Jump to content

  •  

CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.

Photo

Multi-point GoTo alignment software for Tak EM-200?

This topic has been archived. This means that you cannot reply to this topic.
64 replies to this topic

#26 syscore

syscore

    Aurora

  • *****
  • Posts: 4,711
  • Joined: 07 Sep 2015

Posted 21 October 2016 - 09:31 PM

In my case, since I will be using a Losmandy saddle, it will go between the round dec head (with the two bolt holes) and the saddle. I will try to make the saddle more orthogonal.


Edited by syscore, 21 October 2016 - 09:33 PM.


#27 syscore

syscore

    Aurora

  • *****
  • Posts: 4,711
  • Joined: 07 Sep 2015

Posted 22 October 2016 - 08:52 PM

So after a couple hours with an EM-200 Temma 2 Jr, which is the same as any EM-200, except slower, I polar aligned and then manually moved the scope (clutches unlocked) to the area I was going to be in, locked the clutches, took a 10 second (bin 2x2) image, plate solved it, synced to plate solved image, and my GOTOs were great in that area (+/- 20 degrees). I haven't yet done any long GOTOs. Considering how long it takes to slew, I am actually ok with moving long distances manually right now, but I will try a long slew later and see what kind of cone error there is. I also used the polar alignment scope to polar align, and then measured with PemPro and was off 5.9' in az and 0.0' in alt. I am very impressed with the polar alignment scope. However, if I couldn't see Polaris, the alt/az adjusters are the smoothest I have ever used, and I could get there pretty fast with just PemPro or PHD2. I like PemPro's drift align better.



#28 johngwheeler

johngwheeler

    Surveyor 1

  • -----
  • topic starter
  • Posts: 1,712
  • Joined: 13 Jul 2015

Posted 22 October 2016 - 11:53 PM

So after a couple hours with an EM-200 Temma 2 Jr, which is the same as any EM-200, except slower, I polar aligned and then manually moved the scope (clutches unlocked) to the area I was going to be in, locked the clutches, took a 10 second (bin 2x2) image, plate solved it, synced to plate solved image, and my GOTOs were great in that area (+/- 20 degrees). I haven't yet done any long GOTOs. Considering how long it takes to slew, I am actually ok with moving long distances manually right now, but I will try a long slew later and see what kind of cone error there is. I also used the polar alignment scope to polar align, and then measured with PemPro and was off 5.9' in az and 0.0' in alt. I am very impressed with the polar alignment scope. However, if I couldn't see Polaris, the alt/az adjusters are the smoothest I have ever used, and I could get there pretty fast with just PemPro or PHD2. I like PemPro's drift align better.

Sounds like a good result. What software are you using the plate-solving & syncing?

 

Mine seems to have a goto accuracy within about 0.5-1º, i.e. using a 30mm, 70º FOV eyepiece on a C8 (c. 1º FoV), lands within the FoV most of the time. However, it normally misses the FoV using my illuminated reticle eyepiece (12mm with an AFOV of probably about 50 degrees, so about 0.3º FOV). I'll be trying to get multi-point alignment working with Astromist to see if this improves matters.

 

I found the PA scope rather fiddly here in the southern hemisphere, partly because the reticle illumination on my used mount is faulty, so I just try to shine red light into the PA scope to see the reticle. You also have to apply a correction table to invert the time & date values, so it requires an extra lookup.

 

I found the Azimuth knobs to be OK when the mount attachment knob is loosened slightly, but very stiff when the mount is cinched-up tighlty to the tripod and tightening up the mount does shift the alignment slightly, so I'm curious whether you found the same problem. Does your mount have any teflon bearing or similar between the mount and the tripod? I might cut up a plastic milk bottle to make some bearings. Seems like a funny solution on a mount that costs this much!

 

Thanks,

 

John.



#29 syscore

syscore

    Aurora

  • *****
  • Posts: 4,711
  • Joined: 07 Sep 2015

Posted 23 October 2016 - 12:52 AM

I had the bottom screw tight, but not clinched tight. I found the az adjusters much smoother and stronger than my CGEM and CGE Pro adjusters. I will give it a try with it clinched tight, but I don't think that is your main issue. You are probably having issues trying to use the polar scope in the south and with the light messed up. The light works well on mine and the reticle is clean. I did have to look up a star chart to see what the stars near Polaris looked like. I could easily see Polaris with my eye, but when I looked into the polar scope, there were more stars than I expected. It has also been awhile since I looked into a polar scope, but next time I think I could find Polaris on my own, in the polar scope I mean. Not withstanding your issues, I was very impressed with the accuracy. Previously, I thought 1/2 a degree was good with a polar scope, but to get less than 5' is good enough for AP. 

 

My GOTOs were landing within 5 arc-min of the target, and that is slewing from M31 to M45 to M77. About 25 degrees of slew.

 

I am using MaximDL, which uses PinPoint to plate solve. I manually point in the general direction and lock the clutches. Then I click the "Search Sky Online" button, which does a blind solve against Astrometry.net. There is no way PinPoint can solve it without knowing RA/DEC yet. But after the blind solve, I then click the normal (PinPoint) solve, and it solves it easily. I then sync, and am all good to go. I would say that I am satisfied with the GOTO results at this point and as long as the PA doesn't take too long, I am fine with the whole start up procedure. Even the slow slewing is ok actually, although I won't be jumping all over the sky like I do with my CGE Pro.

 

However, I am having some issues with PE and backlash on my EM-200.



#30 syscore

syscore

    Aurora

  • *****
  • Posts: 4,711
  • Joined: 07 Sep 2015

Posted 23 October 2016 - 12:54 AM

This was what I found in PemPro, after using the polar alignment scope...

 

azimuth.jpg

 

And after adjusting azimuth, this was the altitude error...

 

altitude.jpg


Edited by syscore, 23 October 2016 - 12:55 AM.


#31 syscore

syscore

    Aurora

  • *****
  • Posts: 4,711
  • Joined: 07 Sep 2015

Posted 23 October 2016 - 01:08 AM

When you use a one-star pointing model, PA is critical, but if you can get the PA to 5' or less, then you should be fine, for GOTOs and for AP.

 

"You also have to apply a correction table to invert the time & date values, so it requires an extra lookup."

 

What is the correction?

 

Here, I set the longitude scale to +6 deg offset (which is my location). Center the bubble level, then rotate the reticle calendar/day to align with the time. And then center Polaris.

 

How does it differ in the south?



#32 syscore

syscore

    Aurora

  • *****
  • Posts: 4,711
  • Joined: 07 Sep 2015

Posted 23 October 2016 - 02:45 AM

After M45 crossed the meridian, I did a flip (wow, that took a long time). The first image is where I was on M45 before the flip, and the second after the flip. About 1/2 deg off.

 

M45_East.jpg

 

M45_West.jpg

 

 



#33 orlyandico

orlyandico

    Voyager 1

  • *****
  • Posts: 10,014
  • Joined: 10 Aug 2009

Posted 23 October 2016 - 09:00 AM

There's an app called PolarFinder which has reticles for Tak, iOptron, Skywatcher, etc. etc. etc.

 

It costs a buck.

 

https://play.google....larfinder&hl=en

 

Not sure if available for iOS (I'm an android user).

 

Have never had a chance to use it thanks to not being able to see Polaris.  Glad to know the vaunted Tak polar scope is that accurate. I've read so much about it but haven't seen hard numbers till now.



#34 syscore

syscore

    Aurora

  • *****
  • Posts: 4,711
  • Joined: 07 Sep 2015

Posted 23 October 2016 - 09:45 AM

Here is a finished shot of M45. 70 x 120 seconds (2.3 hours)...

 

 

get.jpg


Edited by syscore, 23 October 2016 - 09:46 AM.


#35 orlyandico

orlyandico

    Voyager 1

  • *****
  • Posts: 10,014
  • Joined: 10 Aug 2009

Posted 23 October 2016 - 11:38 AM

looks like it's guided.. 120 seconds should be possible unguided with an EM-200...



#36 syscore

syscore

    Aurora

  • *****
  • Posts: 4,711
  • Joined: 07 Sep 2015

Posted 23 October 2016 - 11:49 AM

looks like it's guided.. 120 seconds should be possible unguided with an EM-200...

What does "guided" look like? 

 

I am starting another thread, so as not to get off John's topic, but yes, I guided last night, but I agree that an EM-200 can do 120s unguided at that focal length pretty easily.



#37 Ken Sturrock

Ken Sturrock

    Cardinal Ximenez

  • *****
  • Posts: 13,232
  • Joined: 26 Sep 2009

Posted 23 October 2016 - 01:18 PM

You can see that the center of the first Closed-Loop-Slew image is off by a little over 22 arc minutes from M1's actual position.


So, in this case, the slew was off by a little over 23 arc minutes.


Wow, 22 arc-minutes is quite a bit.


The first image is where I was on M45 before the flip, and the second after the flip. About 1/2 deg off.


Sounds about right. :flowerred:

#38 syscore

syscore

    Aurora

  • *****
  • Posts: 4,711
  • Joined: 07 Sep 2015

Posted 23 October 2016 - 05:32 PM

You were dead on.:)

 

I'll be honest, you had me scared, but I plate solve and sync about as much with my CGE Pro, just to get even more accuracy. GOTO issues did not come up with me last night, and that 1/2 degree error would not bother me at all. Even though the CGE Pro would be closer, I would still plate solve and sync it after a meridian flip.

 

I think John's issue is that he is doing this with an eyepiece. He can't just click a button and make things perfect again. And that would bother me as well, if I was using it visually and not seeing the targets in my FOV.

 

What bothered me most was having to keep checking on the mount to make sure I wasn't crashing into a tripod leg. Having moved to a pier some time ago, I am not used to that. The meridian flip wasn't an issue. I still go out an watch my CGE Pro flip, in case of cable snags. The only difference with the EM-200 is that I can make a sandwich and have some milk and then go outside just as it is clearing the half way point.:)



#39 syscore

syscore

    Aurora

  • *****
  • Posts: 4,711
  • Joined: 07 Sep 2015

Posted 23 October 2016 - 06:11 PM

 

looks like it's guided.. 120 seconds should be possible unguided with an EM-200...

What does "guided" look like? 

 

I am starting another thread, so as not to get off John's topic, but yes, I guided last night, but I agree that an EM-200 can do 120s unguided at that focal length pretty easily.

 

But a real test of unguidedness is going to be 2800 mm, not 384 mm. :)

 

EM_200_04.jpg

 

If the clouds cooperate. Stripped down like it is, the EdgeHD 11", T-Adapter and Camera is about 35 lbs. I suspect it will guide well, but we will see. I had to add one of my Celestron weights to get it balanced. 3 Takahashi weights would work as well I think, but be further down. Last night I did have to play with balance to get the results I did. 


Edited by syscore, 23 October 2016 - 06:46 PM.


#40 johngwheeler

johngwheeler

    Surveyor 1

  • -----
  • topic starter
  • Posts: 1,712
  • Joined: 13 Jul 2015

Posted 23 October 2016 - 06:25 PM

looks like it's guided.. 120 seconds should be possible unguided with an EM-200...

What does "guided" look like? 
 
I am starting another thread, so as not to get off John's topic, but yes, I guided last night, but I agree that an EM-200 can do 120s unguided at that focal length pretty easily.

But a real test of unguideness is going to be 2800 mm, not 384 mm. :)
 
attachicon.gifEM_200_04.jpg
 
If the clouds cooperate. Stripped down like it is, the EdgeHD 11", T-Adapter and Camera is about 35 lbs. I suspect it will guide well, but we will see. I had to add one of my Celestron weights to get it balanced. 3 Takahashi weights would work as well I think, but be further down. Last night I did have to play with balance to get the results I did.


Beautiful rig! Do you find that the C11 feels "comfortable" on the EM-200? The guy who sold me mine said that he though his RC10 was too much for it, but there seem to be plenty of folks using C11s.

On the subject of gotos, I'm wondering whether there is other software that can do "closed loop" gotos for visual use via plate-solving, apart from The Sky X? I use SGP for imaging, but have never though of using it as a visual GoTo tool. Could I use my guide scope / camera for purpose, or would it need a wider-field camera for the plate solves?

John

#41 syscore

syscore

    Aurora

  • *****
  • Posts: 4,711
  • Joined: 07 Sep 2015

Posted 23 October 2016 - 06:44 PM

I don't know yet how it will handle the 11" (for imaging), but physically, I didn't note anything. And those 2x4's for legs, that I keep making fun of, don't even budge. When I put the 11" on my CGEM, it feels overloaded. 

 

A finder scope would have too wide a FOV I think for plate solving software. I say this because I accidentally put 80 mm in for FL (instead of aperture) and PinPoint said that it required too many stars. But a guide scope should be fine. You can try that with your rig now, just select the guide camera as main camera.

 

Starsense does the exact same thing, but it is built for wide fields of view. Maybe you can set options on the common plate solvers we use to better handle wide fields. 



#42 syscore

syscore

    Aurora

  • *****
  • Posts: 4,711
  • Joined: 07 Sep 2015

Posted 23 October 2016 - 11:13 PM

Well, the clouds are not letting up. I managed to get 3 frames at 60 seconds unguided...

 

get.jpg

 

The rest were either (mostly) ruined due to clouds and looked like this ...

 

Clouds.jpg

 

Or they were ruined due to tracking errors, and there is a bad spot on the worm that looks like this...

 

Worm.jpg

 

If it wasn't cloudy, then I think 50% of the subs would have survived. 25% would be iffy, the other 25% definitely not, and a few of those would have landed on bad worm spots. You can expect at least 15% of 60 second subs to be ruined by the bad worm spots, because the worm cycle is 8 minutes. I am not sure what the cost is for a Takahashi worm.

 

However, those spots didn't affect the 120s guided subs (of M45). None of those got thrown away. But that was at less than 1/7th the focal length.



#43 orlyandico

orlyandico

    Voyager 1

  • *****
  • Posts: 10,014
  • Joined: 10 Aug 2009

Posted 23 October 2016 - 11:52 PM

is the seller aware of that bad spot? that's a pretty bad spot!

 

but it could simply be that the worm end-carriers aren't properly torqued in (or are torqued too tight...) I seem to recall reading about something like this on the Uncensored Tak Group on Yahoo.

 

one other possibility is corrosion or failed lubrication/failed bearing seals on the worm bearings, which causes them to become "gritty" which then again causes periodic errors.

 

I wouldn't be too happy with an EM-200 with that level of RA error/bad spot, but the seller has offered a pretty low price.. maybe it's worth the risk at $2K.



#44 syscore

syscore

    Aurora

  • *****
  • Posts: 4,711
  • Joined: 07 Sep 2015

Posted 24 October 2016 - 12:07 AM

Without that spot the PE is about 8" p-p. I am going to check into possible reasons/cures. But even 8" p-p is not low enough to go more than 60 seconds at 2800 mm. Actually, it's the pixel scale, which with my camera is 0.40" / px. When I was shooting at 384 mm it was 2.88" / px. Now, if EM-200's had PEC, they could easily do 3 to 5 minutes unguided at 2800 mm. The PE is very smooth and constant. PEC would take it to 1" or below quite easily. Except for the bump of course.:) This is the first premium mount I have tested, and the consistency and smoothness of the PE looks a lot different than you see with no premium mounts. What is even more amazing is that it doesn't use worm blocks. It is built (but not machined) just like a CGEM. But it is tight.


Edited by syscore, 24 October 2016 - 12:13 AM.


#45 syscore

syscore

    Aurora

  • *****
  • Posts: 4,711
  • Joined: 07 Sep 2015

Posted 24 October 2016 - 08:16 AM

I had a lot of difficulty guiding the 11", but the weather kept interrupting. Balance seems to be critical with this mount, even with the smaller scope. There was a small break in the clouds and I managed to get 17 minutes of 15 sec subs (unguided) of M42. I thought most of the night would be guided imaging, not unguided, so I didn't drift align. This wasn't calibrated, just registered and stacked.

 

M42.jpg  


Edited by syscore, 24 October 2016 - 08:18 AM.


#46 orlyandico

orlyandico

    Voyager 1

  • *****
  • Posts: 10,014
  • Joined: 10 Aug 2009

Posted 24 October 2016 - 11:30 AM

Surprised that you find balance critical.  I found (on my EM-11) that balance was pretty indifferent.  But then again I only put an 80mm on it (albeit a very heavy for its size 80mm).

 

That M42 must be with the C11?  honestly I think an EM-200 is overloaded with a C11...



#47 johngwheeler

johngwheeler

    Surveyor 1

  • -----
  • topic starter
  • Posts: 1,712
  • Joined: 13 Jul 2015

Posted 24 October 2016 - 03:57 PM

Surprised that you find balance critical.  I found (on my EM-11) that balance was pretty indifferent.  But then again I only put an 80mm on it (albeit a very heavy for its size 80mm).

 

That M42 must be with the C11?  honestly I think an EM-200 is overloaded with a C11...

 

I don't know whether the balance makes a huge difference to tracking accuracy, but the EM-200 is an extremely sensitive mount to any imbalance due to its extremely low friction. I've found that pretty much any scope will have significant balance shifts at different points is often not symettrical due to the offsets from finders, accessorires and even the control electronics on the mount itself - if you loosen the clutches, the mount almost never "stays still",. I've learned to live with this and enjoy the fact that I can move a heavy scope around with finger-tip pressure. I do worry about letting it go though! Generally the balance is good enough to prevent the scope hitting the mount and it finds its equilibrium, but even a small movement can be a bit alarming. 

 

With my AVX I could loosen the clutches and nothing moves because of the enormous friction in the bearings....although this could hardly be considered a good quality!



#48 johngwheeler

johngwheeler

    Surveyor 1

  • -----
  • topic starter
  • Posts: 1,712
  • Joined: 13 Jul 2015

Posted 24 October 2016 - 04:51 PM

Here's an update to my GoTo accuracy trials.

 

I bought Astromist for the iPad ($23) which supposedly supports 2-star and n-star alignments with Takahahsi mounts. I have to report that it has not proven to be a satisfactory solution (so far...). Whilst Astromist has some interesting features, its design is antiquated, having been originally developed for Windows mobile OS in the early 2000s. I could forgive this if it actually worked for alignment...something I have been unable to accomplish. I may be doing something wrong (following the rather brief and imprecise instructions), but it simply doesn't appear to work as advertised and doesn't seem to update the mount position data when the scope is aligned with the alignment stars. It frequently disconnects from the mount and requires a base-line alignment to the zenith (similar to many DSCs), but I've done this over and and over, and it still gets it completely wrong. I understand the process having done this with other mounts and software such as Align Master, but Astromist is completely unintuitive....so back to SkySafari and one-star alignment for now. I'll post my experience to this and other forums to see if anyone can enlighten me. I'm happy to eat humble-pie and feel stupid in return for a solution :-)

 

Back to SkySafari (what a breath of fresh air after fighting with Astromist!), I found that accuracy with shorter slews (up to about 60 degrees) is probably within about 0.5 degree. This is within the FoV of my 22-24mm EPs, so is at least getting me "in the ballpark". I can probably tweak my PA to improve this and I may have a bit cone error, both of which could be rectified. It does often require me to change EPs for the goto though, and would be insufficent for centering on a CCD. Maybe this is "as good as it gets" for the one-star alignment, and any improvement will require n-star alignment software or closed-loop gotos using plate solving.

 

I'm don't know whether there is any planetarium software that includes an n-star goto alignment routine without requiring a camera and plate-solving, apart from The Sky X + T-Point (over $400). I may just have to live with the current accuracy for visual-only use, and use a plate-solving option for imaging or if I require greater goto accuracy.

 

John



#49 orlyandico

orlyandico

    Voyager 1

  • *****
  • Posts: 10,014
  • Joined: 10 Aug 2009

Posted 24 October 2016 - 05:03 PM

1) Yes the EM-11 is also quite frictionless.  When I say "indifferent to balance" I meant that I was able to guide successfully even with a lot of imbalance (due to not having enough counterweights... so long as the mount was biased east-heavy) but yes if you have enough weights, balancing is pretty easy because the mount spins so freely.

 

2) You can probably get a refund for Astromist.  I was looking at it but the antiquated nature turned me off.  Now as it seems multi-star alignment doesn't work (I was planning to use it for polar alignment, since I can't see Polaris) then even less reason for me to consider it.

 

3) I was planning to build my own app for multi-point alignment and modeling (poor man's TPoint) because the Mach1 has exactly the same problem as the EM200. The idea was to command slews to specific bright stars, wait for the end-user to center the stars, then fetch the mount-reported Alt/Az (and store along with the correct/known RA/Dec of the stars).  Having these (Alt/Az, RA/Dec) pairs - many of them - would allow a pointing model to be created.

 

Problem is I'm terrible with Windows programming.  I could put together such an app in a weekend on Linux, but nobody uses Linux for astronomy. My aspiration was to build an Android app, but getting wireless telescope control over bluetooth was too hard.

 

There's been on and off talk of multi-star alignment on the Sky Safari forums for years. It still hasn't materialized.



#50 johngwheeler

johngwheeler

    Surveyor 1

  • -----
  • topic starter
  • Posts: 1,712
  • Joined: 13 Jul 2015

Posted 24 October 2016 - 06:03 PM

1) Yes the EM-11 is also quite frictionless.  When I say "indifferent to balance" I meant that I was able to guide successfully even with a lot of imbalance (due to not having enough counterweights... so long as the mount was biased east-heavy) but yes if you have enough weights, balancing is pretty easy because the mount spins so freely.

 

2) You can probably get a refund for Astromist.  I was looking at it but the antiquated nature turned me off.  Now as it seems multi-star alignment doesn't work (I was planning to use it for polar alignment, since I can't see Polaris) then even less reason for me to consider it.

 

3) I was planning to build my own app for multi-point alignment and modeling (poor man's TPoint) because the Mach1 has exactly the same problem as the EM200. The idea was to command slews to specific bright stars, wait for the end-user to center the stars, then fetch the mount-reported Alt/Az (and store along with the correct/known RA/Dec of the stars).  Having these (Alt/Az, RA/Dec) pairs - many of them - would allow a pointing model to be created.

 

Problem is I'm terrible with Windows programming.  I could put together such an app in a weekend on Linux, but nobody uses Linux for astronomy. My aspiration was to build an Android app, but getting wireless telescope control over bluetooth was too hard.

 

There's been on and off talk of multi-star alignment on the Sky Safari forums for years. It still hasn't materialized.

Thanks for the feedback. I will try to get some user input for Astromist (or write to the author), and give it one last chance. However, I'm not really sure I would want to use this as a planetarium because the it's much less intuitive and visually appealing than Sky Safari. It the 2-star alignment is actually buggy, then I'll ask for a refund. Otherwise, I'm prepared to wear the cost. A bit like a wine that you order in a restaurant which you dislike, but is not actually defective....

 

I'm impressed that you could write an alignment routing in a weekend! Astromist is supposed to do exactly what you describe. You select a star from the sky view, which correctly enters the catalog RA/Dec coords, but when I move the mount, it doesn't pick up the mount co-ordinates correctly. It has a "read mount" button that is supposed to get the encoder values, but doing this on two stars (moving the scope between them) doesn't actually change the value read, so it appears that it is not correctly return the mount position.




CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.


Recent Topics






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics