Calibration and Guiding LX200

[Daniel Low]

I have had trouble setting up guiding with my 8" LX200 Classic.  I have worked for a while to improve various aspects of the setup to get the calibration procedure to be stable and reliable.  Here is what I did.

1) Upgraded to the Audiostar controller from George Dugash (it works great!), this also included an inspection of both the RA and Dec drive motor/gear systems.

2) Balanced the scope, I balance both the RA and DEC axes using fixed and movable weights

3) Upgraded the wedge. I purchased the wedge from Ken Milburn. I had made modifications to the original wedge so that it worked fairly well, but the new wedge is fantastic

4) Used PhD2's polar drift alignment to get my alignment to within 4-6 minutes of arc.  

5) Trained the drive, using 45 minutes of training data.  

With the alignment and training, a star on the celestial equator moves very little by just tracking, maybe 10 s of arc over 10 ish minutes.  

I use an Altair GPCAM2 290M guide camera, usually with an OAG.  For these tests, however, I used only the guide camera with no OAG.  

I disabled declination guiding.  When I requested the PhD2 system to calibrate for guiding using a 250 ms guide pulse, the westward motions were almost nonexistent, just a few pixels, the eastward motions were substantial and the system failed to calibrate.  When I set the guide pulses to 150 ms, the motions were more similar and the system did calibrate.  

Could this be backlash?  When I command east or west motion via the hand controller, the star field moves predictably with the west and east motions appearing to be similar.  The same happens when I use the PhD2 manual guide option. The west and east motion are almost identical.  

I attached a screen grab of my latest calibration where I did both RA and DEC. Note that the West motions are much smaller than the East motions and that the scope has to swing back to the originating point after moving Eastward.

Is there some setting that I have incorrect?  Why would East and West motion speeds be so different? 

The PEC training worked extremely well, I had been blaming the previously nonexistent PEC as a potential source of this problem, but no longer.  It appears that the PhD2 moves the scope a different speed when moving west during calibration than east, so much so that it overshoots dramatically.    

Also, it would be great if I could select a relative guide speed less than 1. Does anyone know if this is possible? That should eliminate backlash as a potential problem.  

Thanks,

Dan

[bw_msgboard]

Hi Dan, sorry you’re having trouble with your mount.  We’ll need to see both the guide and debug logs for a session where you had these problems:

 

https://openphdguiding.org/getting-help/

 

Just to clear up a few other points:

 

1. Backlash is irrelevant in RA so long as you’re using 1x guide speed or less.  
2. PHD2 has no control over the guide speed and doesn’t actually care what the guide speed is.  That really only affects the size of the calibration step that’s used.  If you’ve used the new-profile-wizard and supplied correct values for the focal length, pixel size, and mount-controlled guide speed, the calibration step-size is computed for you.  Alternatively, you can use the ‘Calculate’ button on the Guiding tab of the Advanced Dialog to set it yourself.
3. It’s very important that the RA and Dec backlash settings in the mount be set to zero
4. Some of the old Meade mounts may force the guide speed to be something like 0.5x sidereal regardless of what the hand-controller may tell you.  I believe the default value for Autostar is 0.5x.
5. The PHD2 Manual Guide tool uses exactly the same machinery for moving the mount as calibration does.  However, you may be sending much larger guide pulses depending on what you’ve set in the Manual Guide UI.  So you may be using a calibration step-size that’s much too small.  Again, that can be fixed as described in item 2.

 

Once we have your log data, we can probably have more to say about what’s going on.

 

Cheers,

Bruce

 

---

--
You received this message because you are subscribed to the Google Groups "Open PHD Guiding" group.
To unsubscribe from this group and stop receiving emails from it, send an email to open-phd-guidi...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Daniel Low]

Thanks for getting back to me.

I looked for it, but none was recorded last night.  I didn't do much actual tracking, I spent most of the evening training and then tried a couple of calibrations, one with 250ms and one with 100ms (the example I attached).  I did track (RA only) for about 3 minutes and it worked great, likely because I had trained the drive and the scope was aligned to within a few minutes of arc.  

Given that my tracking session last night wasn't recorded, how long does a tracking session need to be to be recorded?  

I will send the log later tonight or tomorrow.

See answers below:

Thanks!

Dan

On Tuesday, June 12, 2018 at 12:04:50 PM UTC-7, Bruce Waddington wrote:

Hi Dan, sorry you’re having trouble with your mount.  We’ll need to see both the guide and debug logs for a session where you had these problems:

 

https://openphdguiding.org/getting-help/

 

Just to clear up a few other points:

 

1. Backlash is irrelevant in RA so long as you’re using 1x guide speed or less.  Got it...
2. PHD2 has no control over the guide speed and doesn’t actually care what the guide speed is.  That really only affects the size of the calibration step that’s used.  If you’ve used the new-profile-wizard and supplied correct values for the focal length, pixel size, and mount-controlled guide speed, the calibration step-size is computed for you.  Alternatively, you can use the ‘Calculate’ button on the Guiding tab of the Advanced Dialog to set it yourself.  I did use the calculate button.  
3. It’s very important that the RA and Dec backlash settings in the mount be set to zero  They are set to zero.  
4. Some of the old Meade mounts may force the guide speed to be something like 0.5x sidereal regardless of what the hand-controller may tell you.  I believe the default value for Autostar is 0.5x.  That would make sense, the RA drive doesn't stop when one is guiding east. 
5. The PHD2 Manual Guide tool uses exactly the same machinery for moving the mount as calibration does.  However, you may be sending much larger guide pulses depending on what you’ve set in the Manual Guide UI.  So you may be using a calibration step-size that’s much too small.  Again, that can be fixed as described in item 2.  Good to hear. 

[bw_msgboard]

Hi Dan.  If you ran PHD2 at all, there will be a debug log file.  If you attempted a calibration, there will be a guide log.  So I think the logs are available to you.  If you’re running 2.6.5 – which you should be – the Upload Logs menu item will display all the available log files including their timestamps.

 

Cheers,

Bruce

 

---

From: open-phd...@googlegroups.com [mailto:open-phd...@googlegroups.com] On Behalf Of Daniel Low
Sent: Tuesday, June 12, 2018 12:51 PM
To: Open PHD Guiding
Subject: Re: [open-phd-guiding] Calibration and Guiding LX200

 

Thanks for getting back to me.

 

I looked for it, but none was recorded last night.  I didn't do much actual tracking, I spent most of the evening training and then tried a couple of calibrations, one with 250ms and one with 100ms (the example I attached).  I did track (RA only) for about 3 minutes and it worked great, likely because I had trained the drive and the scope was aligned to within a few minutes of arc.  

 

Given that my tracking session last night wasn't recorded, how long does a tracking session need to be to be recorded?  

 

I will send the log later tonight or tomorrow.

 

See answers below:

 

Thanks!

 

Dan

 

On Tuesday, June 12, 2018 at 12:04:50 PM UTC-7, Bruce Waddington wrote:

Hi Dan, sorry you’re having trouble with your mount.  We’ll need to see both the guide and debug logs for a session where you had these problems:

 

https://openphdguiding.org/getting-help/

 

Just to clear up a few other points:

 

1.      Backlash is irrelevant in RA so long as you’re using 1x guide speed or less.  Got it...

2.      PHD2 has no control over the guide speed and doesn’t actually care what the guide speed is.  That really only affects the size of the calibration step that’s used.  If you’ve used the new-profile-wizard and supplied correct values for the focal length, pixel size, and mount-controlled guide speed, the calibration step-size is computed for you.  Alternatively, you can use the ‘Calculate’ button on the Guiding tab of the Advanced Dialog to set it yourself.  I did use the calculate button.  

3.      It’s very important that the RA and Dec backlash settings in the mount be set to zero  They are set to zero.  

4.      Some of the old Meade mounts may force the guide speed to be something like 0.5x sidereal regardless of what the hand-controller may tell you.  I believe the default value for Autostar is 0.5x.  That would make sense, the RA drive doesn't stop when one is guiding east. 

5.      The PHD2 Manual Guide tool uses exactly the same machinery for moving the mount as calibration does.  However, you may be sending much larger guide pulses depending on what you’ve set in the Manual Guide UI.  So you may be using a calibration step-size that’s much too small.  Again, that can be fixed as described in item 2.  Good to hear. 

[Daniel Low]

The log files are at:

https://openphdguiding.org/logs/dl/PHD2_logs_Ghtn.zip

Thanks!

Dan

[bw_msgboard]

Hi Dan.  Let’s start by trying to connect the dots on your configuration.  You said you had removed the OAG and were using the guide cam without an OAG.  I assume that means you simply attached the guide cam to the focuser on the Meade SCT?   I think the normal optical setup for the classic 8-inch Meade SCT was f/10, but the focal length you entered in PHD2 suggests you’re using the f/6.3 focal reducer.  Is that correct?

 

I guess the bad news here is the detailed log data just confirms what you saw visually – the mount is not responding correctly to west guide pulses, they’re not even close.  Even with the last calibration, the one that raised the alert message about inaccurate rates, the numbers are way off the mark.  Using exactly the same sequence of guide pulses of exactly the same length,  this is what PHD2 saw:

    West moves took 20 pulses of 150ms to see the guide star move 28 px

    With the same 20 pulses moving east, the guide star moved by 90 px

    North moves took 8 pulses of 150ms to see the guide star move 29 px

    The same 8 pulses moving south moved the guide star by 20 px – a common symptom of Dec backlash, nothing to worry about at this point

 

This is pretty clearly a problem with the mount controller, the mount RA drive system, or the ASCOM driver you’re using.  That assumes the RA axis really is well-balanced and the clutches are tight.  Since you’re apparently using some kind of 3^rd party Autostar controller, I think that would be the place to start asking questions.  The original classic LX200 mount controller didn’t support pulse-guiding in the normal way if I remember correctly.  So the pulse-guiding functions might be handled in the ASCOM driver or there would have to be custom code in the mount controller.  The driver is reporting that the mount is guiding at 1x sidereal, and the north calibration supports that – it looks about right.  You mentioned before you couldn’t see the problem using the Manual Guide tool.  You should try it again using 150ms pulses – that should duplicate what is done during calibration.  If you find the problem disappears with larger pulse sizes, that will be good information to have.

 

Beyond that, I would suggest binning the guide camera if you can.  With this mount, you are likely to have a lot of trouble trying to guide at 0.5 arc-sec/px.  If you can bin the camera 2x2, you’ll get better SNR ratios on the guide stars, be able to use fainter stars, and generally make life easier for yourself.   Next, you’ve chosen to use non-default guide algorithms for both RA and Dec, which is not advisable at this point.  Your mount will almost certainly do better with the default algorithms – Hysteresis for RA, Resist-switch for Dec.

 

Hope this helps,

Bruce

 

---

From: open-phd...@googlegroups.com [mailto:open-phd...@googlegroups.com] On Behalf Of Daniel Low

Sent: Tuesday, June 12, 2018 1:54 PM
To: Open PHD Guiding

--

[Daniel Low]

To close the loop for other group readers:

I went out two weeks ago and found that my mount worked much better than before.  This was what I did:

I used a star chart to crudely find the north celestial pole, using a 40mm eyepiece to align the mount.  The Ken Milburn wedge allowed me to more easily adjust this position, but the old wedge would have worked, just been more frustrating to set up.  

I used PhD2 polar drift align to adjust the polar alignment to <4 minutes of arc (I used my guide camera as the imaging camera for this process). This took me only about 15 minutes.  

Assembled my imaging gear, including ASI1600MM, OAG, focuser, and filters, balancing the scope after assembling the imaging gear. The RA balance was done in advance (I have two weights I screw into threaded holes I placed in the fork). The Dec balance required me to loosen the dec lock knob. I first positioned the scope at a star near the celestial equator and meridian, loosened the dec lock knob, adjusted the weights, aligned the scope to that star, then locked the dec axis. 

The polar alignment was very stable, I was able to guide without dec guidance, keeping the dec alignment to within 2-3 seconds of arc for 5 minutes at a time.  I used manual motion whenever I needed to tweak dec alignment between images.  I used 150 ms pulses, which led to approx 2 s dec motion.  

My previous experience with the scope was frustrating, I kept seeing mirror "flop", about every 2-5 minutes, with >10 arc s of rapid motion in the declination axis that would ruin my exposure. I had assumed that focusing the scope by always rotating the focus knob in a specific direction would solve this, but it didn't.  This time, I elected to leave the focus knob between clockwise and counterclockwise engagement, hoping that I would leave the mirror without stress, and this proved very successful.  I use an external focusing system for fine tuning, so I don't need the main focuser to be accurately set.  

I ran two nights and had almost no mirror flop (only about 5 times in two nights!).  My RA and Dec guidance were within 2 s and I had only one image ruined by poor guidance (probably an image when mirror flop occurred).  Focus was a bit different, I did have some drift in focus during one of the imaging sessions that I didn't detect or look for. The evidence was that stars at the corners of the image came slightly out of focus.  That may have happened during one of the few flop instances.  I will do more QA during the night to detect defocusing.  Still, my image of the Omega Nebula, acquired using narrow band filtering, came out much better than any previously acquired image.  I enclosed my favorite version of that image.  

Thanks to PhD2 for providing an excellent toolset.  I will continue to find tune the RA guiding algorithms to see if I can get better than 2 s performance, but I am quite pleased with the current performance.

Dan

[mj.w...@gmail.com]

Hi Daniel

Are you still getting different move distances on RA and Dec?

From what I remember of bolting Autostar onto other mounts, the handset is used to set a motor rate for each axis ?

So possibly:

1. The rates entered aren't the same

2. Motors with different gearing have been accidently supplied, in which case you could experiment with the figures until you get a match.

Other than that, I like your image, but I'm surprised you got that with 3 arcsec moves on Dec.

Michael
Wiltshire UK

[Daniel Low]

I don't think my biggest problem is differences in Dec and RA, but differences in motion East and West.  This causes the calibration to miscalculate RA motion, perhaps leading to differences in Dec and RA speeds.  Why the East and West motions are so different is unclear, but it would be great to figure out. Since I needed only RA guiding, the problem calibrating RA and Dec was avoided.  I would like to be able to use Dec guiding, but the alignment process removed that requirement.  

[Daniel Low]

Update and question/request

Since my June posts, I replaced the plastic bearings with needle bearings on my LX200.  Needless to say, the friction is much less.  I went out Saturday to the mountains and set up the scope, polar alignment via PhD2 was 4 minutes of arc.  

I balanced the scope and calibrated RA and DEC. With the stiction greatly improved, I don't get the bizarre calibration behavior I used to, but still have way too much backlash for a good calibration set.  I will work on the declination drive next, tightening it up, but i was wondering, since there are four declination options for guiding, off, auto, north, south, why are there only two effective options for calibration (both and off)?

Can we get a feature that, if north or south guiding is selected, the calibration will only calibrate north or south and ignore the amount of declination backlash it needs to overcome to get back to the original location? That would eliminate my need to greatly reduce the declination backlash. I would align the scope, go to the object I want to image, calibrate with either a north or south selection, offset the polar alignment just a bit to cause the same direction of drift, then image, with the scope only needing to do a one-directional declination drift.  

Any chance that this could be implemented?

In any case, I will continue to work with the drive to remove as much backlash as possible.

Thanks,

Dan

On Tuesday, June 12, 2018 at 10:54:28 AM UTC-7, Daniel Low wrote:

[bw_msgboard]

Hi Daniel.  Here’s an excerpt from the help docs regarding calibration:

 

Although PHD2 moves the guide star in all four directions, only the west and north movements are actually used to compute the guide rates and camera angle. The east and south moves are used only to restore the star roughly to its starting position.

 

So your feature request really isn’t needed – the calibration is based on guiding in only the west and north directions.  If you have a lot of backlash, you can manually clear it before you start the calibration.  Just make sure the last slew direction was north or use the hand-controller to move the mount north immediately before starting the calibration.  If you do a reasonable (not necessarily perfect) job of clearing the Dec backlash this way, your calibration results should be fine.

 

Hope this helps,

Bruce

 

---

From: open-phd...@googlegroups.com [mailto:open-phd...@googlegroups.com] On Behalf Of Daniel Low
Sent: Monday, October 08, 2018 1:51 PM
To: Open PHD Guiding
Subject: [open-phd-guiding] Re: Calibration and Guiding LX200

 

Update and question/request

--

[Andy Galasso]

Hi Dan,

I'm with Bruce in that the feature request is probably not needed. Just make sure you nudge the mount North before starting calibration.

You could also enable PHD2's backlash compensation (use the Guiding Assistant to measure and set a reasonable starting value, and PHD2 will adjust the compensation automatically over time.)

Lots of us image with mounts with lots of dec backlash.  As long as you have a little bit of polar mis-alignment and set your Dec min motion setting high enough that PHD2 is not chasing seeing, there should be very few Dec correction direction reversals. Selecting the spiral dither option can reduce the number of Dec direction reversals caused by dithering.  And when dithering does cause a Dec direction reversal, the dynamic Backlash Comp can quickly work through the backlash.

HTH,

Andy

[Daniel Low]

Wonderful to know!  My dec axis used to have lots of stiction, which caused odd twisting motion in both RA and Dec when switching declination direction. That appears to be gone now with the new bearings, but the backlash is still there.  

I will purposefully tweak tweak the polar alignment to have a slight south-drift bias and try north-only guiding the next time I get out. 

I did try using PHD's backlash but my backlash is still too large (4 ish seconds) for it to be effective. I will be looking at my declination drive to reduce backlash, hopefully with the endpoint that I can get away with using the PHD backlash and guide both directions.  My polar alignment is usually good enough to need only very slight corrections, and I would like to leave the scope unattended during image acquisition.  

In any case, thanks for the feedback, I trust that either doing north-only drift correction or removing backlash will cure my declination issues!

Dan

[Daniel Low]

By the way, I don't mean to sound like I am complaining about PHD, I am not!  I have gotten wonderful photos using PHD2! 

[Andy Galasso]

On Tue, Oct 9, 2018 at 12:25 PM Daniel Low <lowdo...@gmail.com> wrote:

By the way, I don't mean to sound like I am complaining about PHD, I am not!  I have gotten wonderful photos using PHD2! 

No worries, it did not come off that way to me. We appreciate your feedback.

Nice image... perfectly round stars ... no guiding issues there!