too small and it won’t move enough to calibrate. The bad cable suggestion annoys me, too.

Edited by BobE102330, 23 February 2020 - 07:00 PM.


#7 Ricepow

Ricepow

    Lift Off

  • -----
  • topic starter
  • Posts: 10
  • Joined: 28 Nov 2017

Posted 23 February 2020 - 10:34 PM

Fantastic suggestions! Man I love this place!
I will try all of those, and keep this thread updated, maybe it’ll save someone a similar headache. Thanks and keep em coming, I’ll let y’all know when it’s resolved.

#8 Mert

Mert

    Fly Me to the Moon

  • *****
  • Posts: 6,788
  • Joined: 31 Aug 2005
  • Loc: Spain, Pamplona

Posted 24 February 2020 - 01:12 PM

In your logfile I see a calibration stepsize of 400 which seems to be

quit short.

At least start with 1500 or more!

Just a wild guess!!



#9 Ricepow

Ricepow

    Lift Off

  • -----
  • topic starter
  • Posts: 10
  • Joined: 28 Nov 2017

Posted 25 February 2020 - 12:45 PM

Uninstalled and reinstalled PHD2 last night. After getting it all ready, I noticed it again set the calibration step very low......like 300-400 ms, and with this setting it again would not clear backlash in something like 40-50 steps. So I turned it up to 1100 ms (after some trial and error) and it FINALLY resulted in movement of the declination axis, and then after a very ugly dec calibration cycle it "calibrated" with a warning that the dec axis rate was less than half the RA. It was indeed, but at least it had moved and it was orthogonal to the RA, so that issue was now on the ropes. So then I started running the calibration multiple times, as well as manually moving north A LOT prior to initiating the calibration in an effort to get all of the perceived backlash out. No love. 

Then I started to really observe what it was doing, and I realized what I was seeing was the declination drive was stick-slipping, so to speak. It would sent multiple 1100 ms move commands w/ no result, and then it would move a small amount for a couple command cycles, and then it would again dwell for a few more cycles before moving again. When it did move, the rate was very comparable to the RA rate, but it only actually moved less than half the time. Prior to my "axis tuneup" this was not the case. So now I'm feeling like what I may be seeing is that I may now have too much spring pressure on the worm causing it so "bind" until the nylon output gears have "loaded up" with enough tooth strain to force the move.

I added a fairly hefty compression spring to what was already there, probably 3-5 times higher spring constant, and also replaced the grease with the recommended lubriplate/friction grease mix. I doubt the grease was the issue, but I could certainly see the spring load being excessive. I checked balance as I have an large FLI CCD/filter wheel/focuser assembly hung off the end. It's not perfect, but it's also not horrendous, probably less than a lb of counterweight needed, so I don't think it's that. And besides it acts screwy in south as well as north calibration moves, not what you'd expect from a biased gravity load. 

So, the plan tonight is to pull the uber heavy compression springs and swap them out with something a bit lighter to see if that makes a difference. Fingers crossed.

After the janky calibration last night, I did, for the first time, see RMS error hovering around 0.75" for the first time EVER for relatively extended periods. Weak sauce for some of you guys, but absolutely amazing for me. 



#10 Ricepow

Ricepow

    Lift Off

  • -----
  • topic starter
  • Posts: 10
  • Joined: 28 Nov 2017

Posted 27 February 2020 - 10:25 AM

The adventure continues. I pulled the heavy springs and tried lighter springs.....better, but was still getting a 165 to 200+ % delta between RA and DEC rates......So I pulled the auxiliary springs and went back to the original plunger only. If I snug it up to get good preload on the worm/gear interface, it begins exhibiting the stick/slip behavior again. If I back it off to a very light pressure I get almost immediate backlash clearing BUT, when I try to guide after the calibration, it acts like the DEC signals aren't resulting in any motion, so I'm guessing there's a happy compromise somewhere. I'm also beginning to wonder about how sensitive the system is to dec balance, and how good an idea the friction grease was. 




CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.


Recent Topics






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics