I’ve had a look at the log and I can see a plate solve hint position. This seems to be a J2000 position, at least it’s closer to the J2000 slew position than the JNow slew position.
Where does this hint position come from? If it’s read directly from the mount then it could indicate that the mount is using J2000 coordinates, not JNow.
The hint for these solves comes directly from the mount, but SGPro understands it is reading JNow and converts the location to J2000 automatically (because that’s what the solver wants).
I too had one of these similar frustrating nights last night with SGP. No matter what, I could not get the target centered. I was going after NGC 457.
Lenovo X220 Tablet / i7, Windows 8.1
SGP 2.4.3.25
iEQ45 Pro w/iOptron ASCOM
Everything had been working just fine over several prior nights and the mount hadn’t been moved since the last successful imaging session. I did insert a new spacer flange (1.25") from MoonLite on my focuser to help reduce the amount the focus tube extends to get prime focus with my DSLR. I rather doubt that had much to do with this, but wanted to mention it just because it was a ‘change’ in equipment.
I did re-do the 3-star alignment, just to be sure the mount was properly aligned and when I checked my guiding with PHD (forced a re-calibration) it was guiding very well. Didn’t check the drift-alignment, but given the good guiding graph, didn’t seem necessary.
I finally gave up, and decided to bring the mount head inside to warm up since its been out there for several weeks and the temps for this weekend are going down below zero. I’ll try again once the sky’s clear and temps get near freezing again. Any thoughts on why this would occur all of a sudden?
@rjbokleman Your logs don’t really show anything suspicious (like mount failure or the like) so I can’t really be sure. SGPro is doing what is asked of it:
Figure out where the scope is pointing, sync your mount to that position
Slew to requested J2000 position by converting it to JNOW and sending it to your mount
Validating the new position, repeat
If you throw out the initial validation (the one that occurred after a long slew), your error range deviates only about 50 pixels (~50 arc-seconds for you)… it’s pretty consistent actually so that “kind of” rules out “chasing the target” (backlash). Doesn’t rule it out 100%, but I would look at other stuff first.
I am also not super familiar with your mount. I do know there were some issues with early versions of that ASCOM driver. Are those drivers up to date? You also may want to avoid the “hand paddle” alignments all together and just perform a blind sync before you center or manually center the scope on bright star and use a planetarium to sync the mount. Not sure if maybe the hand paddle method is serving as some sort of model that is getting messed up when SGPro attempts to perform a sync.
It’s a pretty popular mount… maybe some other iOptron users have some advice.
I have this mount and I find that centering is within a few pixels (less than 10) usually in one or two solves. Lately I have been using PlateSolve2 with great results. I wonder if your plate solve parameters are set up correctly (e.g. scale)
Do you use a fixed pointing model? I ask because I use EQMOD and found that if I do not load the previous pointing model the plate solves work fine (Pinpoint), but before that I had errors like yours of hundreds of pixels out and it would not deviate much from that. I use an AZ-EQ6 GT mount.
Just realized I replied to this question months ago.
Anyway even with a AZ-EQ6 GT mount I get the position to within a few pixels in 2 iterations. I have it set to aim for better 12 pixels, but often it’s within 5 pixels and I could take that further if I wanted.
I had to wait for the sky to clear around here to see what was going on.
Once I setup the mount again I did a 3-Star alignment using Castor, Pollux and Deneb. I then slewed around to a few objects to check it and what I found was … it was way off. I then tried sync’ing to specific stars. Still it was off based on my reticle EP.
I then decided the mount was foobar and decided to clear its mind using the ‘Clear Alignment Data’ option on the HC.
Once I did that the first star I slewed to, to start the 3-star alignment was nearly dead center again! After completing and then testing the 3-star alignment…everything was working as expected.
I then proceeded to put my DSLR in place and SGP centered to within 14px of NGC 457. Perfect!
Conclusion then is that the iOptron iEQ45 Pro alignment model must have gotten messed up somehow. Perhaps I had a slight pier rub of some kind? Not completely sure, but all worked as expected last night…and if it happens again know what to do to clear it up.
I am pretty much having the same issues with my iOptron CEM60. Spent a few months without being able to use the mount due to bad weather and the last 3 times I tried I failed to be able to solve. It takes the picture, tries to solve, gives me a huge error of over 100 pixels and the error is usually around the same number and fails! I can’t wait to have another clear night, clear my alignment data and give it another try. I really have my fingers crossed and can’t think of anything else that could be causing this problem.
Ladyhawke,
How many attempts have you told SGP to make plate solving? An error of
100 pixels isn’t that much; I have had errors over 1000 pixels and it
continues to try again (up to 10 times in my case) until is solves. It
sounds to me like you’ve only given SGP one chance to get within your
target error, but that’s hard to know without a log file.
I think I have my SGP set to try 5 attempts (I let all 5 attempts go through) and withing 50px error, which always worked flawlessly. I would increase the amount of attempts but the numbers do not change at all so I know it won’t solve no matter how many attempts I try. Unfortunately I am at work right now and I don’t even know where the find the logs since I never needed them (I’m very new to this) but I’ll find them when I get a chance.
I ended up simply clearing the Alignment Data and it appeared to resolve the Centering issue.
Honestly, I think the best solution is to simply keep the HC disconnected as you can run the mount via the iOptron Commander when imaging and only connect the HC when you use it visually. Aka This way I can keep my alignment data in the HC when I want it.
I get down to less than 15px every time now and sometimes as close as 4px.
I’ve been seeing the same problem with my G-11 and Pinpoint for several months. Used to get within 5 pixels. Now can’t get closer than 900 or so (it solves OK). Does the accuracy of the pointing model in the mount have anything to do with this? I haven’t been adjusting that thinking that Center on Target takes care of all mount issues.
SGP knows there’s an error. But it just doesn’t move the mount to correct it. Haven’t tried yet with 2.5.0.23.
We’d need logs to validate that. In most cases we’ve seen SGP issues the slew and the mount doesn’t move or can’t move to get there because of backlash.
In my case and I think others - the problem isn’t due to backlash but the fact that the mount doesn’t land at the ra/dec values it is told to - exactly. If the problem were only backlash, the final ra/dec values would match exactly - but the true pointing location would be off. In my case, and I believe others with non high-end mounts, the final ra/dec values are not exactly the intended ones - and there is no backlash issue at all because the mount does the final slew in the same direction always.
I will try centering with my cge-pro again - but I have already described this in detail. It isn’t trivial to get a mount to land at the exact ra/dec at the exact time - and there may be an inherent error of 1’ or so that has little impact on normal usage - but would make the sync/slew centering approach of sgp simply not work well at all. After one sync/slew it is about as close as it ever will be and it won’t get closer.
No backlash involved at all.
I have suggested two workarounds: nudging for the final centering, and having sgp keep track of the sync offset and correcting for it in sgp rather than syncing the mount. As long as the “miss” is repeatable, the latter should work well and has many advantages since it doesn’t alter the mount model.
From the logs above - I don’t see a log after the slew and before the sync - where you report where the mount thinks it is after the slew. That would be key diagnostic information - because otherwise you are just assuming it went to where it was told to go.
I will try this again when I have a chance, with cge-pro. But if that info isn’t in the logs - I recommend adding it. Otherwise the value will be stomped on by the sync. So I would need to do a separate test to tell the mount to go somewhere - and then query where it thinks it actually landed - based on its own ra/dec values. If those numbers are different - that is a fundamental limitation to accuracy of the sync/slew approach.
I’m back here with a few more attempts on this. Last year I would drag everything to my backyard on a JMI Wheeley Bar, fire SGP and go to bed and everything would work fine. Now I have an observatory, my mount is permanently set up, etc and absolutely nothing works. I am ready to throw the towel. Last week I got 1 night of successful attempts then last night when I was trying with a new telescope (and again with the other one that gave me successful attempts) everything went back to what it was, usually over a thousand px error when before I would never go over 5px and it would solve on the first attempt. I don’t know what is going on, I tried everything I could think of including uninstalling and installing everything and I still keep getting the same errors. I am attaching a couple of logs (I tried all night so I had many logs and I apologize if some settings are different, it’s all because I was trying all that I could think of). So here is what I noticed. I align my mount and tell it to go to a target (Andromeda for example), centered spot on, move the mount and tell it to Center on Andromeda again and I can see how off it is. I tried removing the hand controller like suggested, tried clearing my previous alignment, etc. Any help will be appreciated.
