Saw tooth guiding after Dithering - AP900
Anis Abdul
Anis Abdul
bw_msgboard
Hi Anis. Please post your guide logs as attachments or via DropBox links. We can’t do anything with screen snapshots or log excerpts. The guide logs aren’t very big so there should be no difficulty.
http://openphdguiding.org/getting-help/
Thanks.
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.
bw_msgboard
Oops, my bad – I didn’t see that you’d posted a link to the file at the bottom of your earlier message. We’ll take a look…
Bruce
bw_msgboard
Hi Anis. This looks very much like you have declination backlash compensation set in the mount firmware (not in PHD2, in the mount). If you don’t have a keypad, you need to use the AP ASCOM driver interface to be sure both RA and Dec backlash compensation are set to zero. This is very important, you won’t be able to guide effectively otherwise. Your Dec instability is triggered by a reversal in direction, which is, in turn, caused by a dither in Dec.
Your first calibration has no pixel size or focal length while the later ones do. Did you run the new-profile-wizard to build this configuration? You need to do that. It also looks like you also aren’t using either a dark library or a bad-pix map, which we know are important for the LodeStar guiders. And you did your calibration at a Dec value of 59 degrees, which is not recommended and, in your case, not necessary. So I think you have some homework to do. <g> Please read the Help manual, at least the Basic Use section, and follow the advice given there. Most important, create a new profile using the new-profile-wizard and don’t monkey around with any of the settings. If you do that and also eliminate the backlash compensation in the mount, you should do much better.
If you still see problems, let us know and we’ll take the next steps. If the mount was shipped to you, you may need to check the gear mesh in Dec because that can be affected by shipping. There’s a procedure on the AP web site for doing that.
Good luck,
Bruce
From:
open-phd...@googlegroups.com [mailto:open-phd...@googlegroups.com] On Behalf Of Anis Abdul
Sent: Thursday, September 22, 2016
1:18 PM
To: Open PHD Guiding
Subject: [open-phd-guiding] Re:
Saw tooth guiding after Dithering - AP900
I have attached the PHD2 guidelog as Bruce suggested.
--
Anis Abdul
bw_msgboard
Hi Anis. Of course it’s cloudy – you’ve got a new mount, there’s something you really want to check out, what else could you expect? <lol>
While you’re waiting for clear skies I’ll give you a few other things to think about. If you have a dark library for the guide camera you’re using now, you can import that into your new profile. As long as it’s the same camera with the same sensor size, you can re-use it. Once you’re connected to the new profile, just use the Darks/Import from Profile… menu item. The combo boxes there will show you the other profiles with compatible dark libs/bad-pix maps. Choose the one you want and you’re good to go.
The reason I was urging you to build a new profile with the wizard rather than just adding the pixel-size and focal length is that you may have a set of guiding parameters that aren’t really optimized for your set-up. If you upgrade to the 2.6.2 release, you can accomplish the same thing by using the new ‘reset’ parameters on the algorithms tab of the Advanced Dialog.
I’m fairly optimistic that the Dec backlash compensation in the mount is the problem. To reset it, you’d use this panel in the AP ASCOM driver. The AP driver might show zeroes in these fields already, but you should make sure they are zeroes and then click both ‘set’ buttons.
I did notice that the guiding wasn’t correcting in Dec the way I would expect even before the instability started. I suspect that’s because of polar alignment error. If you were just trying to get going with the mount to see what it would do, maybe you didn’t spend much time doing polar alignment.
One of the things we usually recommend to people who are working with a new mount is to spend 15-20 minutes collecting some data to really understand what the mount is doing. You would do that by running the Guiding Assistant for 15-20 minutes and then looking at the results. You’ll see what sort of residual periodic error you have, how much Dec backlash, how much drift on both axes (including polar alignment error), and you’ll also get a recommendation for tuning the min-move parameters in the guide algorithms. It’s not usually the first thing you’d want to do but it can really save you a lot of time in the long run, and it makes it easier for us to offer advice. If you collect the data, we’ll be happy to help you interpret it.
Hope it clears up soon,
Bruce
Anis Abdul
Anis Abdul
bw_msgboard
Hi Anis. Sounds like you make some good progress. Could you also post your debug log file so we can see what recommendations were made by the Guiding Assistant? You may have misinterpreted the guide speed setting behavior although it’s not really important in this case. PHD2 doesn’t really care about the guide speed in terms of actual guiding, it’s mostly used to calculate the calibration step-size. What did you see in the PHD2 UI that made you think it was set to 0.5X? Were you not connected to the mount at that point? Astro-Physics recommends using the 1X sidereal speed for guiding so there’s probably not reason at this point to change that. Either one should work fine.
When we get your debug log file, we’ll take a look and help you understand what you saw. For example, polar misalignment will cause drift in Dec, not usually RA. If the debug log is large, you’ll probably need to post it via cloud-link.
Anis Abdul
Hi Bruce
Andy Galasso
Anis Abdul
bw_msgboard
Hi Anis. Things are definitely looking up here. You had a polar alignment error of about 9 arc-min, so you can probably improve on that the next time out. The uncorrected periodic error looks like it’s around 5 arc-sec peak-to-peak, well within the normal range for AP mounts. A good periodic error correction will probably reduce this substantially. You might try enabled PEC to use whatever was programmed by the previous owner. You never know, it might be a significant improvement until you can do the PEC yourself. And if it makes things worse, just turn it off, you haven’t lost anything.
The Guiding Assistant failure to measure backlash is kind of an obscure bug, something we’ll fix in the next release. It happened because you started the Guiding Assistant without having a calibration as part of the new profile you built. So the next time out, you should:
- Reset the guide speed in the mount to 1X sidereal. This will speed up your Dec backlash recovery by ½, which is one of the reasons AP recommends the higher speed. We can see there is some amount of Dec backlash in your system, about 1 second at the 0.5x sidereal rate. If you restore the guide speed to 1X, the Dec backlash recovery time will decrease proportionately, and you should be able to use the Dec backlash compensation feature in PHD2 if you want.
- Get a new calibration somewhere near Dec=0 and within an hour of the central meridian. Try to avoiding doing it at high Dec values, where you did it this last time.
- Start guiding
- Run the Guiding Assistant for 3-4 minutes. There’s no need to run it for the long period because we can already see the RA periodic error.
- Let the Guiding Assistant measure the Dec backlash, which should work fine if you follow this sequence
- Refine the polar alignment if necessary and apply the recommendations made by the Guiding Assistant
- Resume guiding and see what you get over a 10-15 minute period
All the evidence I can see suggests the mount is in fine shape and you should get good results. If you want us to look at the data, please post both the guide and debug log files.
Anis Abdul
Hi Anis. Things are definitely looking up here. You had a polar alignment error of about 9 arc-min, so you can probably improve on that the next time out. The uncorrected periodic error looks like it’s around 5 arc-sec peak-to-peak, well within the normal range for AP mounts. A good periodic error correction will probably reduce this substantially. You might try enabled PEC to use whatever was programmed by the previous owner. You never know, it might be a significant improvement until you can do the PEC yourself. And if it makes things worse, just turn it off, you haven’t lost anything.
The Guiding Assistant failure to measure backlash is kind of an obscure bug, something we’ll fix in the next release. It happened because you started the Guiding Assistant without having a calibration as part of the new profile you built. So the next time out, you should:
- Reset the guide speed in the mount to 1X sidereal. This will speed up your Dec backlash recovery by ½, which is one of the reasons AP recommends the higher speed. We can see there is some amount of Dec backlash in your system, about 1 second at the 0.5x sidereal rate. If you restore the guide speed to 1X, the Dec backlash recovery time will decrease proportionately, and you should be able to use the Dec backlash compensation feature in PHD2 if you want.
- Get a new calibration somewhere near Dec=0 and within an hour of the central meridian. Try to avoiding doing it at high Dec values, where you did it this last time.
- Start guiding
- Run the Guiding Assistant for 3-4 minutes. There’s no need to run it for the long period because we can already see the RA periodic error.
- Let the Guiding Assistant measure the Dec backlash, which should work fine if you follow this sequence
- Refine the polar alignment if necessary and apply the recommendations made by the Guiding Assistant
- Resume guiding and see what you get over a 10-15 minute period
All the evidence I can see suggests the mount is in fine shape and you should get good results. If you want us to look at the data, please post both the guide and debug log files.
Good luck,
Bruce
To unsubscribe from this group and stop receiving emails from it, send an email to open-phd-guiding+unsubscribe@googlegroups.com.
bw_msgboard
From: Anis
Abdul [mailto:ani...@gmail.com]
Sent: Saturday, September 24, 2016
5:22 PM
To: Bruce Waddington
Cc: Open PHD Guiding
Subject: Re: [open-phd-guiding]
Re: Saw tooth guiding after Dithering - AP900
Hi Bruce
I definitely will perform these changes and I will be sure to attach the debug and log file for analysis. The PA definitely got worse due to my action I think. The pier feet are on smooth surface and that is a mistake. I will switch it out with bricks and the PA should stay put after calibration.
I wasnt sure if PEC will be valid since I changed the chip in the CP3 box but may be it is. I will be sure to test it. Last night The sky was quite murky and I couldnt get any stars near the equator so I had to switch to the milkyway area to get some stars to show up.
Good point, I forgot you did a chip upgrade so you probably shouldn’t do this. Probably better to leave it alone until you can program it yourself.
Thank you again for your help. I really appreciate it.
No problem – let us know how you do.
Bruce
Anis
On Sat, Sep 24, 2016 at 4:02 PM, bw_msgboard <bw_m...@earthlink.net> wrote:
Hi Anis. Things are definitely looking up here. You had a polar alignment error of about 9 arc-min, so you can probably improve on that the next time out. The uncorrected periodic error looks like it’s around 5 arc-sec peak-to-peak, well within the normal range for AP mounts. A good periodic error correction will probably reduce this substantially. You might try enabled PEC to use whatever was programmed by the previous owner. You never know, it might be a significant improvement until you can do the PEC yourself. And if it makes things worse, just turn it off, you haven’t lost anything.
The Guiding Assistant failure to measure backlash is kind of an obscure bug, something we’ll fix in the next release. It happened because you started the Guiding Assistant without having a calibration as part of the new profile you built. So the next time out, you should:
1. Reset the guide speed in the mount to 1X sidereal. This will speed up your Dec backlash recovery by ½, which is one of the reasons AP recommends the higher speed. We can see there is some amount of Dec backlash in your system, about 1 second at the 0.5x sidereal rate. If you restore the guide speed to 1X, the Dec backlash recovery time will decrease proportionately, and you should be able to use the Dec backlash compensation feature in PHD2 if you want.
2. Get a new calibration somewhere near Dec=0 and within an hour of the central meridian. Try to avoiding doing it at high Dec values, where you did it this last time.
3. Start guiding
4. Run the Guiding Assistant for 3-4 minutes. There’s no need to run it for the long period because we can already see the RA periodic error.
5. Let the Guiding Assistant measure the Dec backlash, which should work fine if you follow this sequence
6. Refine the polar alignment if necessary and apply the recommendations made by the Guiding Assistant
7. Resume guiding and see what you get over a 10-15 minute period
Anis Abdul
From: open-phd...@googlegroups.com [mailto:open-phd...@googlegroups.com] On Behalf Of Anis Abdul
bw_msgboard
You didn’t get the guide speed in the mount changed. Don’t get confused here, you don’t do that by changing a setting in PHD2. You have to do it from the AP ASCOM driver. PHD2 never sets the guide speed, it only reads it from the mount.
Anis Abdul
bw_msgboard
Ok, but it didn’t seem to work. Both the calibration and the subsequent guiding show the mount’s guide speed set at 7.5 arc-sec/sec, which is 0.5X sidereal. Maybe you didn’t actually get it set in the ASCOM driver interface. I don’t use that driver and I don’t remember exactly how it’s done. I’m referring to the calibration done at 21:25:03 – any possibility that you changed the guide speed setting after that point in time?
Anis Abdul
bw_msgboard
Nice work tracking this down. The “RG2” parameter sets the guide speed to 1X sidereal, so you should be good to go. So for whatever reason, it didn’t get done in the first try. Yes, you need to re-do the calibration, for sure. And if you want to get a better measure of your backlash, you should re-do the GA as well…
Anis Abdul
bw_msgboard
Sounds good, glad to hear you’re happy with the results. That’s not surprising, really, I think the underlying problem was really just the bogus backlash compensation setting in the mount. If you decide to make further measurements down the road and want some help in evaluating the results, let us know.
Good luck,
Bruce
Anis Abdul
bw_msgboard
Hi Anis. Your results look good, and congrats on getting the PEC dialed in so well. A residual peak-to-peak PE of 1.9 a-s should be easy to guide out, so you should be good to go. I do recommend applying all the recommendations made by the GA, that’s the best advice we have. <g> You can keep an eye on the backlash compensation setting to be sure you don’t get instability in Dec guiding, but a comp setting of 1.5 sec should work ok. If you do see problems, you can disable the Dec backlash compensation in the Guiding tab of the brain dialog, no need to re-run the GA. By the way, you *are* now guiding at 1X sidereal, so that issue is also behind you.
I think from this point forward, you can just keep an eye on the guiding and imaging results, but I expect you’ll be a happy camper. So get some good images!
Have fun,
Andy Galasso
You can keep an eye on the backlash compensation setting to be sure you don’t get instability in Dec guiding, but a comp setting of 1.5 sec should work ok. If you do see problems, you can disable the Dec backlash compensation in the Guiding tab of the brain dialog, no need to re-run the GA.