PHD2 connection to Celestron and calibration issues.
Ed Bianchina
I have been using PHD2 for quite a while now and it does a great job most of the time. There are a couple of issues. First my equipment. Orion Starshooter guide camera w/ST4 port, Celestron CGE mount, NexRemote control software. I use the ST4 guiding option. Currently I am using an 8" Astro-Tech Newt as a guide scope (my normal guide scope is a wide field refractor but I am taking some pics with it right now).
Problems:
1) There are times that the DEC pulse will not move the mount. I can manually move the mount with the mount software but PHD2 will not do it. I think if I shut everything down and reconnect it works. The reason I said think is that it work twice failed once. I let the program choose the pulse time (200 ms) from the calculation. Dec backlash is present but not to an extreme. The calibration works sometimes but not other. I ALWAYS approach a target from the west and the north so it should at least be consistent.
2) I use the ST4 method to guide but I would really like to go through the NexRemote software so everything is automatic (DEC correction, flip, etc). How would I do that. I know Celestron control is a little wonky!
3) I typically use 1 sec exposure to guide but I think that is too fast. Is there a way to determine the correct exposure time to use to get the best results? I can usually get a total error of less then 1" but last night is was all over the place and the error was about 2 with some really large excursions. Is that seeing popping its ugly head?
Thanks in advance.
Ed
John Stiner
John Stiner
Ed Bianchina
On Monday, November 14, 2016 at 3:54:08 PM UTC-6, John Stiner wrote:
Ed, are you using the On-Camera setting on PHD or are you routing a USB/serial cable from a USB port on your laptop to the bottom of your hand controller? If the former, you'll need that USB/serial connection through your HC. You'll likely also need to assign a virtual port on the NexRemote and then have PHD run through that virtual port. I'd get NexRemote working but couldn't get PHD to fire up (or vice Versace) because they were trying to use the same COM port. That's where the virtual port on NexRemote comes in and lets you do both.
I am using the on camera setting. I use a USB to serial to run from the computer to mount. I do not go through the hand set. There is no problem with the virtual port I just don't know how to hook it.
bw_msgboard
Hi Ed. As I mentioned on the other forum, we need to see matching guide and debug log files. Over here, we don’t like to just stab in the dark and make guesses about the problems people are having. <g> Why don’t you start with logs that will show a calibration failure and we can take it from there. FWIW, slewing *from* the north before calibration is the wrong thing to do. To clear backlash, you want to finish by having the scope moving north, which is the direction PHD2 will move it during calibration.
As far as getting an ASCOM connection going, take a look at the post from Peter Wolsley on this thread:
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.
Kevin S
Andy Galasso
3) I typically use 1 sec exposure to guide but I think that is too fast. Is there a way to determine the correct exposure time to use to get the best results? I can usually get a total error of less then 1" but last night is was all over the place and the error was about 2 with some really large excursions. Is that seeing popping its ugly head?
John Stiner
Bryan
Ed Bianchina
Bruce
Thanks for the help. Attached is the log file. I have no idea how to read it but honestly I haven't spent a ton of time trying to decipher it. So last night I was able to connect through NexRemote and everything looked like it was working. When I went to guide, I had the same issues as usual. RA calibration was good but the DEC didn't want to calibrate well. I hade to increase the calibration step do get it to move at all. When it did calibrate I was able to guide but the Dec was pretty noisy and the RA was horrible.
ANY help is greatly appreciated. Anything else you might need to help I can get. I really want to learn how to troubleshoot so thanks to all the help I can get!!!
Ed
bw_msgboard
Hi Ed. I don’t think any log files got attached. But we need *both* the guide and debug log files from your session last night. They both live in the same directory with the same timestamps embedded in the file names. We need the guide log to analyze the performance and the debug log to understand more about how you are configuring things.
I don’t think many people read the guide log directly. Your best bet is to use the PHDLogViewer tool for that: http://openphdguiding.org/phd2-log-viewer/
To start coming up to speed on what it will show you, take a look at this tutorial:
http://openphdguiding.org/tutorial-analyzing-phd2-guiding-results/
If you can upload both those log files – DropBox works well for big files – I’ll take a look at them. Also, there’s a big section in the Help file that talks about calibration problems and how to deal with them. I’m guessing your problem is likely to be covered. J
http://openphdguiding.org/manual/?section=Trouble_shooting.htm#Problems-_Mount_Control
Good luck,
Bruce
From: open-phd...@googlegroups.com [mailto:open-phd...@googlegroups.com] On Behalf Of Ed Bianchina
Sent: Sunday, November 20, 2016
11:20 AM
To: Open PHD Guiding
--
bw_msgboard
Well, that’s half of what we need… J
Bruce
From:
open-phd...@googlegroups.com [mailto:open-phd...@googlegroups.com] On Behalf Of Ed Bianchina
Sent: Sunday, November 20, 2016
12:19 PM
To: Open PHD Guiding
Subject: [open-phd-guiding] Re:
PHD2 connection to Celestron and calibration issues.
It helps if you actually attache the files! DUH.
--
Ed Bianchina
https://drive.google.com/file/d/0B3FvCADdv6oAVXpjM2RheFFPUDg/view?usp=sharing
Ed
bw_msgboard
Thanks. One question up-front: in your initial post, you said you were using an Astro-Tech 8” scope for the guide scope. Isn’t that an f/4 system with a focal length of 812mm? Or do you have some kind of thread-on focal reducer in the guider optical train? I just don’t want to waste time here if the specified focal length of 558mm is wrong.
Bruce
From:
open-phd...@googlegroups.com [mailto:open-phd...@googlegroups.com] On Behalf Of Ed Bianchina
Sent: Sunday, November 20, 2016
12:27 PM
To: Open PHD Guiding
Subject: [open-phd-guiding] Re:
PHD2 connection to Celestron and calibration issues.
Here is the debug file.
https://drive.google.com/file/d/0B3FvCADdv6oAVXpjM2RheFFPUDg/view?usp=sharing
Ed
Ed Bianchina
bw_msgboard
Hi Ed. I adjusted the settings in the log file so we could use the correct image scale.
It looks like you have something above 5 seconds of backlash in Dec, which is probably partly responsible for your calibration issues. Of course, the incorrect focal length also meant the computed calibration step-size was too small by 30% or so. To get a better calibration, move the mount manually at guide speed NORTH for 15-20 seconds before starting your calibration. You can do that with the hand controller or the PHD2 manual guide tool, whichever you prefer.
With respect to the guiding performance, it’s a little hard to tell because you’ve created a complete hash for guiding parameters. These are remarkably bad J As we said in some of the reference documents, it’s better to leave these things alone unless you have a clear understanding of what they do and why you might need to change them. As you saw, the RA guiding performance is 50-100% worse than the Dec guiding, so that’s the thing to focus on. The bogus parameter settings tended to make that worse, so I can’t tell what the baseline tracking performance really is. It does look like you are getting a lot of high-frequency motion in RA which is something I don’t expect to disappear by just fixing the guiding parameters. Using 2-second exposures will probably help with that but I expect this will still be a limiting condition. But we need to get a clean measurement of it before we can try to mitigate it. High-frequency movement like this is not something that can be guided out, it will need to be improved or eliminated at the source if possible.
To make any headway, I think you need to start over:
1. Use the new-profile-wizard to create a new equipment profile for the test, being absolutely sure the guide scope focal length and camera pixel size are correct. Those should be 810mm and 5.2u, and the guide speed setting in your mount is currently 0.5X.
3. Slew to a good guide star near Dec=0 and within an hour of the central meridian. Set the exposure amount to 2 seconds, nothing less.
4. Manually move the scope NORTH at guide speed for 15-20 seconds, then start the calibration. Let PHD2 auto-select the guide star in the field, and don’t monkey around with any settings during calibration. If the guide star still doesn’t move enough, it’s ok to increase the calibration step-size, but I doubt you’ll have to do that once the correct equipment parameters are in place.
5. Run the Guiding Assistant for 10-15 minutes and apply whatever recommendations it makes – particularly with respect to min-moves. Let it measure your Dec backlash. You may need to use a large tracking region for this test in case the mount tracking causes a loss of the guide star – just be sure there aren’t multiple stars in the tracking rectangle.
6. Guide on a suitable guide star for 10-15 minutes, just letting it run so long as there aren’t gross errors from wind or other “mistakes.”
7. Do NOT change any of the guiding parameters while the test is being done.
One this is all done, package up both the guide and debug log files, and we’ll take another look.
Good luck,
Bruce
Ed Bianchina
Ed
bw_msgboard
Hi Ed. It’s good to see you’re making some progress. Here’s the next set of questions and comments:
- Your calibrations are still triggering alerts because the Dec movement is not behaving as expected. I wonder if you really did what I suggested: did you move the scope manually north for 20 seconds before starting *each* calibration. If you tried to do this, how did you go about it? You will need to do this every time you do a calibration, but that shouldn’t happen very often if you can get a good one. The point of doing the manual moves is to clear most of the Dec backlash in the mount. OTOH, if you did all this and still have the calibration alerts, we may be looking at a different problem (there is some hint of “cogging” during the Dec calibration).
- The Guiding Assistant run shows you have about 10 seconds of Dec backlash in the mount. If you increase the guide speed setting in the mount to 1x sidereal, that will be reduced to something more like 5 seconds. This is not a giant amount of backlash, much less than we see in many of these mounts, so it would be worth trying.
- The GA also shows you have an uncorrected periodic error of about 25-30 arc-secs, which is again not bad for these mounts. You should apply periodic error correction (PEC) to the mount which will probably reduce it significantly – assuming this version of the mount supports periodic error correction.
These are all fairly easy problems to fix. Moving ahead to the 14 min guiding run, we see kind of a mixed bag. It’s definitely better than what you got before. I attribute that to using longer exposures and getting the guiding parameters back to something reasonable. But the RA tracking is still 2x worse than Dec. Here are two graphs showing the behavior on the two axes – this is with *no* guiding, just letting the mount run with guiding disabled.
Now, the Dec motor isn’t running at all for this sequence, so the movements we see here are mostly due to seeing. You can see the Dec motion stays in a region of about +/- 0.5 arc-sec. But the RA tracking is much worse with an overall envelope of about +/- 2 arc-secs. What’s worse, the fluctuations happen rapidly, so it’s not practical to guide these out. I think this is the limiting condition for your guiding, and it shows up again during the actual guiding sequence starting at 22:02. What’s causing this? It could be coming from roughness in the gears or some kind of vibration in the RA drive train. Some of it might be coming from local heat convection depending on your particular set-up and where you were pointing in the sky. Unfortunately, there’s not much I can do to help you track this down, but this is not a guiding problem and not something that can be fixed by tweaking guiding parameters.
At some point, you’ll have to judge what is “good enough” to support your imaging goals. You may decide that what you have is sufficient for what you’re trying to accomplish. If not, you’ll probably have to figure out what’s causing the mechanical problems on the RA axis. You may be able to get other ideas from people who are experts with the Celestron mounts, which appear to have a considerable variety of problems.
Ed Bianchina
On Monday, November 21, 2016 at 11:55:11 PM UTC-6, Bruce Waddington wrote:
Hi Ed. It’s good to see you’re making some progress. Here’s the next set of questions and comments:
- Your calibrations are still triggering alerts because the Dec movement is not behaving as expected. I wonder if you really did what I suggested: did you move the scope manually north for 20 seconds before starting *each* calibration.Now that I think of it not on everyone. I did on the last one I but it was not for 20 seconds but rather enough to take the backlash out of the mount as signified by star movement. I did this by moving the mount manually using the up arrow until the star mover several diameters.. If you tried to do this, how did you go about it? You will need to do this every time you do a calibration, but that shouldn’t happen very often if you can get a good one. The point of doing the manual moves is to clear most of the Dec backlash in the mount. OTOH, if you did all this and still have the calibration alerts, we may be looking at a different problem (there is some hint of “cogging” during the Dec calibration). I found it odd that the angle between the DEC and RA axis was between 9 and 12 degrees off? Not sure what to make of that.
- The Guiding Assistant run shows you have about 10 seconds of Dec backlash in the mount. If you increase the guide speed setting in the mount to 1x sidereal, that will be reduced to something more like 5 seconds. This is not a giant amount of backlash, much less than we see in many of these mounts, so it would be worth trying. There is a procedure for reducing the backlash on the DEC axis for the mount and it is pretty minimal so I will give that a try when I have a chance. It involves removing the telescope and weight so is a pain to do.
- The GA also shows you have an uncorrected periodic error of about 25-30 arc-secs, which is again not bad for these mounts. You should apply periodic error correction (PEC) to the mount which will probably reduce it significantly – assuming this version of the mount supports periodic error correction. I didn't realize the PEC would help on a guided set up. I will give that a try as well.
These are all fairly easy problems to fix. Moving ahead to the 14 min guiding run, we see kind of a mixed bag. It’s definitely better than what you got before. I attribute that to using longer exposures and getting the guiding parameters back to something reasonable. But the RA tracking is still 2x worse than Dec. Here are two graphs showing the behavior on the two axes – this is with *no* guiding, just letting the mount run with guiding disabled.
Now, the Dec motor isn’t running at all for this sequence, so the movements we see here are mostly due to seeing. You can see the Dec motion stays in a region of about +/- 0.5 arc-sec. But the RA tracking is much worse with an overall envelope of about +/- 2 arc-secs. What’s worse, the fluctuations happen rapidly, so it’s not practical to guide these out. I think this is the limiting condition for your guiding, and it shows up again during the actual guiding sequence starting at 22:02. What’s causing this? It could be coming from roughness in the gears or some kind of vibration in the RA drive train. Some of it might be coming from local heat convection depending on your particular set-up and where you were pointing in the sky. Unfortunately, there’s not much I can do to help you track this down, but this is not a guiding problem and not something that can be fixed by tweaking guiding parameters.
At some point, you’ll have to judge what is “good enough” to support your imaging goals. You may decide that what you have is sufficient for what you’re trying to accomplish. If not, you’ll probably have to figure out what’s causing the mechanical problems on the RA axis. You may be able to get other ideas from people who are experts with the Celestron mounts, which appear to have a considerable variety of problems. Overall the guiding is not a problem in my images. I due up to 5 min subs and that is really limited by my sky pollution. I don't have issues with oblong stars or anything unusual. You have definately helped and given me a direction to head! I will keep you posted. Thanks again. I'll remember you when I am rich and famous! ;)
bw_msgboard
Hi Ed. I’ll just wrap this up with a few final comments:
From: Ed
Bianchina [mailto:stard...@gmail.com]
Sent: Tuesday, November 22, 2016
1:13 PM
To: Open PHD Guiding
Cc: stard...@gmail.com;
bw_m...@earthlink.net
Subject: Re: [open-phd-guiding]
Re: PHD2 connection to Celestron and calibration issues.
Bruce, your response is overwhelming! Thank you so much for helping. Wow is all I can say. I will try to answer the questions below in red but some of them I will have wait until the clouds clear which could be a while (winter in Chicago!). I did have several issues where the computer gave me the blue screen of death when I was working. It would happen in the ASCOM mode but not on the ST4 mode. I reinstalled the software and ran the ASCOM checker and it seemed to work after that.
On Monday, November 21, 2016 at 11:55:11 PM UTC-6, Bruce Waddington wrote:
Hi Ed. It’s good to see you’re making some progress. Here’s the next set of questions and comments:
1. Your calibrations are still triggering alerts because the Dec movement is not behaving as expected. I wonder if you really did what I suggested: did you move the scope manually north for 20 seconds before starting *each* calibration.Now that I think of it not on everyone. I did on the last one I but it was not for 20 seconds but rather enough to take the backlash out of the mount as signified by star movement. I did this by moving the mount manually using the up arrow until the star mover several diameters..
That’s why the last calibration was better. You need to do it each time, and I recommend doing it for the full 20 seconds. You can get fooled by just looking at the star and besides, 20 seconds is not that long a time in the grand scheme of things.
If you tried to do this, how did you go about it? You will need to do this every time you do a calibration, but that shouldn’t happen very often if you can get a good one. The point of doing the manual moves is to clear most of the Dec backlash in the mount. OTOH, if you did all this and still have the calibration alerts, we may be looking at a different problem (there is some hint of “cogging” during the Dec calibration). I found it odd that the angle between the DEC and RA axis was between 9 and 12 degrees off? Not sure what to make of that.
The angle error is just a symptom of other problems. In cases where the Dec backlash isn’t cleared, the star just wanders around in the beginning because of seeing effects rather than getting moved as it should. In your second calibration, you can see kind of a sine wave pattern in the Dec movement. That often happens because of uncorrected periodic error in RA.
2. The Guiding Assistant run shows you have about 10 seconds of Dec backlash in the mount. If you increase the guide speed setting in the mount to 1x sidereal, that will be reduced to something more like 5 seconds. This is not a giant amount of backlash, much less than we see in many of these mounts, so it would be worth trying. There is a procedure for reducing the backlash on the DEC axis for the mount and it is pretty minimal so I will give that a try when I have a chance. It involves removing the telescope and weight so is a pain to do.
It may not be worth the trouble for you. Your guiding limitation is clearly in RA, and the existing Dec backlash is not bad for your mount.
3. The GA also shows you have an uncorrected periodic error of about 25-30 arc-secs, which is again not bad for these mounts. You should apply periodic error correction (PEC) to the mount which will probably reduce it significantly – assuming this version of the mount supports periodic error correction. I didn't realize the PEC would help on a guided set up. I will give that a try as well.
Yes, you should apply PEC. The general rule is to have the guiding software do as little work as possible.
These are all fairly easy problems to fix. Moving ahead to the 14 min guiding run, we see kind of a mixed bag. It’s definitely better than what you got before. I attribute that to using longer exposures and getting the guiding parameters back to something reasonable. But the RA tracking is still 2x worse than Dec. Here are two graphs showing the behavior on the two axes – this is with *no* guiding, just letting the mount run with guiding disabled.
Now, the Dec motor isn’t running at all for this sequence, so the movements we see here are mostly due to seeing. You can see the Dec motion stays in a region of about +/- 0.5 arc-sec. But the RA tracking is much worse with an overall envelope of about +/- 2 arc-secs. What’s worse, the fluctuations happen rapidly, so it’s not practical to guide these out. I think this is the limiting condition for your guiding, and it shows up again during the actual guiding sequence starting at 22:02. What’s causing this? It could be coming from roughness in the gears or some kind of vibration in the RA drive train. Some of it might be coming from local heat convection depending on your particular set-up and where you were pointing in the sky. Unfortunately, there’s not much I can do to help you track this down, but this is not a guiding problem and not something that can be fixed by tweaking guiding parameters.
At some point, you’ll have to judge what is “good enough” to support your imaging goals. You may decide that what you have is sufficient for what you’re trying to accomplish. If not, you’ll probably have to figure out what’s causing the mechanical problems on the RA axis. You may be able to get other ideas from people who are experts with the Celestron mounts, which appear to have a considerable variety of problems. Overall the guiding is not a problem in my images. I due up to 5 min subs and that is really limited by my sky pollution. I don't have issues with oblong stars or anything unusual. You have definately helped and given me a direction to head! I will keep you posted. Thanks again. I'll remember you when I am rich and famous! ;)
Sounds good. If you’re happy enough with your imaging results, I wouldn’t recommend spending a bunch of time trying to work on the guiding. The guiding is simply a means to an end. J
Have fun,
Bruce