Bad Performance in DEC while Guiding, iOptron CEM25P

182 views
Skip to first unread message

Mine Be

unread,
Apr 22, 2020, 9:36:52 AM4/22/20
to Open PHD Guiding
Hey guys,
I bought an iOptron CEM25P as my entry level mount for astrophotography in january. Due to very bad weather, I could only use the mount the first time 4 weeks ago. I did my first imaging session on M51 unguided, which went okay. But then I decided I want autoguiding and longer exposures, so I researched how to handle PHD2 guiding, but that's when the problems arised last week on my second session on M101. I had very bad guiding in Dec and had to throw away tons of subs due to weird looking star shapes (example below). I've read a lot of possible solutions and tried to eliminate user errors already.

I'm using this equipment:
iOptron CEM25P
TS Photoline 80/560mm f/7
Long Perng Adjustable Field Flattener
Orion Mini Deluxe Guidescope 50/162mm
ZWO Asi 120mm Mini
Canon 600D

And so far I've done the following to eliminate any possible user error:
I've checked double if the guidescope is aligned with my main scope in daylight, I've got into polar alignment with Sharpcap and got roughly ~3 arcminutes of polar alignment error, I've set up the mount completely level and stable, I've done cable management so no dragging or stuck cables, I've balanced the mount very carefully and even in DEC and RA as that's what you are supposed to do with iOptron CEM mounts, I'm using pulse guiding via USB/serial cable, so no ST4 guiding, I've downloaded the correct drivers and reinstalled them again already from the iOptron and Zwo sites, I've carefully focused my guidecam with Sharpcap, I've tried guiding in one direction (tried both nord and south, the star either drifts away and then no guide impulses occur), did calibration near the celestial equator, etc...
However, I often got the little south movement detected, and according to the guiding assisstant I've got HUGE backlash in DEC with anything from 14000-20000 ms. I'm throwing last nights guiding log here in as well, in RA I got no problems, but DEC is horrendous.

My workflow last night was like this:
1) Level tripod very carefully with a level
2) Put on mount head, adjust to 50 latitude
3) Assemble Counterweight and Telescope with everything
4) Calibrated guidescope so it shows the same FOV as my main scope by daylight
5) Install every cable and carefully manage them so nothing can get stuck and is able to move freely
6) Balance the mount in DEC and RA
7) Get Polaris in the middle of the polar scope
8) Get Polaris to appear in the Canon Liveview as good as centered as possible
9) Set as Zero Position
10) Focus the Guidecam in Sharpcap
11) Polar Alignment with Sharpcap, got down to roughly 3.5 arcmins error
12) PHD2 Calibration near the celestial equator
13) Let the Guiding Assistant run for a couple of minutes
14) Sync to Target with Dubhe on M81
15) Center M81 in Liveview
16) Start guiding
17) Start imaging

However nearly every frame had stars like this and my guiding graph looked horrendous in DEC last night. I'm putting the log from last night here as well, and I'm also showing an example image of the horrible stars.

I don' t know what to do anymore, I just want to got out and image but all I do right now is trouble shooting...
Is my mount broken?

Thanks and clear skies!!
PHD2_GuideLog_2020-04-21_100935.txt
weird stars.jpg

Richard Beck

unread,
Apr 22, 2020, 10:55:43 AM4/22/20
to Open PHD Guiding
When you get out again, make sure you accept all GA (guiding assistant) recommendations.  It appears from my initial inspection that the backlash compensation, at a minimum, from the GA was not accepted.  If GA didn't give you a checkbox for backlash compensation, look for potential issues which cause the apparent backlash measured by GA to be reduced to a value that does provide an accept button.

Mine Be

unread,
Apr 22, 2020, 11:07:46 AM4/22/20
to Open PHD Guiding


Am Mittwoch, 22. April 2020 16:55:43 UTC+2 schrieb Richard Beck:
When you get out again, make sure you accept all GA (guiding assistant) recommendations.  It appears from my initial inspection that the backlash compensation, at a minimum, from the GA was not accepted.  If GA didn't give you a checkbox for backlash compensation, look for potential issues which cause the apparent backlash measured by GA to be reduced to a value that does provide an accept button.


Thanks! I'm having backlash reported somewhere between 14000-20000 ms each time I ran the GA, so backlash compensation isn't even showing up anymore...

Richard Beck

unread,
Apr 22, 2020, 11:33:05 AM4/22/20
to Open PHD Guiding
Seeing you have the spring-loaded clutches, your issue probably isn't a DEC clutch that is too tight, which was one of my issues.  Other things that I worked on were balancing the DEC axis as well as I could and minimizing any issues of drag on the DEC axis.  When balancing the DEC axis, it's important to have the focuser near optimum focus and all lens caps off so the center of gravity of the imaging train while balancing is consistent with the center of gravity while imaging. 

There is also a possibility of differential flexure between your guide scope and your imaging scope.  Tighten these down as well as you can.  You can also share a photo of your set up where folks on the forum could provide suggestions based on your set up.

bw_msgboard

unread,
Apr 22, 2020, 12:18:16 PM4/22/20
to open-phd...@googlegroups.com

Hi Mine.  I think the situation is as you suspected,  the very poor mechanical performance of your Dec axis is causing your imaging problems.  Here’s a typical sequence of guiding showing only the Dec in green:

 

 

What you see here is a long delay before the axis reverses direction followed by an immediate over-correction in the opposite direction.  This often implies two problems: a large backlash plus stiction (static resistance) on the axis.  We can’t really distinguish between them, the whole sequence might be caused by just binding and stiffness of the Dec axis.  But you can see the very bad effects this will have on your imaging – each of these cycles is letting the guide star wander around by as much as 15 arc-sec.  Unfortunately, many other users have encountered these sorts of problems with brand-new iOptron mounts.  As a first step, perhaps you should go back to iOptron and ask them to replace the mount or help you fix it.  If that doesn’t work out, you can probably do some reasonable imaging by guiding in only one direction for Dec.  It looks like you might have tried that at one point but got the direction wrong.  There’s a fairly length discussion in the Help guide for how to do uni-directional Dec guiding.  But if it’s a new mount, I would start by asking the manufacturer to correct the problem.

 

Hope this helps,

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.
To view this discussion on the web visit https://groups.google.com/d/msgid/open-phd-guiding/eaeee1f9-8357-4d04-87a9-c1e822281a32%40googlegroups.com.

image002.jpg

Mine Be

unread,
Apr 22, 2020, 12:32:01 PM4/22/20
to Open PHD Guiding
I did this, usually on my 1:10 focuser the ideal point is at 25mm, so I balance while it's at this point. I also have everything already attached then.

Mine Be

unread,
Apr 22, 2020, 12:37:10 PM4/22/20
to Open PHD Guiding
Hey,
thanks for your answer!
Isolating the Dec axis in the graph makes it look horrendous sadly...I'm from Germany, so i tried to talk to my vendor (teleskop-express.de) today where I ordered the mount, they suggested a broken USB/Serial cable which I don't believe as I did my fair share of research to order one that's supported under Windows 10 and is durable and high quality, some other amateur astronomers use the same one I ordered, I don't have any driver problems or random disconnects either. I tried to explain what I suspect but they waved off, saying the mount does track normally (which it does) so there can't be anything wrong. Sending to iOptron from Germany isn't an option for me due to high shipping costs. I'm aware there's a PDF file from iOptron Technical services (https://drive.google.com/file/d/1-wSYganf13_zmSb7mHK5TUUqCZalP5n8/view) which shows how you can tighten the belt, I'm gonna check the belt tension now, and try to do it myself, fingers crossed.
Thanks for your very indepth and good explanation to what's happening with my mount.

Actually I tried to either guide north only and south only, but the mount stopped in both cases to give any guiding impulses and the star just wanders off, I think you can see it in one of the later guiding attempts in my logs.

To unsubscribe from this group and stop receiving emails from it, send an email to open-phd...@googlegroups.com.

bw_msgboard

unread,
Apr 22, 2020, 1:58:06 PM4/22/20
to open-phd...@googlegroups.com

As you know, the response from Teleskop-Express is just wrong.  The mount *does* respond (eventually) to guide commands in all 4 directions so there’s nothing wrong with the USB cable.  And of course no errors are reported when PHD2 communicates with the mount through the night.  Maybe they should stare at the graph I sent you and re-think their response.

 

The basic approach to uni-directional guiding goes like this:

  1. Don’t hyper-tune the polar alignment – alignment error in the 5-10 arc-min range will make things easier for you.  It will also be easier if most of your polar alignment error is in altitude.
  2. After calibration, run the Guiding Assistant for a long enough time that you can clearly see the direction of drift *in Declination*.  If the direction is “up” on the graph, you want to do *south* guiding and vice-versa.
  3. Expect that you will need to switch direction at some point in the sky.  If most of your polar alignment error is in altitude, that will happen sometime around when you do a meridian flip.
  4. You can use either Resist-Switch or LowPass2 algorithms for Dec but don’t be overly aggressive with them.  The guide star may never quite reach the starting point, but that doesn’t matter.  What you care about is not getting reversals in direction and getting a Dec RMS value that’s reasonably close to RA.  On most mounts, that isn’t hard to do.

 

If the mount is only a few months old, I think you should demand a refund from the manufacturer but that’s just me.

 

Good luck,

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/c1bd1703-00bf-466c-9c43-f3a75dff0470%40googlegroups.com.

Duncan Thomas

unread,
Apr 22, 2020, 2:48:13 PM4/22/20
to open-phd...@googlegroups.com
In case you ever have to ship your mount to the US, check with Darkframe UK on their shipping methods.  When I check directly with regular package carriers in the US (Post Office, FedEx, USPS) the rates to the UK are crazy, and that is just one-way.  However, Darkframe uses a carrier that has much more reasonable prices (I think less than $300 US dollars round-trip), and I don’t see why they would not share their source with you. Hopefully you won’t have to go that route though. Good luck.

On Apr 22, 2020, at 10:58 AM, bw_msgboard <bw_m...@earthlink.net> wrote:

As you know, the response from Teleskop-Express is just wrong.  The mount *does* respond (eventually) to guide commands in all 4 directions so there’s nothing wrong with the USB cable.  And of course no errors are reported when PHD2 communicates with the mount through the night.  Maybe they should stare at the graph I sent you and re-think their response.
 
The basic approach to uni-directional guiding goes like this:
  1. Don’t hyper-tune the polar alignment – alignment error in the 5-10 arc-min range will make things easier for you.  It will also be easier if most of your polar alignment error is in altitude.
  2. After calibration, run the Guiding Assistant for a long enough time that you can clearly see the direction of drift *in Declination*.  If the direction is “up” on the graph, you want to do *south* guiding and vice-versa.
  3. Expect that you will need to switch direction at some point in the sky.  If most of your polar alignment error is in altitude, that will happen sometime around when you do a meridian flip.
  4. You can use either Resist-Switch or LowPass2 algorithms for Dec but don’t be overly aggressive with them.  The guide star may never quite reach the starting point, but that doesn’t matter.  What you care about is not getting reversals in direction and getting a Dec RMS value that’s reasonably close to RA.  On most mounts, that isn’t hard to do. 
 
If the mount is only a few months old, I think you should demand a refund from the manufacturer but that’s just me.
 
Good luck,
Bruce

From: 'Mine Be' via Open PHD Guiding [mailto:open-phd...@googlegroups.com] 
Sent: Wednesday, April 22, 2020 9:37 AM
To: Open PHD Guiding
Subject: Re: [open-phd-guiding] Bad Performance in DEC while Guiding, iOptron CEM25P
 
Hey,
thanks for your answer!
Isolating the Dec axis in the graph makes it look horrendous sadly...I'm from Germany, so i tried to talk to my vendor (teleskop-express.de) today where I ordered the mount, they suggested a broken USB/Serial cable which I don't believe as I did my fair share of research to order one that's supported under Windows 10 and is durable and high quality, some other amateur astronomers use the same one I ordered, I don't have any driver problems or random disconnects either. I tried to explain what I suspect but they waved off, saying the mount does track normally (which it does) so there can't be anything wrong. Sending to iOptron from Germanyisn't an option for me due to high shipping costs. I'm aware there's a PDF file from iOptron Technical services (https://drive.google.com/file/d/1-wSYganf13_zmSb7mHK5TUUqCZalP5n8/view) which shows how you can tighten the belt, I'm gonna check the belt tension now, and try to do it myself, fingers crossed.

Mine Be

unread,
Apr 23, 2020, 7:03:17 AM4/23/20
to Open PHD Guiding
Hey, thanks for your answer. I opened up my mount yesterday in the afternoon and tightened the belt in the Declination axis which was very loose. That fixed it for now, I'm down to 2700ms backlash according to the guiding assistant which could be compensated by turning on backlash compensation with the recommended values, I was hovering between 0.5-0.8" in both Axis last night, so finally the mount performs as it should. I went with resist switch for now, which worked great, turned the aggression down to 80 I think. Thank you for explaining unidirectional guiding to me, very helpful just in case it gets worse, but last night was really successful.

I couldn't return the mount as it's already 3 months old, and the vendor doesn't offer refunds in that case as he said the mount is working and it's likely the cable. That angered me quite a bit and I went with the belt tightening procedure.

All subs were great from last night, except the last 20 because a tree got in the telescopes way, but even then, guiding was still okay as the guiding cam apparently got a glimpse through the leaves, I'm really impressed.
PHD2_GuideLog_2020-04-22_162124.txt

Mine Be

unread,
Apr 23, 2020, 7:11:56 AM4/23/20
to Open PHD Guiding
For the first three calibration attempts in the log, something was wrong with my computers USB port or the cable, as the mount somehow didn't move, but I just closed the app and disconnected and reconnected mount and camera again, then it worked.

bw_msgboard

unread,
Apr 23, 2020, 12:31:18 PM4/23/20
to open-phd...@googlegroups.com

Hi Mine, this is really good news, nice job!  Obviously there’s still a lot of reversal delay in the axis but the PHD2 Dec backlash compensation seems to be controlling that quite well.  You can probably leave well enough alone at this point and just do some imaging.  But one thing you could consider doing is to increase the guide speed setting in the mount from the current 0.5x sidereal to a value closer to 1x.  That could further reduce the reversal delay by a proportional amount.  If you do that, just be sure to then repeat both the calibration and the GA backlash test.  In any case, the backlash compensation algorithm has a lot of work to do with your mount, so you may find that it takes a little time for the Dec guiding to stabilize after you do a slew.  So you might want to keep an eye on it and not start an image until you can see there are no oscillations in Dec.

 

Thanks for reporting back!

Bruce

 


From: 'Mine Be' via Open PHD Guiding [mailto:open-phd...@googlegroups.com]
Sent: Thursday, April 23, 2020 4:03 AM
To: Open PHD Guiding
Subject: Re: [open-phd-guiding] Bad Performance in DEC while Guiding, iOptron CEM25P

 

Hey, thanks for your answer. I opened up my mount yesterday in the afternoon and tightened the belt in the Declination axis which was very loose. That fixed it for now, I'm down to 2700ms backlash according to the guiding assistant which could be compensated by turning on backlash compensation with the recommended values, I was hovering between 0.5-0.8" in both Axis last night, so finally the mount performs as it should. I went with resist switch for now, which worked great, turned the aggression down to 80 I think. Thank you for explaining unidirectional guiding to me, very helpful just in case it gets worse, but last night was really successful.

 

I couldn't return the mount as it's already 3 months old, and the vendor doesn't offer refunds in that case as he said the mount is working and it's likely the cable. That angered me quite a bit and I went with the belt tightening procedure.

 

All subs were great from last night, except the last 20 because a tree got in the telescopes way, but even then, guiding was still okay as the guiding cam apparently got a glimpse through the leaves, I'm really impressed.


Am Mittwoch, 22. April 2020 19:58:06 UTC+2 schrieb bw_msgboard:

As you know, the response from Teleskop-Express is just wrong.  The mount *does* respond (eventually) to guide commands in all 4 directions so there’s nothing wrong with the USB cable.  And of course no errors are reported when PHD2 communicates with the mount through the night.  Maybe they should stare at the graph I sent you and re-think their response.

 

The basic approach to uni-directional guiding goes like this:

1.      Don’t hyper-tune the polar alignment – alignment error in the 5-10 arc-min range will make things easier for you.  It will also be easier if most of your polar alignment error is in altitude.

2.      After calibration, run the Guiding Assistant for a long enough time that you can clearly see the direction of drift *in Declination*.  If the direction is “up” on the graph, you want to do *south* guiding and vice-versa.

3.      Expect that you will need to switch direction at some point in the sky.  If most of your polar alignment error is in altitude, that will happen sometime around when you do a meridian flip.

4.      You can use either Resist-Switch or LowPass2 algorithms for Dec but don’t be overly aggressive with them.  The guide star may never quite reach the starting point, but that doesn’t matter.  What you care about is not getting reversals in direction and getting a Dec RMS value that’s reasonably close to RA.  On most mounts, that isn’t hard to do.

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/0c40950e-8788-43b4-a69b-cddd46192da8%40googlegroups.com.

Mine Be

unread,
Apr 24, 2020, 10:45:51 AM4/24/20
to Open PHD Guiding
Hey Bruce,
I'm gonna try to change to guiding speed soon, and see if it reduces the amount of backlash after calibrating in the GA test.
Yeah, I realised that, the first three pics weren't good, but after that, guiding stabilized and not a single bad shot due to egg shaped stars.


Thanks again!
Reply all
Reply to author
Forward
0 new messages