Dialing in my Hobym Crux 140

186 views
Skip to first unread message

neomorpheus

unread,
Nov 6, 2021, 11:02:05 AM11/6/21
to Open PHD Guiding
Good morning. Nice new mount for my Radian 61 portable setup. I'm told it is capable of excellent guiding performance.I'm using an OAG with ZWO ASI 174mm guide camera.

After several sessions of trying I finally got the PA good enough to try some imaging. I got PA error within 10 arcmin or so, which usually results in sub-1" guiding performance on my CEM60 / Esprit 120.


My first target started around 9pm and at first guiding error was in the 1.3" range so I let it run. It deteriorated and most of the night the guiding was poor.

Any suggestions? 

Thank you!

neomorpheus

unread,
Nov 6, 2021, 11:05:07 AM11/6/21
to Open PHD Guiding
One more thing - I did notice that every dither seemed to mess up the guiding for a minute before it settled down. I also got a related error from NINA: "PH2 timed out warning for the guider to settle"

bw_msgboard

unread,
Nov 6, 2021, 6:11:37 PM11/6/21
to open-phd...@googlegroups.com
Hi Neo.  There seem to be quite a number of things you will want to sort out:
 
1.  You didn't start with a fresh calibration so we can't see what you were working with for guiding.
2.  I would say there's a large amount of instability in the rig, probably with the mechanical fittings on the guide scope/guide camera assembly.  Here is just one of many examples (RA is red):
 
 
3.  You have many lost-star-events for some reason.  You need to set the min-HFD value to something that makes sense for your system, disable star-mass detection, and make sure the guide camera is as carefully focused as you can get it.
4.  The mount appears to have a lot of Dec backlash although you apparently didn't measure it with the Guiding Assistant.  That much backlash (reversal delay) makes calibration more difficult and causes settling times to be very long.
5.  To the extent we can trust any of the data, there appears to be a very large uncorrected periodic error in RA:
 
 
Most of this is coming from two frequencies in the RA drive system:
 
 
The very large resonance is at 430 sec, with the slower one somewhere around 1094 sec.  This is pretty severe and you will probably need to get some help from the mount vendor - this much tracking error can't be guided out very well.
 
If you want to establish a good measurement baseline for the new mount, you can follow this procedure:
 
 
I suspect you may need to spend an extended time measuring and trouble-shooting the mount's performance if you want to get major improvements.
 
Good luck,
Bruce
 

From: open-phd...@googlegroups.com [mailto:open-phd...@googlegroups.com] On Behalf Of neomorpheus
Sent: Saturday, November 06, 2021 8:02 AM
To: Open PHD Guiding
Subject: [open-phd-guiding] Dialing in my Hobym Crux 140

--
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.
To view this discussion on the web visit https://groups.google.com/d/msgid/open-phd-guiding/b6993b6a-000d-4a2f-9697-65edb9ca3fc8n%40googlegroups.com.
Outlook.jpg
Outlook.jpg
Outlook.jpg

neomorpheus

unread,
Nov 7, 2021, 11:13:10 AM11/7/21
to Open PHD Guiding
Thank you so much Bruce for your comments. 

Agreed I have work to do!  Since I posted this yesterday I did some more research on Hobym mounts and did some testing I'd like to share. It's getting better every time and I'm starting to get usable results I think, really appreciate your help.

I'm not sure if you're aware, but Hobym mounts us strain wave drives, and from what I gather they behave a bit differently. This may explain the PE?

Big improvement:  I found that reducing the guide exposure time from 3s (which works great on my CEM60) to 0.5s, with much larger minmo values closer to 30 than 12. This improved guiding error from 3" + to the 0.6" range!

re periodic error - I don't understand this yet, but I believe this is characteristic of the strain wave drive. The mount has a PEC Periodic Error Correction feature which I found fought with PHD2 so I turned it off. This is an area of further research, maybe my performance can be further improved. If PE is still bad but RMS error is good, is that a problem?

Odd thing: at two points in the evening I started guiding, it was excellent for a minute, then suddenly the Dec correction failed and guiding was off the chart bad (see screenshot). Did not self-correct. Stop and restart guiding, rebooting PHD2 both resulted in this pattern repeating. Only when I recalibrated PHD2 did this get back on track.



Screen shots attached.




Screen Shot 2021-11-06 at 9.00.17 PM losing Dec.png
Screen Shot 2021-11-07 at 2.53.53 AM losing Dec 2.png
Screen Shot 2021-11-07 at 2.52.25 AM good guiding.png

neomorpheus

unread,
Nov 7, 2021, 11:24:19 AM11/7/21
to Open PHD Guiding
One more significant change of note:  I switched from my OAG/174mm guide cam to a separate 120mm guidescope/ZWO120mm guidecam around the time I observed the big improvement in guiding performance.

bw_msgboard

unread,
Nov 7, 2021, 9:56:52 PM11/7/21
to open-phd...@googlegroups.com
Maybe I don't understand.  You have used the same PHD2 profile and the same guide scope focal length for the entire night, the same as on 11/5..  That shows a guider focal length of 250mm.  I find that suspicious if you went from an OAG to a separate guide scope.  When you make configuration changes like this, you *must* use separate PHD2 profiles - run the new-profile-wizard for each new profile and give them a unique name.  If you've made this sort of mistake, you can't draw any conclusions from your recent session.
 
Bruce

From: open-phd...@googlegroups.com [mailto:open-phd...@googlegroups.com] On Behalf Of neomorpheus
Sent: Sunday, November 07, 2021 8:24 AM
To: Open PHD Guiding
Subject: Re: [open-phd-guiding] Dialing in my Hobym Crux 140

neomorpheus

unread,
Nov 8, 2021, 7:42:42 AM11/8/21
to Open PHD Guiding
Interesting. Now that you mentioned it, though I re-ran darks and bad pixels and re-calibrated when I changed cameras, I did not re-run the wizard so it would not have updated the focal length of the guide scope (makes me wonder why the guiding was ok ...). 

I will start with the baseline procedure you shared tonight before testing again.

Thanks

bw_msgboard

unread,
Nov 8, 2021, 10:37:31 AM11/8/21
to open-phd...@googlegroups.com
That means the "improved" guiding was probably mostly illusory.  The guiding results are normally shown in units of arc-seconds which means the guider image scale must be known.  In your case, the scale was wrong so all of your results are probably off by the factor of new_focal_length/old_focal_length. 
 
With regard to your question about the mount being "different" while having a large periodic error: there's no place to hide on this.  In the end, for the purposes of guiding and imaging, only two things really matter - how accurately the mount can track the night sky and how well it responds to small guide corrections in all 4 directions.  It doesn't matter if the drive system is fancy robotics gear or ropes and pulleys - all that matters is the end results.  The RA tracking behavior I showed you in the original message looks quite bad and will probably degrade your imaging.  That's why I suggested you show it to the mount vendor and ask for help from them - I doubt they would tell you the behavior is expected.
 
Good luck, 
Bruce

From: open-phd...@googlegroups.com [mailto:open-phd...@googlegroups.com] On Behalf Of neomorpheus
Sent: Monday, November 08, 2021 4:43 AM

neomorpheus

unread,
Nov 9, 2021, 11:06:34 AM11/9/21
to Open PHD Guiding
I spent some time last night working on the guiding after getting the equipment profile right i.e. 120mm focal length on the guide cam. Overall results were stable but still in the 1.2" to 1.7" range. 

I started by getting a clean baseline calibration and let it run for 20 min or so. Turns out PEC was on at this point. I then tried a few variations, including PEC on/off to confirm PEC was helping a bit.

Sometimes for no apparent reason the mount will just stop making RA adjustments (you can see this at the end of some of the 20 min baseline tests early in the evening).

https://openphdguiding.org/logs/dl/PHD2_logs_59md.zip

Could I bother you to have one more look at this clean data and see what conclusions can be made? 

Thank you so much.

Mike


bw_msgboard

unread,
Nov 9, 2021, 8:22:44 PM11/9/21
to open-phd...@googlegroups.com
Hi Mike.  The guiding didn't stop at the times you're talking about.  Instead, something caused gigantic excursions in RA, as large as 5-7 arc-minutes.  When those occurred, the mount subsequently didn't respond to the guide corrections.  You'll have to determine the origin of these giant tracking errors - something wrong with the mount, the guiding assembly moving around, a cable pulling, whatever.  These are not guiding problems.
 
I don't see anything here to change my earlier opinions.  The mount has a huge amount of tracking error, as much as 40-60 arc-sec peak-peak.  Until you can improve this, I don't expect you'll see any signiicant improvement in guiding results.  If you've decided you don't want to get the mount fixed, you can try using the PHD2 PPEC guiding algorithm for RA - start it with a period length of around 440 seconds and let it run.  You've also adopted a guide scope approach that leaves you very vulnerable to mechanical weakness and lack of rigidity.  A movement of the guide camera by 4 microns - less than 10% of the thickness of a human hair - creates an apparent guide star excursion of over 6 arc-sec.  Your Dec drift rate also seems to be all over the map suggesting to me that the mount isn't holding its alt/az position after you make adjustments for polar alignment.
 
I'm sorry but I really don't know what else to tell you.  Good luck with it,
Bruce

From: open-phd...@googlegroups.com [mailto:open-phd...@googlegroups.com] On Behalf Of neomorpheus
Sent: Tuesday, November 09, 2021 8:07 AM

neomorpheus

unread,
Nov 10, 2021, 7:16:31 AM11/10/21
to Open PHD Guiding
Thanks so much Bruce, I really appreicate your assistance in diagnosing the performance.

I thought maybe with cleaner data the conclusion may have changed but it appears there's something wrong. The images don't look so bad compared to the big errors you're showing me, but just the same this performance is unacceptable for a $4000 mount. 

May I ask what tool you used to generate the "Analysis" charts?  I can take it from there.

Thanks again,

Mike

bw_msgboard

unread,
Nov 10, 2021, 10:09:33 AM11/10/21
to open-phd...@googlegroups.com
Hi Mike.  On this page:
 
 
Look for
 
 
 
If you get any useful help from the mount company, maybe you could report back - you probably won't be the only person having this problem.
 
Good luck,
Bruce

From: open-phd...@googlegroups.com [mailto:open-phd...@googlegroups.com] On Behalf Of neomorpheus
Sent: Wednesday, November 10, 2021 4:17 AM

neomorpheus

unread,
Nov 10, 2021, 2:47:31 PM11/10/21
to Open PHD Guiding
I have been using that logviewer shown in the Basic Analysis tutorial you shared. You also showed me some "analysis" graphs that showed the PE error? The one with buttons "drift corrected" and frequency analysis (sample attached). 

Happy to report back. A lot of guiding issues with these mounts.

Thanks



Outlook-2.jpg
Outlook-3.jpg

Brian Valente

unread,
Nov 10, 2021, 2:51:55 PM11/10/21
to Open PHD Guiding
Mike

right-click on the graph window in Log Viewer, you will see option for Analyze selected raw RA
image.png

choose Frequency Analysis button and you will get some analysis of PE with corrections backed out. please remember this is a PHD estimate of errors

image.png



To view this discussion on the web visit https://groups.google.com/d/msgid/open-phd-guiding/e9a9844c-db79-4714-8ea3-e98b5f064d81n%40googlegroups.com.


--
Brian 



Brian Valente

neomorpheus

unread,
Nov 10, 2021, 6:55:45 PM11/10/21
to Open PHD Guiding
Thanks Brian.

neomorpheus

unread,
Nov 10, 2021, 7:49:14 PM11/10/21
to Open PHD Guiding

Regarding cables, how much resistance makes a difference?

I can pretty much rule out snags, it's a simple system and 3 cables nicely hang from the rig. That said there is *some* weight to one of them. How much pull is enough to affect guiding?

Brian Valente

unread,
Nov 10, 2021, 9:02:34 PM11/10/21
to Open PHD Guiding
Hi Mike

>>>How much pull is enough to affect guiding?

It could impact it some, but the picture that Bruce is painting suggests to me that your mechanical issues (or i guess more accurately your non-guiding related issues) are far more than cable weights would account for, particularly given your description

Brian

To view this discussion on the web visit https://groups.google.com/d/msgid/open-phd-guiding/be084f23-3d83-470b-a5fb-a7bea738df19n%40googlegroups.com.

neomorpheus

unread,
Nov 13, 2021, 5:46:42 PM11/13/21
to Open PHD Guiding
I've been speaking with the manufacturer, nothing conclusive yet. He keeps telling me to read the appendix on the manual p48-52  . Myself I don't find anything actionable, but it does talk to unique characteristics of the strain wave drive that may cause confusion for users. May be of interest.

Reply all
Reply to author
Forward
0 new messages