I’m not able to center better than 70 px or so with G11 and 1422 mm FL. Please see attached images. Is this expected or should I try some different setup parameters? Any help is appreciated. I’m preparing to do a large Mosaic.
This looks like the same problem I have with cge-pro. The slew does not land exactly where it is told to and there is always a well defined offset that is not removed by syncing/slewing.
Two solutions are to use nudge for the final motions - or to figure out the slew error and correct for it in the subsequent goto. As long as it is repeatable - and it is for you and me - it should allow much greater accuracy - and you don’t need to sync anymore.
I have already entered it as a feature request.
I think for now the only workaround is to increase the allowed error in the centering operation.
Frank
I’m using a G11 with a scope at 920mm or 1200mm, and plate solving usual
takes 2-3 attempts to get to single digit accuracy. How tight is your
gear mesh, is there a lot of play in your mount? Also make sure that
your epoch (J2000 or Jnow) is set consistently across your programs and
equipment. Sorry I don’t have more to offer, I just know that with my
G11 I get very accurate results with just 1-3 plate solves.
Thank you Joel. Three questions if you don’t mind…
- My G11 has “Gemini expects J2000 co-ordinates” unchecked. Do I need to
change that or something else in SGP? - Is there a reference on how to check and adjust the G11 gear mesh? I
just assumed that was taken care of by the mount design. - With PHD2 I find RA tracks very tight while Dec moves around a lot,
especially when the scope is pointing up with no load on Dec. See the graph
below. Maybe this is related?
- In my experience setting the “Gemini expects J2000 coordinates” isn’t
really necessary. In fact there is a small bug with that particular
setting in the ascom driver that is being dicussed on one of the yahoo
Gemini user groups. Don’t worry about that for now. - One way to judge your gear mesh is to feel the movement at the end of
the counterweight shaft. By hand you should be able to oscillate the
counterweight shaft by a couple of mm. More movement than that would
signal that your mesh is a little loose. But again, I wouldn’t worry about
that for now unless it’s really bad. - Your guide graph doesn’t look too bad. Neither then Dec nor RA RMS
error is bad. The RA oscillation is a bit high.
I started to get better results when I put a counterweight on the mount so
that the the mount was always east biased. This places the load always on
one side of the gear teeth and should cut down on the RA oscillation. What
I did was tie a cable on the DEC shaft up by the RA block, then looped the
acround the west sid of the mount, looping over the top of the RA block and
down the east side of the mount, and put a 4lb weight on the end of the
cable. This always pulls the mount on the east side.
What is your guide exposure? It should be 3-5sec.
Another thing I have found critical is the settle time of the mount. I
have found it necessary to have the mount settle for 20 sec before the next
plate solve image is taken, otherwise the plate solve results will be off.
So under the control panel/Telescope tab, set your mount settle time to
20sec. Your mileage may vary.
Try the mount settle time and see if that improves your plate solve
accuracy.
Thanks Joel. Trying to stay on topic, just two more questions…
- Could you please share an image of what you did with the 4 lb weight?
- My guide exposure is 0.5 sec. I found if I go more, the Dec tends to
drift down too much. I believe I’m polar aligned well, so cannot explain
the downward drift.
Here’s my setup…
Here’s a link to a pic. This isn’t mine but it’s what I did.
http://www.astrobin.com/102020/B/
0.5sec is much too short for guide exposures. Your just chasing seeing
conditions, which may help explain some of the high RA oscillation. I
really don’t know why the dec would drift with 3sec exposures. I keep
the dec slightly off balance to make it nose heavy.
Thanks Joe. Went to 10 second settling time, 3-sec guiding and did drift
alignment. Much better results now with centering under 50 pixels.
Looks like my centering error is back. G11 just seems to stop about 200
pixels away in either RA or Dec. Attached is the center results and log
files. Any thoughts on what’s happening here?
I have the latest version of SGP and am using Pinpoint with UCAC4.
On Sun, Sep 6, 2015 at 10:23 PM, Kurious George curiousgmonk@gmail.com
wrote:
CenterError.txt (33.7 KB)
Rather difficult to say, but my guess would be backlash. It looks like your max error is always in RA. So my guess would be that you need to keep things a little more east biased to keep the gears engaged.
Thanks,
Jared
Ok, will try that. Thanks!