Is the relatively poor Dec guiding a consequence of PE in RA?
Geoffrey Lewis
I had a couple of nights imaging ICs 59&63 and overal had pretty good guiding, but my Dec in particular appeared to display a regular pattern similar to a PE curve, but surely PE doesn't apply with the Dec axis, or am I wrong about that. On the 1st night I messed about with Dec algorithms to little avail (and yes, Bruce and Andy, this was an unstructured, try this try that approach). The 2nd night I just let the rig run for 6+ hours and today I checked the guide log using the phdlogview tool, which showed that the Dec and RA curves pretty much followed the same pattern, albeit Dec deviations are higher than RA. With total RMS <1" I'm not complaining, but if I can I'd like to improve on it. I have not applied any PEC, so that is on my to do list, but I'm interest to learn whether the Dec guiding is merely related to PE in RA or indicative of something else that I should investigate please. I have attached the GuideLog from 25th fyi, so please let me know if there is anything I can do to improve the performance, or if I should just accept what I'm getting.
Many thanks,
Geof
bw_msgboard
Hi Geoff. To answer your easy question, no there is no periodic error in Dec. The Dec motor should not be running at all for considerable amounts of time so there’s no “period.” <g> I also don’t think uncorrected PE has anything to do with your Dec guiding. That is something that can happen during calibration but we don’t generally see it happen during guiding. And whatever residual RA periodic error you have is being well-corrected by guiding. Beyond that, your Dec guiding data is a little puzzling. In my experience, you generally expect Dec guiding to be at least as good as RA simply because the motor isn’t running all the time and there’s less to do. This can change as you move closer to the pole because up there variations in RA tracking have much less effect on overall tracking. So in your first session, up at Dec = 65, things look somewhat normal although there’s probably too much direction reversal in Dec. But then you started a second session at Dec = -2 and things got considerably worse. We’d expect the RA tracking to get a little worse, but in your case, it’s the Dec guiding that went bad. I think this is worth pursuing because you should get better results with your set-up.
I don’t think we know how much backlash you have in Dec, but I would expect it to be pretty small with an AP mount. There’s some history in the log to suggest it might be around 1 sec but I don’t know if that’s accurate. If it’s anything like 1 second, it shouldn’t be a problem. My first guess is that you were chasing the seeing with such a low setting of min-move. To start, I’d run the Guiding Assistant again and see what it recommends – I’d include measuring Dec backlash again just to get a good number. With Dec, you really don’t want to chase seeing because it will create excessive direction reversals and whatever backlash you have can become an issue. I don’t know whether the pointing position of Dec = -2 moved you into a zone of very bad seeing or some other local thermal behaviors. As you point out, the oscillation in Dec is something you want to eliminate. So one possibility is that you can reduce this by using a larger value of min-move, and that’s where I would start. You really only want to correct for slow changes in Dec, which should be mostly in one direction.
If that doesn’t help, you’d have to start looking at the mechanics. The graph shows typical sequences that follow a pattern:
1. React to a guide star deflection and issue a bunch of short guide pulses that might total up to 1 second
2. The corrections over-shoot and the guide star moves to the other side of the x-axis.
3. Now issue another sequence of short guide pulses in the opposite direction, again presumably having to clear some backlash
It looks to me like the delays in getting the mount moving are probably just backlash, but there’s always the possibility that something is interfering with the mount’s ability to react more quickly. Have you tinkered around with the mesh on the Dec axis? Any cables or other impediments in the way of Dec motion? Are you working in frigid conditions? It’s probably also worth checking the balance in Dec. AP mounts are very tolerant of load imbalances, but there are limits to everything. One experiment you could try would be to run a guiding session at a Dec value equal to your latitude. That will put the scope pointing straight up with respect to declination so Dec loading should be neutral. But as I said, the place to start is probably by increasing the Dec min-move and trying to keep Dec corrections moving in the same direction.
Good luck,
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.
Geoffrey Lewis
I very much appreciate the fast and detailed response. Thanks for confirming that there is not and cannot be any PE in Dec; that was my understanding, then I looked at the graph using the log viewer and saw that the RA and Dec lines seemed to trace together so I wondered if I was seen Dec mirroring RA patterns.
On the previous night I tried a bunch of different settings, including increasing Dec min move and reducing aggression, but with little impact either better or worse. I did run the guiding assistant without measuring backlash and accepted the recommended change to min move which was to the 10 that I retained throughout the session on the 25th.
I have measured backlash previously and accepted a recommended backlash comp of 1311, but the guiding session was a nightmare with much smoother guiding after I turned it off, so I was confused by that, but left it off ever since. I've attached the GuideLog from that session on 18 Nov to show what I mean. Once again I note that the Dec and RA curves (dx/dy) broadly coincide, so why is this? It is significant, normal, or just coincidence?
I will check my scopes balance as that very well may be something that I should address. I originally balanced everything with the intention of imaging through my C14, but these past few sessions I've been using the 4" TSAPO which piggy-backs on the C14, so the camera was moved from the C14 to the TSAPO and the corrector lens cover was left in place on the C14, undoubtedly making it somewhat nose heavy. Do you think that this would be sufficient to cause these Dec guiding errors?
BTW my latitude is 52.5N so imaging at Dec 61 does put the scope in an almost vertical position as it transits the meridian.
Best regards,
Geof
bw_msgboard
Hi Geoff, comments below.
From:
open-phd...@googlegroups.com [mailto:open-phd...@googlegroups.com] On Behalf Of Geoffrey Lewis
Sent: Sunday, November 27, 2016
12:35 PM
To: Open PHD Guiding
Cc: geof...@hotmail.com;
bw_m...@earthlink.net
Subject: Re: [open-phd-guiding] Is
the relatively poor Dec guiding a consequence of PE in RA?
Hi Bruce,
I very much appreciate the fast and detailed response. Thanks for confirming
that there is not and cannot be any PE in Dec; that was my understanding, then
I looked at the graph using the log viewer and saw that the RA and Dec lines
seemed to trace together so I wondered if I was seen Dec mirroring RA patterns.
On the previous night I tried a bunch of different settings, including
increasing Dec min move and reducing aggression, but with little impact either
better or worse. I did run the guiding assistant without measuring backlash and
accepted the recommended change to min move which was to the 10 that I retained
throughout the session on the 25th.
I have measured backlash previously and accepted a recommended backlash comp of
1311, but the guiding session was a nightmare with much smoother guiding after
I turned it off, so I was confused by that, but left it off ever since. I've
attached the GuideLog from that session on 18 Nov to show what I mean. Once
again I note that the Dec and RA curves (dx/dy) broadly coincide, so why is
this? It is significant, normal, or just coincidence?
The 11/18 log is pretty interesting, I think there’s something not quite right here. We’ve seen this sort of thing before, a situation where a move in Dec causes a corresponding deflection on the RA axis. I think you should run the star-cross test, it’s described in the Help docs. I wonder if you might get something that looks like this – notice the “loops” on the RA axis instead of straight lines:
I will check my scopes balance as that very well may be something that I should
address. I originally balanced everything with the intention of imaging through
my C14, but these past few sessions I've been using the 4" TSAPO which
piggy-backs on the C14, so the camera was moved from the C14 to the TSAPO and
the corrector lens cover was left in place on the C14, undoubtedly making it
somewhat nose heavy. Do you think that this would be sufficient to cause these
Dec guiding errors?
This is something you should straighten out before doing anything else. It should only take a few minutes and it will eliminate the need for guesswork. If you are located at latitude 52N, then the imaging session at -2 Dec must have put the OTA in a very horizontal orientation. I’d expect that to bring out the worst effects of a significant Dec weight imbalance.
Let us know what you discover,
Bruce
Geoffrey Lewis
I'll definitely attend to the balance issue and thanks for the star cross suggestion which is something I've never done, so I'll read up and give it a try. If I do see those looped RA lines, what does that mean?
The brief session at Dec -2 was a test shot on IC434 (Horse Head) which I'm contemplating for another couple of sessions as I'm thinking that should frame well in the with QSI.TSAPO combo and yes at my latitude the scope is pretty level, so maximum torque is being applied across the Dec axis when unbalanced.
Regards,
Geof
bw_msgboard
From:
open-phd...@googlegroups.com [mailto:open-phd...@googlegroups.com] On Behalf Of Geoffrey Lewis
Sent: Sunday, November 27, 2016
1:26 PM
To: Open PHD Guiding
Cc: geof...@hotmail.com; bw_m...@earthlink.net
Subject: Re: [open-phd-guiding] Is
the relatively poor Dec guiding a consequence of PE in RA?
Thanks again Bruce,
I'll definitely attend to the balance issue and thanks for the star cross
suggestion which is something I've never done, so I'll read up and give it a
try. If I do see those looped RA lines, what does that mean?
Let’s burn that bridge when we come to it. J In my case, it required sending the Dec assembly back to AP for an adjustment, but I was told it was a pretty rare problem. Hopefully, you’ll see an improvement by getting the balance straightened out.
Bruce
Geoffrey Lewis
Cheers,
Geof
Ray Gralak
Geoffrey Lewis
Thanks for chiming in. I've previously downloaded your PEMPro log viewer though not the latest version I saw that you posted today, but there's too much going on in there for my small brain to comprehend... :-(. Thanks for posting the YouTube which I've watched a few times, but it goes too fast for me to follow what you did.
I believe that the camera is in the same orientation as when I last calibrated which was at the start of the session on 18th Nov, which GuideLog I also uploaded later in this thread. Please could you review that for me and advise whether that shows the same 'false periodic error'?
Also would you take a phone call to talk me through your tool and the YouTube that you posted as I'm keen to learn what I can from experts like yourself and Bruce? Please PM with your contact details if that is an option.
Regards,
Geof
Ray Gralak
Geoffrey Lewis
OMG, I don't know how many YouTubes I've watched but I never new that you could change the playback speed, so thanks for that. Yes I can now follow it, though I still don't understand what you did. I know that you moved the rotation angle slider, but how did you know how far to move it, i.e. what were you looking at to decide too far, not enough, just right? Also I don't understand the significance of this.
I've never paid much attention to the camera angle, other than it currently sits about vertical to the horizon when the scope is in its park position with the RA axis horizontal and scope pointing south. That of course will change from target to target as I try to optimise framing. However, please note that I'm using the QSI's integrated OAG so the Lodestar guide camera is turned in the guide camera housing on the QSI to whatever position it needs to achieve focus whilst the imaging camera is in focus. Having done that I locked it down, so it very well may be at a weird orientation to the imaging camera and the mount axes. I'm new to using an OAG and I've found it very tricky to get focus with the guide camera and certainly getting the imaging camera and guide camera orientation the same is something that I've not been able to achieve. This is made some what difficult in that my obs has a warm room where my laptop lives and I can't see the laptop screen from the mount meaning that I'm going back and forth between mount and laptop just trying to fine tune focus without worrying about orientation of the cameras to each other and the mount axes.
No worries about the call, just easier for me to follow with verbal Q&A.
Regards,
Geof
Ray Gralak
Yes I can now follow it, though I still don't understand what you did. I know that you moved the rotation angle slider, but how did you know how far to move it, i.e. what were you looking at to decide too far, not enough, just right? Also I don't understand the significance of this.
Bryan
Geoffrey Lewis
Got it, I noticed the levelling out of the yellow frequency graph, so wondered if that was what i should be watching. I also noticed the smoothing of the Dec graph, so i think I now follow that part, so thanks for your patience in walking me through that. I also found you YouTube on the logviewer so have now watched that at x0.5 speed, which was also helpful.
Can I go back to your first post on this thread where you said....
'However, if you can confirm that the camera angle was correct, there might be some cross talk between the axes and you should contact A-P.'
Having confirmed that the camera angle was 31.6 ever since calibration, do I have a problem that I should discuss with AP? What is the relationship between the measured camera angle of 31.6 and the assessed Dec graph rotation angle of 65? I'm tempted to add them together to get close to the 90 of a right angle but maybe I'm just flailing around trying to make sense of numbers that I don't understand....
Regards,
Geof
Geoffrey Lewis
That was also my understanding, though I know that there are benefits in having them aligned and is something that I always tried to do when I was guiding with a separate guide scope. It was easy to eyeball the alignment of the guide and imaging cameras both to each other and the mount axes, something that I'm now struggling to do with my laptop physically separated from the mount by a the obs / warm room dividing wall. Something that I didn't consider in the build design...!!!
Regards,
Geof
bw_msgboard
Hi guys. Let’s please not drive off into the ditch on this topic <g>. PHD2 doesn’t care at all how the guide camera is oriented with respect to the main imaging camera. Some people like to keep them roughly aligned because it’s easier for them to see what’s going on, but that’s just a personal preference. It doesn’t matter to PHD2 at all. Since you’re using what amounts to an OAG, there are two things that can create the need to re-calibrate: if you rotate the entire camera assembly, including the guider or 2) you rotate the guide camera in its adapter because you’re fooling around with focus. If you haven’t done either of these things, then the camera angle computed by PHD2 won’t change. That said, many of us with OAGs do have to rotate the assembly in order to pick up a guide star – in which case we re-calibrate. But since you’re working with a refractor here, I would imagine you had plenty of stars to choose from.
Bruce
From: Geoffrey
Lewis [mailto:geof...@hotmail.com]
Sent: Sunday, November 27, 2016
4:35 PM
To: Open PHD Guiding
Cc: geof...@hotmail.com; bw_m...@earthlink.net
Geoffrey Lewis
I'm fairly confident that I didn't move the camera assembly, or the guide camera within the OAG housing since I last calibrated on 18 November, but its possible that I did. You're right I had lots of stars from which PHD2 auto selected the guide star. However, the discussion has given me food for thought, so I'm just trying to refine my understanding of the components so to speak. My intention remains to follow your recommendation to adjust balance and then run the star cross test, but I will probably also see if I can get a better orientation of both guide and imaging cameras to the mounts axes first, then recalibrate somewhere reasonably close to Dec 0 before running the star cross test. I'll then share my findings.
Just as an fyi the finally processed image of ICs 59&63 that I captured shows round stars, so at the imaging scale of my 4" TSAPO the guiding is perfectly adequate, but I am conscious that I will want to image with the C14 which will be less tolerant, hence my desire to better understand and if possible improve guiding performance. I had other issues with the image, due to poor transparency, but if anyone is interested you can see it at http://www.astrobin.com/273506/.
Best regards,
Ray Gralak
Geoffrey Lewis
A new calibration is definitely on my to do list along with a couple of other suggestions from Bruce. I'm going to be travelling the next few days, so I may not get onto this right away, but I've gleaned a lot of useful information that I hope to build into my personal, currently somewhat limited, knowledge base. I'll update the thread with my findings in due course.
Many thanks,
Geof
Geoffrey Lewis
I popped down to the observatory today to check the guide camera angle and adjust the scope balance. I can confirm that the guide camera was some 30-40 degrees offset from both the imaging camera and the RA/Dec axes, so just for neatness I'll try to realign that the next time I get a clear sky. I have a drop down shutter opening in my obs/warm room dividing wall which I installed to be able to observe mount movements from the warm room. I've established that I should be able to observe the laptop in the warm room from the mount through that when the scope is pointing in the general direction of Polaris, so I'll slew to there to fine tune guide camera rotation and focus, before slewing back to the south around Dec 10 to run a new calibration. The mount is now well balanced in both RA and Dec, so Ill see what difference that makes to guiding.
Thanks again,
Geof
On Monday, 28 November 2016 01:28:53 UTC, Ray Gralak wrote:
Geoffrey Lewis
I rebalanced the mount, rotated the guide camera to better align with the mount axes (PemPro Log Viewer now shows a camera angle of 1.6), recalibrated and then guided for a about 4 hours on ngc281 (Packman). The Dec guiding was much improved and indeed both Dec and RA ran very well for about 45 minutes, then I started to see significant oscillations in RA. I'm wondering whether that is due to the RA axis being too well balanced, i.e. not east heavy as the oscillations seemed to start at about the time that the target passed the meridian then continued thereafter. I was guiding through the meridian without any flip so ended with the counterweights high, but if the mount is balanced I'm not sure how that would make any difference. Anyway I've posted the guide log fyi, so will be interested to know what you think.
BTW the short session at the end of the log was me messing with the counterweights to see if moving them one way or another made any difference. I was tired and cold so probably this was a waste of time and I need to take my time with more structured diagnosis.
Best regards,
Geof
Ray Gralak
peter wolsley
peter wolsley
Geof,
I attached a graph that illustrates the strange phenomenon happening with your mount. This graph is of a exerpt from your PHD2_GuideLog_2016-11-18_183040.txt logfile. The GuideStar DEC Deviation and GuideStarRA Deviation traces are the RA and DEC traces that you normally see in PHD2.
The Culmulative DEC Movement trace is an visualization tool that I use to give me insight as to how your mount's DEC axis is being commanded to move. For this graph it clearly illustrates how your mount is moving back and forth thru it's DEC backlash. Using your guider camera's pixelscale it shows that your mount has roughly 20 arc-seconds of backlash. This assumes that your ASCOM driver for your mount is truely giving you the correct autoguiding speed.
The Culmulative RA Movement trace is an identical visualization tool to show you how your RA axis is being commanded to move. What is unusual here is that the Culmulative RA Movement is oscillating in sync with the Culmulative DEC Movement. It would normally be expected that these two axis would act independent of each other...DEC should be catering for backlash while RA is unaware of DECs movements and simply catering for things like PE, RA drift , seeing, etc.
This specific graph illustrates how your mount responds when DEC backlash comp is disabled. I'm sure that when you disable DEC backlash comp your comments would be that while you no longer see RA spikes you now see significant DEC deviations.
When DEC backlash comp is enabled on your mount the DEC deviations disappear but the RA spikes appear.Peter
Geoffrey Lewis
Thanks for you reply and I apologise for my delay in responding, but I'm currently travelling with limited access to the internet. I will certainly check the guide rates set in PHD2 and the Mount driver when I get home to see if they are different, though I'm not sure where to look for that in PHD2. What is curious to me is why I have periods of extremely good guiding, then this severe oscillation in RA kicks in without any changes in guide parameters.
Regards,
Geof
Geoffrey Lewis
Thanks very much for your 2 posts and for the graphical analysis that you've provided, which does seem to confirm what I observed, i.e. that changes in Dec are 'mirrored' by changes in RA or vice versa. Since I am currently travelling there is not much I can do to investigate, but I'll get back on to this when return home in a few days time. I'm somewhat fearful that I may be seeing a potentially expensive mechanical issue with the mount, but I'll continue to investigate other possibilities before going down that road, including performing the star cross test that Bruce originally suggested, but which I completely forgot about doing last night in my eagerness to capture data on another target.
Regards,
Geof
Bryan
Bruce Waddington
Hi again guys. I think this is another blind alley, not really germane to the problem. The calibration was nearly perfect, the guide speeds didn't change after the calibration was done, it simply isn't an issue.
Let's rewind the tape and see if we can get back on track. <g>
1. You said you rebalanced the scope. Did you find it was significantly out of balance in Dec before you adjusted things? This was a suggestion made to see if we could get the Dec guiding to settle down a bit, not something that would directly affect the RA guiding. Also, it's generally not necessary with these big AP mounts to worry about fine-tuning the balance after a meridian flip. Remember, these things are used routinely by people doing automated imaging, often in remote locations, and they're not out in the dome futzing with balance after a meridian flip. <g>
2. It's a shame you didn't run the star-cross test because we would probably be much further along. So if you can do that, it will definitely be helpful. I *think* it will show the "cross-talk" behavior you're seeing where a large Dec correction causes a shift in RA. There's no sense in getting a case of nerves about this, not until we've isolated the problem and you're able to talk to AP. In my particular case, the repair amounted to an adjustment and the cost was minimal. I know you're located outside the US, but the AP guys are very good at what they do and well accustomed to working with non-US users. Obviously, your problem might be entirely different than mine, but I don't think you should be overly pessimistic about it. What we're trying to do here is develop a very simple, very clear demonstration of the problem so they can look at it. By using the star-cross test, we're eliminating PHD2 guiding as a source of the problem, so you won't get dragged into long discussions about how the calibration is wrong or the guiding parameters are wrong, etc, etc.
Ok, so now let's look ahead and assume you've learned what the problem is but for some period of time you need to live with it and just mitigate the effect it has on your guiding. The first place I'd start would be to disable the Dec backlash comp in PHD2. While this isn't causing the problem, it may be exacerbating it. Although I don't have your debug log file and therefore can't see what the Guiding Assistant told you, it seems pretty clear your Dec backlash is pretty small. One of the problems you're having is that you're getting a lot of direction reversals in Dec, which is what triggers the apparent cross-talk problem. You will need to change things so those direction reversals happen rarely, if at all. One option would be to de-tune your polar alignment a bit so there's a higher Dec drift rate in one direction. Many imagers do this, it doesn't cause problems with guiding but it might introduce field rotation if the alignment error gets too large. The Best Practices document explains this and points you to an online calculator that will let you compute what your upper limit should be. You might also need to explore the option of doing uni-directional Dec guiding. But that's something to be dealt with down the road. As a first step, I would disable the backlash comp, de-tune the polar alignment a bit, and see what happens.
Good luck,
Bruce
From: open-phd...@googlegroups.com [mailto:open-phd...@googlegroups.com] On Behalf Of Bryan
Sent: Tuesday, November 29, 2016
5:26 PM
To: Open PHD Guiding
--
Geoffrey Lewis
I can't physically connect to the rig in the observatory as I'm currently travelling, but I have my laptop with me and have checked the rate settings in both the AP Ascom V2 Driver and PHD2 which confirms that both are set at 1.00.
Regards,
Geof
Geoffrey Lewis
Firstly thanks for being the calming influence in my escalating concern about the mount....
secondly, I was really mad at myself when I read Peter's post and realised that I had not performed the star cross test. That will now have to wait until I get home and next have a clear night. I have, however, checked the guide rate settings in PHD2 and the ASCOM V2 driver, which are both set to 1:00 as expected since PHD2 reads that from the driver.
Regarding the scope rebalance, actually the Dec balance was fine, I think that the heaver QSI camera in pace of the lighter Canon DSLR compensated for the weight of the corrector plate cover on the C14, but the heavier camera had thrown the RA balance out a bit to west heavy so I adjusted that put the RA axis back in balance.
Whilst there's nothing more I can do to test things until I get home, I do have access to the debug log on my laptop, so I've that uploaded it to Dropbox where hopefully you can download it via the below link.
https://www.dropbox.com/s/8qdipf08qk3mzq5/PHD2_DebugLog_2016-11-28_181708.txt?dl=0
It tried attaching it to this post, but got a warning message that I'd exceeded attachment file size.
I ran the guiding assistant for about 6 minutes immediately after running calibration; I then accepted the three recommendations on RA & Dec min moves and backlash compensation then let the rig run the entire 4 hours without changing any settings. If you see anything significant from analysis of the debug log for me to test in addition to the star cross test then please let me know.
Regards,
Geof
Ray Gralak
peter wolsley
Michael Garvin
No expert here, but perhaps some food for thought.
It’s been a full year since I had a scope set up with the move and all. Haven’t found the time to build my new observatory yet. Being out of touch and adding an old rusty brain, I’ll share some thoughts on the Dec movement/occilations with RA commands. I’m poor at communicating with the keyboard, so please try to make sense of this.
Lets take a setup using a Gem mount, Refractor, some form of draw tube extension be it reducer/flattener, spacers and/or other devices. Visualize with me a configuration where your mount is oriented with the saddle on the top side. Now, imagine the refractor setting there in all it’s glory with all this “stuff” hanging out the draw tube, complete with two miles of cable. From here, there are a few Dec issues that can, and usually will occur in our average setup.
In the above orientation, imagine a guide command, or even a sharp slew command and watch the reaction as it heads down the pike. A couple “what just happened” might go as follows.
Lets say your a bit out of balance in Dec, where some purposely unbalance Dec. A RA command will kick RA, and as a result, the heavy end of your assembly will temporarily kick in Dec. Sometime, when you have the time, disable Dec guiding and you likely will see this. It will level our pretty fast, but when your guiding, PHD sees the movement and sends a command. From there, the problem will exasperate itself by trying to guide out of balance movements where you are actually creating a wave.
Lets say you have a perfect balance with your cables screwed down tight. What can go wrong now? For kicks and giggles, grab hold of your tube in one hand and the camera in the other and start pulling it around. Remember, it only takes a few microns of movement to move a number of pixels on your camera. Same scenario as above, with this draw tube movement, an Ra command will likely kick off the slop occilations in Ra and Dec .
Solutions? Aside from the typical two setscrews and tightening up your light train, cables etc. consider something that is staring some of us in the face Imagine a cross section of your typical focuser, especially the low profile type found in Newts. Notice you have two rollers assemblies on one side and the adjustment bar/roller on the other. Problem is, the bar/roller is only offset by maybe 3/4” in front of the rear set of rollers. This is pretty much the same for all focusers. With the flex test above, you might notice it is hard to flex in one direction, but might get plenty movement in the other. Add our possible Dec issue I described above, I imagine this to be a viable source of Dec flexure, especially with a couple pounds of camera hanging of the back end. If your focuser has the option to rotate, try putting it in an orientation where Ra will not hammer this slop around. Sadly, if present, it will only move it to the RA side. Try a 45 degree rotation? My solution was to cannibalize a focuser and put an extra wheel car at the draw tube end of the focuser. Somewhere I have a picture of this.
If any of this is correct, perhaps an algorithm to kick out these outliers in Dec? Doesn't completely solve the problem, though it might cut down some of the oscillations, if only in magnitude?
If this has been covered, chalk it up to me not being current on this topic - sorry if I reinvented the wheel.
Michael Garvin
Ray Gralak
Hi Peter,
I hear what you're saying but the RA and Dec axes of the A-P 1200 mount are completely independent from each other so this type of behavior is extremely unlikely.
I think the likelihood that the camera angle is somehow wrong (via a software application bug, a cable tugging the autoguider, or something else) seems greater than a mechanical correlation between the axes. I'm not saying a mechanical reason is impossible, just very unlikely. The mount is massive and I do not think that tiny Dec moves would likely cause the RA axis to also move. In fact the Dec axis completely disconnects from the RA axis and the RA/Dec motors are completely independent. See pictures here:
http://www.astro-physics.com/index.htm?products/mounts/1200gto/1200gto
-Ray Gralak
peter wolsley
On Wednesday, November 30, 2016 at 11:14:25 PM UTC-5, Ray Gralak wrote:
peter wolsley
Geoffrey Lewis
I returned home from my travels to heavy cloud yesterday evening, but this morning opened the observatory to check a couple of things, particularly after reading Ray's observations on camera angle. I was concerned that maybe the camera had rotated since I knew that I'd changed the guide camera cable. The night that I rebalanced the scope and rotated the camera to align with the mount axes I'd been experiencing frequent comms drop outs, which I'd isolated to the guide camera cable. Unfortunately I only had a shorter replacement USB cable, which I thought was long enough, but today I found that the guide camera was rotated in the QSI OAG, with the entire camera assembly also rotated, so for sure it seems likely that the guide camera at some point was no longer in tune with the angle at which calibration had been performed. I've reviewed several of the sub exposures and cannot identify any rotation between first and last images, so maybe the camera assembly didn't rotate until after the imaging run had completed, but clearly the guide camera was rotated maybe 10-15 degrees in the OAG. One can only guess when that happened, but maybe the sudden change in guiding performance indicates the point of failure. Needless to say, today I've purchased a longer replacement USB cable, so next time I get a clear sky (more heavy cloud and rain again today) I'll install that and run some more tests including the cross star test.
Watch this space...
Regards,
Geof
Kent
One other thing to consider is the gear mesh of the mount. I have an AP 1200GTO and I had a similar issue that I finally resolved after a year or two. My Dec deviations would follow my RA deviations. The gear mesh is incredibly touchy with my mount. I finally gave up meshing the gears according to AP instructions and developed a more quantitative way of doing it. If the mesh is too tight, it seems you can get deviations in the other axis due to the axis moving around very slightly in it's bearings. Counter-intuitively, a tight mesh also appears to increase the backlash. You can read about what I did here:
https://groups.yahoo.com/neo/groups/ap-gto/conversations/topics/54721;_ylc=X3oDMTJybmtvYTZkBF9TAzk3MzU5NzE1BGdycElkAzEyOTI4OTMEZ3Jwc3BJZAMxNzA1MDgyODA2BG1zZ0lkAzU0NzIxBHNlYwNkbXNnBHNsawN2bXNnBHN0aW1lAzE0Nzk4MDMzNjE-
It is message 18 of that thread.
BTW: The worm bearing pre-load is another thing that can affect the issue.
I hope you get your problem resolved.
Regards,
Kent
Geoffrey Lewis
Thanks for your thoughts and the link to your previous post about your own experience with an AP1200GTO. Unfortunately that all seems to be above my pay grade; I have no idea what a dial gauge is let alone how to use one. The idea of me opening up the Dec gear assembly gives me the screaming heebie jeebies.... If I have a problem in there now then for sure I'd have an even bigger one if I poked around in there..... :-(.
Cheers,
Geof
Kent
I understand your trepidation; I had the same feeling. If you ever feel slightly adventurous, however, it is not at all hard or invasive to just look and feel. You can at least check if the mesh is binding. If you take the covers off the gear motor housings, you can, with your fingers, rotate the large spur gear (it's right there, easily accessible). It should move really easily - almost no resistance. It is hard to describe how it should feel, but if you have to turn it with some force, it is definitely binding and the mesh should be corrected. If it moves really easily, then you're probably fine although you may have excessive backlash.
Just a thought, if you reach the point where you're all out of other ideas.
Regards,
Kent
Geoffrey Lewis
Thanks again for your encouragement. I'll see if I can find some documentation with photos of what I'd be looking at if I decide to 'open the box' on the Dec gear assembly. I'm hoping that my findings earlier today with the fully extended guide cable and rotated camera are what caused the latest poor guiding episode, so I'll follow that option through first. I'm usually pretty good with mechanical things when I know what I'm looking at, so its not impossible for me to try to adjust the gear mesh, but I'd rather observe someone do it before I try it. My wife could tell you that there have been too many occasions when I started on a repair project not knowing what I was doing that didn't end well.... and this would be by far the most expensive equipment that I took that risk on. Lets hope that it is just the rotated guide camera...
Thanks again,
Geof
Ray Gralak
Kent
Good to hear from you again after several years. You've done some impressive things in the meantime.
I don't have a rotator and I have positioned my camera fairly close to parallel with RA/DEC. The angle is currently 179 degrees. Although I know there is some PE, I don't see it in my 1200 mount. I did use your PemPro when I got the mount in 2012 to look for it. There was none apparent so I didn't look the gift horse in the mouth any more.
There is a difference between Geof''s excursions and mine; mine were on a small scale. Whenever there was an excursion or a correction in RA, there was a corresponding one in Dec. This could be due to wind also, and probably some of it was, but when I reduced the gear mesh, it would go away.
The movement of the 1200 is amazing in that I have to balance my (fairly long) OTA within 2 ozs of weight on the end if I want perfect balance. It moves really easily. So I figure there may be just the tiniest bit of wiggle room in the axes that might be responsible for the correlation.
I am in a windy area and need extremely precise gear mesh to minimize backlash as much as possible when wind gusts occur. I got some of that Aeroshell high pressure grease that other AP users have been using and I think that has helped to get a slightly tighter mesh without restricting movement. I also have developed a more quantitative way of adjusting the mesh. All this is probably more important for me than other users because of the wind I get here in New Mexico. The system has been working well now for the last couple of years so I'm happy (finally).
Thanks for your offer to look at my former issue, but I think I'm good now. It I find I'm not, I will remember your kind offer and perhaps bug you.
Regards,
Kent
Geoffrey Lewis
Tonight I had clear skies again so hooked up the guide camera with the longer USB cable and reset the guide camera broadly square to the mount axes. I then ran the cross star test using manual guide pulses in PHD2 (see attached image) followed by another imaging session to collect more data on ngc281, being the same target as my last session. I started the session much later into the evening with the target already past the meridian, so the scope flipped on acquisition and I let it run from there for about 3 hours. On this occasion, however, I did play around with various settings, particularly Dec compensation and both RA and Dec min moves, including for a short period turning Dec guiding off completely. My guiding performance was much more stable, but as you will see from the GuideLog there were periods of instability which seemed to show the same pattern previously witnessed of RA excursions following a Dec guide pulse correction. I eventually opted to guide without Dec compensation, a Dec min move of 25 and RA min move of 15, which yielded fairly lengthy periods of good guiding, but even then there were periods of guiding instability. I checked that cables were not dragging, or at least I saw no evidence of that. There was one huge excursion, which looks like something hit the mount and I wonder if a bird (owl maybe) perched on the rig for a moment. I checked the guide camera and imaging camera assembly orientations at the end of the session, before parking the scope and everything was exactly the same as at the start of the session. I did not, however, perform a new calibration as I'd put the guide camera as close as I could to the same orientation as the previous session, so used the previous calibration.
The cross star test produced what seems to me many perfect cross star patterns, but please let me know if I'm interpreting that correctly. The attached jpeg image is a stretched version to better show the result of the test. Any further observations on this latest guiding performance will of course be very welcome.
Best regards,
Geof
bw_msgboard
Hi Geoff, this is pretty interesting. I agree, the star-cross test result looks fine – I hate it when I guess wrong. L Very quickly though, I can’t see any evidence that you ever disabled declination backlash compensation. Are you sure you did? I’m concerned that you got confused and changed “declination compensation”, which is a completely different animal. What I was hoping you would try was to disable *backlash* compensation – that’s on the algorithms tab:
I’ve only taken a quick look, but what I see is that you continue to get an RA excursion whenever there is a *large* Dec correction. Small dec corrections don’t appear to have this effect. The inverse is not true, large corrections in RA have no apparent effect on Dec. If you turn off the declination backlash compensation (just un-check the above checkbox), most of these large Dec guide pulses will go away. It would be very interesting to see what that does for your overall guiding.
Bruce
From:
open-phd...@googlegroups.com [mailto:open-phd...@googlegroups.com] On Behalf Of Geoffrey Lewis
Sent: Sunday, December 04, 2016
6:02 PM
To: Open PHD Guiding
Cc: geof...@hotmail.com; bw_m...@earthlink.net
Subject: Re: [open-phd-guiding] Is
the relatively poor Dec guiding a consequence of PE in RA?
Hi everyone,
--
Geoffrey Lewis
Well you may have been wrong about the star-cross test, but you were spot on with me turning off Use 'Dec compensation' on the guiding tab instead of 'Use backlash comp' on the Algorithms tab - doh...!!! I just fired up PHD2 and sure enough 'Use Dec compensation' was unchecked and 'Use backlash comp' was checked with a revised amount of 1057. I can't believe that I did that, but my excuse is that I'd been out all day at a pre Christmas family gathering and had drunk a 'few' glasses of wine with my lunch followed by some after dinner drinks. My teetotal cousin provided the taxi service for me and my wife, so I was ok on that score, but perhaps the adage 'don't drink and drive' applies equally well to telescope mounts as it does to cars....!!! It was a superb clear night last night and who knows when we'll get the next one, so I wasn't going to pass that up....
Ok, well the good news is that I still have something to check....
One other thing that I noticed when I parked the scope was that the slew speed of 600 was very sluggish. I have We had a very heavy frost last night and whilst I don't have an outside thermometer I suspect that the ambient temperature was sub freezing, maybe -3C (27F), so it is possible that the spikes in RA that were not the result of the large Dec corrections are from stiction in the mount gears due to cold temperatures, not that -3 is particularly cold compared with what many mounts must experience in many parts of the world?
Power to the mount is provided via a variable (3v-15v) 40a DC power supply(see below link), which is set to 13.8v.
https://www.rapidonline.com/rapid-sps-9400-209mg-smps-switch-mode-power-supply-15v-40a-with-digital-display-85-1828?gclid=COCo3_rj3NACFcad7QodeFEJNw
Usually the mount draws only 6a-7a, but I noticed that it was drawing ~10a under slew when parking, but that is still well within its 40a capacity, so I suspect this is just a red herring, but I want to provide as much data as possible about the conditions.
Regards,
Geof
Thanks again,
Geof
Bryan
Geoffrey Lewis
I've reviewed the AP power supplies, but they are manufactured for the USA market and 110v-115v mains power, whereas the UK is 220v-240v, so I'd also need a 240v-115v transformer, which together with the international shipping costs makes the AP models prohibitively expensive. The power supply I've purchased is what has been recommended to me from a UK based astronomy supplier, albeit purchased direct not through the astro supplier. I was advised that it should be more than adequate for any conditions that I'd experience in the UK, but I'll continue to monitor it. I only noticed the sluggish park slew as I wanted to check the cameras and cables before parking the scope to make sure that they had not moved during the imaging session. I had noticed that the cameras had rotated after parking for my previous session, but I'm aware that that may have happened during the parking slew, hence this time I checked it before parking. I then watched the mount park to see if any cables snagged, which was when I observed the slightly sluggish slew. I checked the mount's power light and that remained on steadily so I think all was ok, but maybe its just something else to consider. There seem to be so many gotchas in this hobby...!!
Regards,
Geof
Bryan
peter wolsley
~ 1.3 to 2.5 amps at 13.8 volts - both motors slewing at 1200x
Power requirements Nominal 12 volt DC supply at minimum of 5 amps continuous
Geoffrey Lewis
Happy Christmas to you too ;-).
Geof
Bryan
Geoffrey Lewis
Thanks for your observations about my power supply and the current drawing requirements of the AP1200. I should explain that the power supply does not only run the mount, but also a number of items of ancillary equipment via a powered hub, which also provides a fused break point between the supply and the mount. In addition to the mount there are three dew heater straps and of course the cooled CCD camera, albeit at present I'm running that off a separate regulated supply, but I have the option to hook that up via the hub in future. I estimated that the maximum draw on the power supply unit with everything running would be maybe 20a-25a and the price difference between a unit supplying 25a and 40a was minimal so I opted for the larger unit.
Regards,
Geof
Geoffrey Lewis
Thanks for reproducing the guidance from the AP manual which is similar, if not identical, to that which I've read about the AP1200, as I direct consequence of which I've limited my slew speeds to 600x. I can set the voltage on my PS, which is currently fixed at 13.8v, but I could increase that to the max 15v if necessary. I'm no electrician so watts, amps and volts are not familiar currency to me - I'd need to think back some 50 years to my school days for that, so I've opted for taking the advice given by someone that I trust who makes astro electronics equipment, who actually installed my observatory and most of the equipment. That is not to say that I don't value the excellent advice that I'm getting from you and others on this and several other forums, all of which is very much appreciated.
Many thanks,
Geof
Bryan
Geoffrey Lewis
I now live about 100 miles NE of London in a very rural location about 30-40 miles from the Norfolk coast. You are correct in that we have had some lower temperatures here the past few days, but certainly not exceptional. Depending on wind direction it can get very cold here with night time temperatures well below freezing not uncommon, though average winter monthly lows (Dec through Feb) are general a few degrees above freezing 2C-3C (34F-37F). Those averages, however, reflect the fact that the prevailing SW wind direction for all of the UK is warm moist air, but when we have high pressure zones sitting over the north Atlantic during the winter then the prevailing wind direction over the UK flips round to N/NE pulling freezing air straight down from the Arctic, so -5C (23F) is not uncommon and certainly there have been occasions at or below -10C (14F). I will have to monitor how the rig performs if I venture out to image during any such freezing periods, though I suspect that may be unlikely...!!
Regards,
Geof
Geoffrey Lewis
First let me wish everyone a very happy Christmas and health, happiness, plus of course dark skies in 2017.
Due to constant poor weather its been nearly 3 weeks since I've been able to open the observatory, but with clear skies promised I eagerly set up for an imaging session last night. In the end there was fog forming so the data captured was very poor, but I did want to try guiding with backlash compensation disabled to see how the rig performed. I've attached the GuideLog for anyone that wants to take a look, though I suspect not this side of Christmas. You'll note that I tried a period with backlash compensation disabled which wasn't particularly good guiding, followed by a period with it enabled, when the previously seen pattern of RA spikes following large Dec corrections returned, followed by a longer session with backlash compensation again disabled. You will also note that I experimented with both Dec Aggression and Dec Min Move and I seemed to get the best guiding of the session with those set to 70% and 35 whilst Dec compensation was disabled. There was one huge excursion which I suspect was due to something external - I live in a very rural farming location and we have owls hunting at night, so I suspect that one of them thought that my scope would make a good perch....!!! There a few other less significant Dec excursions, but other than the suspected 'owl' excursion my RA and Dec guiding were 0.6" and 0.7" respectively.once I'd settled on those guiding parameters.
As mentioned seeing was poor with a very active jet stream racing right over the UK, so no doubt that will have contributed to poor guiding, however, I remain disappointed with the performance of the mount as it actually seems worse than my NEQ6, albeit that carried a much smaller payload.
I'd be interested on any further input and guidance as to whether I should be contacting AP to see if the mount needs to be serviced, either by me or perhaps even returned to them. The guiding performance is currently not adversely impacting the images taken with the 4" TSAPO100Q, but I'll be getting an Optec telecompressor (focal reducer / field flattener) to pair with my C14 and I'm concerned as to whether the guiding is going to be good enough for that combo.
Best regards and TIA,
Geof
On Monday, 5 December 2016 02:40:25 UTC, Bruce Waddington wrote:
Ray Gralak
Geoffrey Lewis
I used the AP Log Zipper, which I've never used previously, so is the attached what you want? I look forward to learning what you discover from it.
Regards,
Geof
bw_msgboard
Hi Geoff. I’m sorry to hear you’re still struggling with this. But working under poor seeing conditions with a known active jet stream overhead if not likely to give you much useful information. Anyway, here’s my take on where you stand at the moment.
Dec Backlash Compensation Off
With backlash compensation disabled, you had a run of about 40 min at the end of the session with total guiding RMS of 0.73 a-s. IMO, that’s pretty typical performance for a night of poor seeing and I have seen much worse with high jet stream activity. And we know from earlier logs that your mount can deliver performance in the 0.5-0.6 a-s RMS range under better conditions. These all extend across many periods of the RA worm gear so they are not isolated instances of good performance.
Looking back at your history, I see a couple of things that we probably should have talked about earlier. Assuming I’ve found the right info, it looks like you’re currently working with a 100mm TS-optics refractor. That scope evidently weighs nearly 11 pounds, which is a pretty heavy piggyback load on the Celestron. Remember that we’re looking at deflections on the order of a few arc-seconds, so any sort of flexure at all in the refractor assembly could contribute to the problem. More on that later.
Dec Backlash Compensation On
As you say, the earlier pattern is evident when the backlash compensation is enabled. This took me back to the star-cross test you ran earlier, the one I said looked ok. Well, maybe not so fast. J You sent us a compressed jpg image, so we don’t know anything about the image scale. If I’m guessing right that you’re using a QSI main camera with 5.4u pixels at a fl of 580mm, that results in an image scale of around 1.8 a-s/px. That’s pretty coarse when the RA deflections we’re seeing are in the range of 2-3 arc-sec. So we’d be looking for a 1-2 pixel shift in a compressed jpg image, not something I have much confidence in. In fact, if I magnify the image in MaximDL and look very hard – well, I’m just not so sure. J
Next Steps
If you want to better understand what the mount is doing, I think you should start doing tests with the long focal length Celestron OTA. Since you’re using an OAG, you won’t have to worry about mirror flop and differential flexure. That will give you a finer image scale for seeing these small deflections and will also eliminate the heavy refractor as a source of trouble. You’ll need to do this when the seeing conditions are at least average and probably not when you’re trying to image something. Maybe you can get some test sessions done during a bright-moon period in the next few weeks. In the meantime, if you want to continue imaging with the refractor, just do that with backlash comp disabled and don’t obsess about the mount’s performance. If you have reasonable seeing, you’re likely to get nice round stars and decent results. My impression is that you’re happy enough with the refractor images but are fretting about what will happen when you move to the long focal length. So I would say, quit fretting and just move on to measuring it. J
If you decide to pursue the testing, I recommend running the star-cross test again through the large OTA making sure you save the full-resolution, uncompressed FITs rendition. You may need to adjust the duration of the guide pulses to make sure that enough of the star cross patterns don’t run off the edge of the field. I’d start with something like a total of 12 seconds in each direction. If you want to download the latest dev build of PHD2, there’s now a star-cross tool menu item that will do most of the work for you. J But that’s not important, just so long as you capture the image – the “how” doesn’t matter.
Good luck and happy holidays,
Bruce
From: Geoffrey
Lewis [mailto:geof...@hotmail.com]
Sent: Friday, December 23, 2016
8:48 AM
To: Open PHD Guiding
Cc: geof...@hotmail.com; bw_m...@earthlink.net
Subject: Re: [open-phd-guiding] Is
the relatively poor Dec guiding a consequence of PE in RA?
Hi Bruce and all,
peter wolsley
Geoffrey Lewis
Happy holidays to you too.
Yes, I'm still struggling with the guiding performance, but that is mainly because this is the first time that I've had a usable sky to run any tests since my last post and even then it was a poor seeing session. Thanks for your feedback and suggestion to use the C14 for further tests.
I understand the problems of testing under a fast jet stream, but the location of the UK makes that a very common experience, so I have to use whatever clear skies are available.
I'm a little surprised about your comment concerning the TSAPO piggy packed on the C14 being too heavy as I thought that it would be well within the weight limit range of the AP1200, or are you just saying that it would be better for me to mount the APO side by side with the C14 instead of piggy backing it on the C14? I thought that using an OAG took flexure out of the equation, but perhaps I misunderstand how flexure contributes to guiding performance. That said I can see how mounting the TSAPO on a plate side by side with the C14 would move it much closer to the mount which can't be a bad idea. I'll see if I can get some aluminium plate and fabricate a bespoke mounting bracket for both scopes.
I haven't used the C14 much for DSO imaging thus far as I've been waiting to receive a good quality telecompressor from Optec, which arrived a few days ago and I will be 'opening' it as a Christmas present from my wife tomorrow. I couldn't guide the C14 via an OAG at is full F11 (4m focal length) as I couldn't find any usable guide stars, so I'm hoping that the new telecompressor will solve that problem as well as giving me a much better FOV and flatter field for imaging.
I uploaded a JPEG of the star cross test as the FIT file was 16M exceed the group's posting fie size limit, so I've attached a link to it in my Dropbox from where you can download it if you want to check it out.
https://www.dropbox.com/s/y4rljmflorenfv3/QSI-TS100_StarCrossTest-001-L-10C_120Sec_Bin1.fit?dl=0
I will, however, repeat the test with the C14 as soon as I can get that set up with the new Optec telecompressor.
Thanks for your continued support and guidance and I again wish you a very happy Christmas.
Geof
Geoffrey Lewis
Season's greetings to you too and thanks for your analysis even if 'guestimates'.
Its arguable whether I'm actually doing any imaging, certainly this last session didn't yield any usable image data due to fog, but at least I could try to test the guiding performance. As I just replied to Bruce I hope to soon be able to use the C14 for DSO work and as Bruce states that FOV should give me much finer results for testing so that is my next step. I may also even revert to mounting the TS APO on its own to see how the rig then performs, so still several options to work on.
Best regards,
Geof
bw_msgboard
Hi Geoff. My comment about the weight of the guide scope has nothing to do with mount capacity – the AP1200 won’t even feel that. J My point was that you might be getting a small amount of movement between the refractor and the mounting rings or between the mounting rings and the dovetail plate. In other words, the refractor has several mechanical interfaces between itself and the mount. I really don’t think this is a problem, but I’m just trying to be thorough and eliminate as many things as possible. I think if you start testing with the C14 at the finer image scale, we will be better able to see what’s happening. The FITs image you sent is helpful, that’s how we’ll need to look at things going forward.
As for the UK jet stream problems, I understand. Of course, if you’re typically going to have 3+ arc-sec seeing, you are probably already getting guiding results that make you seeing-limited. In that case, you may not want to spend a lot more time looking at the mount performance.
Cheers,
Geoffrey Lewis
All understood. I will definitely run tests using the C14 if I can get suitable guide stars using the new Optec telecompressor. Watch this space....
Cheers,
Geof
Bryan
Geoffrey Lewis
Sorry for my tardy reply, not only did a fun family Christmas get in the way, but for some reason my email account treated your post as junk so I only just saw it in my junk folder. I hope that you and yours also had an enjoyable holiday season.
Yes, I understand that the Optec lens will reduce the FL of my C14 with a corresponding change in image scale, but it will still give me a FL of 2.5m so over 4 times as long as the 580mm of the TSAPO, with a corresponding image scale. I've tried guiding the C14 at its full 3.9m (F11) with an OAG, but have not yet found a suitable guide star on selected targets, though I guess I could try to guide on a bright star like Aldeberan, Rigel, Betelgeuse, etc., just to test the rig.
Thanks for the feedback.
Geof
Geoffrey Lewis
I hope that you all had relaxing and enjoyable Christmas and New Year holidays and I wish you all clear skies during 2017.
Well tonight I set up my C14 with the new Optec telecompressor. I thought it was advertised as F6.3, but having solved a resulting image on Astrometry.net I think it is just over F7 (F7.12 to be precise). Perhaps someone could check my calculations, but I used Ron Wodaski's CCD calulator to match the Astrometry.net calculated FOV of 30.2 x 20.1 arcmin (see JPEG file attached). My scope is a C14 (aperture 356mm), sensor was Canon 600D (T3i), so an array of 5184 x 3456 pixels with pixels at 4.3μm.
Anyway, as Bruce suggested I wanted to test guiding with the C14, hence taking the piggy backed 4" TSAPO, with any associated flexure, out of the loop. Guiding was through an OAG, but the results were not great. I have attached the GuideLogs fyi, which shows that I tried a bunch of things to see what difference any changes in settings made. There are two logs as I disconnected everything to change the OAG prism and guide camera orientation after the first session as the OAG prism was casting a huge shadow in the captured images, which I wanted to see if I could improve, but tonight was more about guide testing than data capture. I tried to attach the debug logs, but that exceeded the allowed files size limits, so let me know if you need them and I'll share them via Dropbox.
Clearly the large Dec guide pulse induced RA excursions remain, whenever I enabled Dec compensation, so I tried with that both on and off. I ran the guiding assistant a couple of times, as one was aborted when I lost the guide star during the Dec backlash calculation. The GA shows that my polar alignment is pretty good, so I tried guiding with Dec set to none and also unidirectional guiding South, as any drift appears to be North (well during most of this session anyway). The smoothest guiding seems to be when using Dec guiding = South. There seems to be a lot of RA drift, which I don't understand, so any feedback on this would be great, especially if there are any settings that I could adjust to counter this, as I am running out of ideas on how to tame this mount's guiding performance.
As always TIA for your feedback and recommendations.
Best regards,
Geof
bw_msgboard
Just some quick comments pending the debug logs:
From: Geoffrey
Lewis [mailto:geof...@hotmail.com]
Sent: Monday, January 02, 2017
4:54 PM
To: Open PHD Guiding
Cc: geof...@hotmail.com; bw_m...@earthlink.net
Subject: Re: [open-phd-guiding] Is
the relatively poor Dec guiding a consequence of PE in RA?
Hi and Happy New Year everyone,
I hope that you all had relaxing and enjoyable Christmas and New Year holidays
and I wish you all clear skies during 2017.
Well tonight I set up my C14 with the new Optec telecompressor. I thought it
was advertised as F6.3, but having solved a resulting image on Astrometry.net I
think it is just over F7 (F7.12 to be precise). Perhaps someone could check my
calculations, but I used Ron Wodaski's CCD calulator to match the
Astrometry.net calculated FOV of 30.2 x 20.1 arcmin (see JPEG file attached).
My scope is a C14 (aperture 356mm), sensor was Canon 600D (T3i), so an array of
5184 x 3456 pixels with pixels at 4.3μm.
This is normal for moveable mirror systems like SCTs. The manufacturer doesn’t tell you this, but as the mirror is moved up and down in the OTA to achieve focus, the focal ratio changes. So if the specified focal ratio is f/8 or f/10, that is really only correct for a particular focal plane position, usually the one that’s used for visual observing. With these scopes, you typically have a huge amount of back focus range, but the price you pay is that the system doesn’t stay at its optimum optical spacing. This isn’t likely to cause you any trouble, it is what it is. Also, the focal reducer’s effect depends strongly on the separation between the sensor and the FR – small changes in distances from the optimum can make a substantial change in the reduction factor.
Anyway, as Bruce suggested I wanted to test guiding with the
C14, hence taking the piggy backed 4" TSAPO, with any associated flexure,
out of the loop. Guiding was through an OAG, but the results were not great. I
have attached the GuideLogs fyi, which shows that I tried a bunch of things to
see what difference any changes in settings made. There are two logs as I
disconnected everything to change the OAG prism and guide camera orientation
after the first session as the OAG prism was casting a huge shadow in the
captured images, which I wanted to see if I could improve, but tonight was more
about guide testing than data capture. I tried to attach the debug logs, but
that exceeded the allowed files size limits, so let me know if you need them
and I'll share them via Dropbox.
Yes, I’d like to see the debug logs so I can see what the GA had to say. For example, I wonder about the min-move settings, they seem pretty low.
Clearly the large Dec guide pulse induced RA excursions remain,
whenever I enabled Dec compensation, so I tried with that both on and off. I
ran the guiding assistant a couple of times, as one was aborted when I lost the
guide star during the Dec backlash calculation. The GA shows that my polar
alignment is pretty good, so I tried guiding with Dec set to none and also
unidirectional guiding South, as any drift appears to be North (well during
most of this session anyway). The smoothest guiding seems to be when using Dec
guiding = South. There seems to be a lot of RA drift, which I don't understand,
so any feedback on this would be great, especially if there are any settings
that I could adjust to counter this, as I am running out of ideas on how to
tame this mount's guiding performance.
If you want to pursue the problem with dec backlash compensation, I’d like to see another star-cross test at this long focal length. I know it’s a pain in the butt, but the earlier result was ambiguous because of the coarse image scale. As for the RA drift, I can think of a couple of possible reasons:
1. There is something sagging in the optical train – this is pretty common. “Flop” of the primary mirror often causes this sort of thing with SCTs but so can any sort of looseness or gravitational sag in the camera and OAG assembly. You might have a lot of weight out there on a long moment arm. I’ve even seen a problem where the back plate of the OTA was just a tiny bit loose – but enough to create drift at the long focal length. As it turns out, this doesn’t appear to be causing you a problem, it’s the sort of thing that’s pretty easily guided out.
2. Something dealing with the mount: tracking not set to sidereal or loose clutches – very unlikely I would say. You should be able to rule this out by running the GA on both sides of the meridian. If the direction of RA drift (E/W) reverses, you’ll know it’s not the mount. Also, if the problem is gravitational, it should get better as you measure closer to the zenith. At this point, I wouldn’t worry much about it.
Cheers,
Bruce
Geoffrey Lewis
The two debug logs to go with the previously uploaded guide logs from last nights session may be downloaded from my Dropbox via the below link.
https://www.dropbox.com/sh/qex51mh5vdqj2pq/AADkhpaILOZxE--3r06bFnqRa?dl=0
Regards,
Geof
bw_msgboard
Thanks, Geoff. I can see now why the min-move settings are set where they are. I was forgetting how big the pixels are in the Lodestar, plus you must have had pretty decent seeing. I still have the feeling there’s something not quite right with the Dec assembly on the mount. The Dec backlash seems a bit high compared to what we normally see with AP mounts, which is something that might be improved by re-meshing the Dec gear train – I can’t recall if you’ve done that. But overall, there seems to be too much movement in Dec - the apparently random movements there are generally larger than RA. That’s opposite to the normal case. Consider this 10-min session:
The green is Dec, the red is RA. Overall, the envelope of the Dec movement is at least 5 arc-sec while the RA is more like 2 arc-sec. What is causing these big deflections in Dec – the motor isn’t even running at those points. My instinct is that there’s a mechanical problem on the Dec axis assembly, possibly a problem with a bearing. An inadequate pre-load on the bearing, for example, could also cause the problem you see with backlash compensation enabled.
Also, you fairly frequently get some very large, abrupt deflections in Dec – what is causing those? If you push the scope by hand with the clutches tightened, can you detect any sort of slop or wobble on the Dec axis?
Bruce
Geoffrey Lewis
Many thanks for continuing to work through this with me. No I have not tried to re-mesh the Dec gears, I have no idea how to tackle something like that. Is there an idiot's guide somewhere. I also don't want to introduce another variable into the equation until I'm more certain that the issue is mechanical, not user (with me that's quite likely) or software related. As far as I can detect by trying to rock the OTA there is no discernible movement in either Dec or RA with the clutches tightened and FWIW I am only hand tightening them.
I know that this is a side bar to the Dec guiding issues, but why is my RA drifting so much compared to Dec, or is this normal? It is very visible when running the GA, where Dec stays close to the origin, but RA wanders off. I have the pulse and guide rate setting at x1 in the AP ASCOM driver.
Regards,
Geof
peter wolsley
Geof,
Geoffrey Lewis
I agree that currently unidirectional guiding (South) does seem to give the best results. Thanks for your suggestions on Dec Aggression settings, indeed I have experimented with that all the way between 10% and 100%. My usual setting during testing, however, has been 70%, but I ended up at 100% last night as I wondered whether the mount might respond better by applying the full pulse. As you and others will probably appreciate, some of the changes to settings that I've made have been more or less desperation to find a solution, any solution...!!
Thanks for continuing to make suggestions, they are all welcome and I'm sure that I'll exit this episode much better informed about guiding in general and my mount more specifically.
Regards,
Geof
Geoffrey Lewis
It has been a couple of months since I posted to this thread, but that doesn't mean that I've not been working to resolve the guiding issues that I was experiencing with my AP1200 mount. I am please to say 'was' as it seems that the problem has been cured and I wanted to provide feedback and thanks to those that helped me out.
At Bruce Waddington's suggestion we took the discussion off line as he had previously experienced a similar problem with his AP mount and felt that he could work with me to resolve the problem. After further diagnosis we agreed that a call to Astro-Physics was probably appropriate with Bruce continuing to facilitate the discussions with the A-P team. To summarise, I had several helpful email exchanges and telephone calls with Howard Hedlum at A-P, culminating in a Skype video call via iPhone on 14 Feb (one week ago), where Howard and Wally (A-P mount supervisor, who it transpired also build my mount back in '07) took me through some diagnosis and maintenance. The diagnosis suggested that the Dec gear assembly had been over tightened causing stiction, which presented as a large backlash in PHD2 with over correction when the required large (>1000ms) guide pulses eventually moved the Dec axis. We also discovered that the locking bolts that secured the RA and Dec axes together had worked loose, so very likely that was also a contributory factor.
The last couple of evenings have been the first opportunity to run my rig to assess whether we have solved or at least improved the mount's performance. Last night was windy, but I had over 5 hours guiding on M109 with guiding mostly falling with RMS of 1 arc second and extended periods of 10-20 minutes when the wind dropped where guiding was with 0.5 arc minutes. If you are interested the below link is to the final processed image from that session on my Astrobin page...
http://www.astrobin.com/287953/C/
Tonight was less windy and guiding was superb (well superb to me) average RMS on each axis approx 0.3 arc seconds over 5 hours with peaks of just over 2 arc seconds. PHD2 is now also reporting backlash of under 200ms. I'll take those stats all day long.
For those interested I have attached the guide log from tonight's session. I have not processed the image data yet, but due to cloud a lot of subs were ruined, so I may have less data to work with than last night.
Anyway, as I stated in my opening comments the main reason for this post is to close the loop on this topic, with a huge gratitude to Bruce for his continued support over the past nearly 3 months. My thanks are not just to Bruce and the A-P team, but also to the great product that is PHD2, which very much helped diagnose the mechanical problems with my mount.
Best regards,
Geof
Andy Galasso
Geoffrey Lewis
The 'customer care' that you, Bruce and others on this group provide puts many of our big name commercial service providers to shame. It is difficult to put into words how valuable your knowledge, expertise and time freely given is to likes of me, but it is hugely appreciated.
Cheers,
Geof