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

Dumb question about DEC backlash on CGEM.

  • Please log in to reply
12 replies to this topic

#1 Borodog

Borodog

    Soyuz

  • *****
  • topic starter
  • Posts: 3,784
  • Joined: 26 Oct 2020

Posted 14 January 2022 - 01:58 PM

According to the PHD2 Guiding Assistant, I have a LOT of backlash in DEC, as in > 9000ms. It suggests I might want to guide in only one direction. However, I never seem to have much trouble with my DEC guiding. I'd say with good seeing and no wind it's typically around 0.5" (RA is another story). It doesn't seem to care that the DEC backlash is large.  This is with DEC set to Auto, not one direction as it suggests. Can anyone explain to me how the DEC backlash can take 9s to take up but my DEC guiding is good anyway? And if it's good, should I care about the large backlash? It seems overly complicated to correct and I am not particularly mechanically inclined. I much prefer to be mechanically reclined.


Edited by Borodog, 14 January 2022 - 01:59 PM.


#2 bobzeq25

bobzeq25

    ISS

  • *****
  • Posts: 27,006
  • Joined: 27 Oct 2014

Posted 14 January 2022 - 02:16 PM

The guiding assistant backlash measurement is often questioned, it's not that precise. 

 

That said, 9000ms is _far_ beyond the usual range.  Something is going on.

 

When the guiding direction in DEC changes, does it take several corrections to get back on track?  Sounds like not.



#3 rgsalinger

rgsalinger

    Voyager 1

  • *****
  • Moderators
  • Posts: 10,249
  • Joined: 19 Feb 2007
  • Loc: Carlsbad Ca

Posted 14 January 2022 - 02:18 PM

I don't trust the guiding assistant's backlash measurement to be a mount issue because that's not what it's measuring. It's measuring the entire system. So, to determine if the mount is the problem,  I strongly suggest that you actually do a star test using an eyepiece and see what happens. My bet is that it shows less than 1 second of backlash. Or if you insist, use the star cross test in PHD rather than the assistant. I don't trust that either.

 

When I test something I don't really want millions of lines of code between me and the result if I didn't write the code and/or I don't need to use the code to test. Very old fashioned but.............

 

Now, there's one other possibility which is that your DEC guiding never reverses. If that's the case then you will never see the backlash. That can happen I think if the polar alignment error causes sufficient mistracking to over come the mechanical and seeing errors in the mount. Just a theory. 

 

Rgrds-Ross


  • bobzeq25 and Borodog like this

#4 Lkealey

Lkealey

    Lift Off

  • -----
  • Posts: 16
  • Joined: 24 Dec 2021
  • Loc: Sugar Land, Texas

Posted 14 January 2022 - 02:35 PM

I would have to see your setup to competently make a good suggestion… having said that…

 

if the Moments of Inertia are large, you’ll have more backlash in general. By larger moments of inertia, what I mean is that much of the mass of the setup is farther from the center of mass. For example, if you have one counterweight that is all the way out on  an extension bar, you will have a much larger moment of Inertia than if you pulled your counterweight up closer and added another smaller counterweight.  The torque varies with the mass and distance (the counterweight is there to balance the torque); whereas, the inertia varies with the mass and the SQUARE of the distance. Thus the inertia increases exponentially as you move the weight out. Inertia is important because it represents the resistance to change in momentum. (Velocity - be it linear ear rotational). Depending upon how the mass is distributed, it could affect backlash in one or both directions.

 

the example often used to describe the properties of Inertia is the figure skater who starts to spin. The skater enters the spin with a certain amount of rotational energy, with their arms spread wide. As the skater pulls their arms in close to their body, they start spinning much faster…. Without any force being applied (I.e. constant energy). 
 

so it takes a lot less energy and force to turn the telescope, and stop it If the mass is distributed more closely to the center - like when the skater has their arms held close… 

 

when I read your post, that was the first thought that came to mind. I hope I explained it well enough without being too technical…. 
 

thus, with a motorized mount, it is usually better to add an extra counterweight than to add an extension bar.  Less stress on the motors when starting and stopping the movement of the scope.

 

hope this helps!

 

-Kealey



#5 Borodog

Borodog

    Soyuz

  • *****
  • topic starter
  • Posts: 3,784
  • Joined: 26 Oct 2020

Posted 14 January 2022 - 03:08 PM

The guiding assistant backlash measurement is often questioned, it's not that precise. 

 

That said, 9000ms is _far_ beyond the usual range.  Something is going on.

 

When the guiding direction in DEC changes, does it take several corrections to get back on track?  Sounds like not.

It doesn't seem to, generally. Now, there have been occasions where something crazy seems to happen but I have generally been able to chalk that up up things like wind, forgetting to level the tripod, a light cable drag, etc. I generally seem to have fairly round stars, although my RA guiding can be 2X or more times my DEC guiding sometimes so they can get a bit eggy sometimes. And my setup is pretty unforgiving at ~1300mm, image scale of 0.38"/pixel, and guide scale:imaging to imaging scale of 13X (I can't prove it but it seems like binning in post rather that during capture seems to provide better alignment, so I leave it).

 

The payload is a 1985 C8 + reducer + dual speed Crayford + ASI183MC + ADM Losmandy D rail for mounting + GLP as finder + 30mm mini guidescope + SV305 as guide camera + dew shield + Reflectix jacket. I haven't weighed it all but I would guess it's something like 20 pounds. I use a single 17 lb counterweight on the bar. It's not especially far down the bar. About half way.



#6 Borodog

Borodog

    Soyuz

  • *****
  • topic starter
  • Posts: 3,784
  • Joined: 26 Oct 2020

Posted 14 January 2022 - 03:10 PM

Another thing to note is that when I use the hand controller or onscreen version via ASCOM to move the mount, it is very snappy north, south, and east. To the west it moves slowly and drifts several seconds before coming to rest, perhaps as long as 5 seconds.



#7 Borodog

Borodog

    Soyuz

  • *****
  • topic starter
  • Posts: 3,784
  • Joined: 26 Oct 2020

Posted 14 January 2022 - 03:15 PM

I don't trust the guiding assistant's backlash measurement to be a mount issue because that's not what it's measuring. It's measuring the entire system. So, to determine if the mount is the problem,  I strongly suggest that you actually do a star test using an eyepiece and see what happens. My bet is that it shows less than 1 second of backlash. Or if you insist, use the star cross test in PHD rather than the assistant. I don't trust that either.

 

When I test something I don't really want millions of lines of code between me and the result if I didn't write the code and/or I don't need to use the code to test. Very old fashioned but.............

 

Now, there's one other possibility which is that your DEC guiding never reverses. If that's the case then you will never see the backlash. That can happen I think if the polar alignment error causes sufficient mistracking to over come the mechanical and seeing errors in the mount. Just a theory. 

 

Rgrds-Ross

See my previous post; when I moved the mount north or south, there is no visible backlash at all. It's only when I move west that it moves sluggishly and drifts before stopping.



#8 rgsalinger

rgsalinger

    Voyager 1

  • *****
  • Moderators
  • Posts: 10,249
  • Joined: 19 Feb 2007
  • Loc: Carlsbad Ca

Posted 14 January 2022 - 05:54 PM

I would not try to guide a C8 with a 30mm guide scope even at 1.3 meters. If you have eggy stars, then I would chalk that up to not using an OAG.

 

Many moons ago I was getting very marginal guiding with my GSO RC8 and an OAG. So I bought the fanciest guides scope system and put my SBIG-STi in it. Guiding got even worse. Lesson learned. 

 

Rgrds-Ross



#9 Borodog

Borodog

    Soyuz

  • *****
  • topic starter
  • Posts: 3,784
  • Joined: 26 Oct 2020

Posted 14 January 2022 - 09:26 PM

Well, it's what I have for the moment. Also, my stars are usually round enough for me these days, unless there is wind or some other confounding factor. The real game changer there was discovering multi-star guiding. Before that I had consistently eggy stars, never round. Afterward, mostly round stars, occasionally a bit eggy. Oh, and ditching the Scope Rollers, unfortunately. I discovered they let the entire scope vibrate. PHD2 was actually chasing the vibration.

 

If it gives me guiding less than the seeing, and it typically does, that seems good enough for me.

 

gallery_346195_18307_282395.jpg



#10 Borodog

Borodog

    Soyuz

  • *****
  • topic starter
  • Posts: 3,784
  • Joined: 26 Oct 2020

Posted 14 January 2022 - 09:27 PM

This is an example from when I was using single star guiding while on the Scope Rollers.

 

gallery_346195_18307_1730.jpg



#11 rgsalinger

rgsalinger

    Voyager 1

  • *****
  • Moderators
  • Posts: 10,249
  • Joined: 19 Feb 2007
  • Loc: Carlsbad Ca

Posted 14 January 2022 - 11:17 PM

When I make that kind of comment it's about looking quantitatively at the raw data - measuring it. My view is that if you used an OAG rather than a guide scope with that configuration the raw data would be measurably better. I need to be more specific about that in the future I guess. Since the issue is mirror flop, guiding numbers can be terrific but the final raw data not so hot. The guiding system is working perfectly but the main imaging system is moving differentially. 

 

I dont really see how multi-star is going to eliiminate mirror flop. If it you can isolate that change from the rollers it would make a terrifc thread in imaging all on it's own. People don't like to use OAG's, so if there was a way to use a guide scope with an SCT. 

 

Rgrds-Ross


  • Borodog likes this

#12 Borodog

Borodog

    Soyuz

  • *****
  • topic starter
  • Posts: 3,784
  • Joined: 26 Oct 2020

Posted 15 January 2022 - 11:17 AM

When I make that kind of comment it's about looking quantitatively at the raw data - measuring it. My view is that if you used an OAG rather than a guide scope with that configuration the raw data would be measurably better. I need to be more specific about that in the future I guess. Since the issue is mirror flop, guiding numbers can be terrific but the final raw data not so hot. The guiding system is working perfectly but the main imaging system is moving differentially. 

 

I dont really see how multi-star is going to eliiminate mirror flop. If it you can isolate that change from the rollers it would make a terrifc thread in imaging all on it's own. People don't like to use OAG's, so if there was a way to use a guide scope with an SCT. 

 

Rgrds-Ross

Thanks; can you elaborate on the mirror flop issue? I know that can be a problem with focus when the scope transits over a large angle, but I don't understand how it can elongate stars during (relatively) short subs up to a few minutes.

 

My understanding was that the problem with guide scopes and SCTs is differential flexure. The 30mm mini guide scope seems pretty good about that in my particular setup; the guide scope is hard mounted to the Losmandy plate, is lightweight and doesn't use true mounting rings with skinny thumbscrews to hold it in place; it's pretty tightly contained and rigidly mounted. The main optical train is pretty light, not having a big filter wheel or a cooled camera, and the scope doesn't have a big finder scope hanging off the rear cell, just a small GLP.

 

If mirror flop is really an issue I'd love to understand it. 



#13 rgsalinger

rgsalinger

    Voyager 1

  • *****
  • Moderators
  • Posts: 10,249
  • Joined: 19 Feb 2007
  • Loc: Carlsbad Ca

Posted 15 January 2022 - 01:47 PM

OK. So there are no mirror locks on your OTA. As the vector of the pull of gravity changes, the mirror moves a bit in response. If you use short enough exposures or larger enough pixels, then you won't see the problem. As the exposures get longer, eventually you'll get 1 pixel of elongation. You can see if this affects you or not by just doing a set (on a moon night) of 1 - 15 minute guided exposures. That assumes that you're getting good guiding which you say that you are. I sometimes wonder if the new CMOS cameras with high QE which give us shorter exposures mean that my "knowledge" of the matter is slightlyh obsolete. 

 

Again, you can't judge any of this from the finished product. Take a look at this finished image taken recently with my TV127is that needs to be collimated. At normal screen size it looks terrific. It you were to see one of the subs, the corners look awful. Stacking can cure a lot of ills. 

 

Rgrds-Ross


  • Borodog likes this


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