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

Steppers and resolution

  • Please log in to reply
11 replies to this topic

#1 jessebear

jessebear

    Vostok 1

  • *****
  • topic starter
  • Posts: 188
  • Joined: 01 Oct 2019
  • Loc: Grand Junction, CO

Posted 12 November 2020 - 11:47 PM

Hello everyone,

I'm hoping this is the right place to ask this. I built a go-to dob with OnStep this summer and using their calculation sheet it has a resolution of 0.6 arc seconds on each axis. Altaz isn't great for imaging, so I built a small barn door tracker from some plans I found online. Now I want to build a poncet platform and try something bigger.

Using parts I've got around the shop I could get my poncet platform to a resolution of 1.95 arc seconds (200 step motor, 32 microsteps, 27:1 planetary, 8:1 gear reduction). I don't think this is nearly appropriate for imaging. Another $50 for a 200 step motor with 100:1 planetary gears gets me to 0.38 arc seconds, and $80 for a 400 step version gets me to 0.19 arc seconds. I'm not expecting fantastic results with an AD10 on a homemade poncet platform running either an ancient T2i or an SVBony 305 camera, I'm just looking to get my feet wet.

Is my assumption that 1.95 arc seconds is unusable correct? If so, is it worth it to go for the 0.19 arc seconds setup over the 0.38, or is that overkill for a dob on a poncet?

Thank you!

#2 galacticinsomnia

galacticinsomnia

    Ranger 4

  • *****
  • Posts: 303
  • Joined: 14 Aug 2020
  • Loc: Pacific Northwest - Oregon

Posted 18 November 2020 - 02:18 AM

Hello everyone,

I'm hoping this is the right place to ask this. I built a go-to dob with OnStep this summer and using their calculation sheet it has a resolution of 0.6 arc seconds on each axis. Altaz isn't great for imaging, so I built a small barn door tracker from some plans I found online. Now I want to build a poncet platform and try something bigger.

Using parts I've got around the shop I could get my poncet platform to a resolution of 1.95 arc seconds (200 step motor, 32 microsteps, 27:1 planetary, 8:1 gear reduction). I don't think this is nearly appropriate for imaging. Another $50 for a 200 step motor with 100:1 planetary gears gets me to 0.38 arc seconds, and $80 for a 400 step version gets me to 0.19 arc seconds. I'm not expecting fantastic results with an AD10 on a homemade poncet platform running either an ancient T2i or an SVBony 305 camera, I'm just looking to get my feet wet.

Is my assumption that 1.95 arc seconds is unusable correct? If so, is it worth it to go for the 0.19 arc seconds setup over the 0.38, or is that overkill for a dob on a poncet?

Thank you!

I may be incorrect, but it feels like your math is wrong somewhere.
But maybe mine is too.. lol

Most steppers
1 full step = 1.8 degrees
32 micro steps = 1.8 degrees
1 micro step = .05625 degrees or 15.625 arc seconds straight from the motor.
So you just need a 10 to 1 or better gear ratio to get you 1.5 arc seconds or better
My math may be wrong but I'm giving it a shot..

 

 

To determine your needs work out the highest resolution necessary for your best case scenario you wish to image with, or with what you have.

The rule is normally 1 to 2 arc seconds per pixel resolution.  I agree with this and shoot for 1 second, but to determine how well your images will come out, you can be all scientific or more exacting by figuring your angular resolution of your gear.  But it's not necessary if you just shoot for 1 to 2 arc seconds. :)

(Pixel Size (μm) of camera / Focal length of telescope) * 206.3 = angular resolution or arcseconds/pixels



#3 jessebear

jessebear

    Vostok 1

  • *****
  • topic starter
  • Posts: 188
  • Joined: 01 Oct 2019
  • Loc: Grand Junction, CO

Posted 18 November 2020 - 11:28 PM

Thank you for the info! You are completely right - the math doesn't check out on my first post for straight calculations of resolution. The figures I reported are from the calculation sheet of the OnStep program, which tries to give a 'real world' estimate of actual resolution when using microstepping (which can be imprecise).

I made another post about this in the ATM forum and got some guidance there as well. Probably my biggest issue is the image scale I'm working with on my available cameras. Basically it'll be no smaller than 0.9, so I either need to have pretty accurate tracking or get a camera with bigger pixels. More accurate tracking is, for now, more cost effective - so that's the road I think I'm headed down.

I have parts and a design ready that should be good for a calculated resolution of 0.03 arc seconds, and a 'real world' estimate of 0.38. I need to just build the thing instead of playing with designs.. I'm mulling one over now that would be fairly easy to make and would bring me down to a calculated resolution of 0.0005 arc seconds. Seems a bit of overkill for a homemade EQ platform though!
  • galacticinsomnia likes this

#4 galacticinsomnia

galacticinsomnia

    Ranger 4

  • *****
  • Posts: 303
  • Joined: 14 Aug 2020
  • Loc: Pacific Northwest - Oregon

Posted 20 November 2020 - 09:51 PM

Thank you for the info! You are completely right - the math doesn't check out on my first post for straight calculations of resolution. The figures I reported are from the calculation sheet of the OnStep program, which tries to give a 'real world' estimate of actual resolution when using microstepping (which can be imprecise).

I made another post about this in the ATM forum and got some guidance there as well. Probably my biggest issue is the image scale I'm working with on my available cameras. Basically it'll be no smaller than 0.9, so I either need to have pretty accurate tracking or get a camera with bigger pixels. More accurate tracking is, for now, more cost effective - so that's the road I think I'm headed down.

I have parts and a design ready that should be good for a calculated resolution of 0.03 arc seconds, and a 'real world' estimate of 0.38. I need to just build the thing instead of playing with designs.. I'm mulling one over now that would be fairly easy to make and would bring me down to a calculated resolution of 0.0005 arc seconds. Seems a bit of overkill for a homemade EQ platform though!

It would be pretty awesome to get that worked out.  Good luck on your project.  Lot of great information out there.
 



#5 klaussius

klaussius

    Viking 1

  • *****
  • Posts: 834
  • Joined: 02 Apr 2019
  • Loc: Buenos Aires

Posted 20 November 2020 - 10:06 PM

I may be incorrect, but it feels like your math is wrong somewhere.
But maybe mine is too.. lol

Most steppers
1 full step = 1.8 degrees
32 micro steps = 1.8 degrees
1 micro step = .05625 degrees or 15.625 arc seconds straight from the motor.
So you just need a 10 to 1 or better gear ratio to get you 1.5 arc seconds or better
My math may be wrong but I'm giving it a shot..

 

Um... 1 micro step = 1/32 degrees = 60/32 arc-minutes = 3600/32 = 112.5 arc-seconds.


  • galacticinsomnia likes this

#6 galacticinsomnia

galacticinsomnia

    Ranger 4

  • *****
  • Posts: 303
  • Joined: 14 Aug 2020
  • Loc: Pacific Northwest - Oregon

Posted 21 November 2020 - 03:50 AM

Um... 1 micro step = 1/32 degrees = 60/32 arc-minutes = 3600/32 = 112.5 arc-seconds.

If you are sure, I have to believe you.  :)  Though it still seems a bit off, even though I don't have the correct maths myself.  Usually pretty good at this stuff but the dementia gets the best of me sometimes.
Thank you for the input.

 



#7 genehunter29009

genehunter29009

    Explorer 1

  • *****
  • Posts: 66
  • Joined: 01 Apr 2017

Posted 21 November 2020 - 07:18 AM

I am working on a project myself where I want to have a mount that is completely automated. I have been looking at all alternatives because I do not want to spend 3k on a mount only to leave it outside all 24x7 365 days a year in the weather. So I started looking on ebay. I found not one but 2 older alt az robotic camera mounts for get this $100 plus shipping. They are rated at 60 lbs. All i am saying is look around and find something used perhaps. I have to write my own code to get this thing tracking but I assume you also already know how to do that or else you would not be playing with steppers which are not entirely easy for  a newb.

 

a few photos of my project. these things are built like a tank, no plastic parts whatsoever at all.

Attached Thumbnails

  • 122661950_3463876220322599_8354801928960363089_o.jpg
  • 122833542_3463876283655926_7534767601211341639_o.jpg
  • 122590777_3463876190322602_6357210653036882938_o.jpg


#8 kathyastro

kathyastro

    Vanguard

  • *****
  • Posts: 2,408
  • Joined: 23 Dec 2016
  • Loc: Nova Scotia

Posted 21 November 2020 - 08:59 AM

Math:

 

1 step = 1.8 degrees.

32 microsteps = 1.8 degrees.

1 microstep = 1.8 / 32 = 0.05625 degrees.

1 microstep = 0.05625 * 3600 = 202.5 arcseconds.

 

A 100:1 gearbox gives you 202.5 / 100 = 2.025 arcseconds per microstep.

 

Kids, don't try this at home.


Edited by kathyastro, 21 November 2020 - 09:02 AM.

  • galacticinsomnia likes this

#9 jessebear

jessebear

    Vostok 1

  • *****
  • topic starter
  • Posts: 188
  • Joined: 01 Oct 2019
  • Loc: Grand Junction, CO

Posted 21 November 2020 - 10:26 AM

Lots of good input here - thank you all! I've played around with several configurations and I'm still mulling a few things over but I've gotten some advice from the Arduino folks as well as OnStep. Though I can really push this design by either moving to a servo or just increasing gear reduction, I'm really limited to imperfections in the north bearing rollers that will be magnified at the eyepiece (or camera) regardless of how fine I get the resolution.

 

Kathy, I'm familiar with the calculation for determining arc-second resolution, despite having kids of my own :) My initial post referenced calculations made by the OnStep spreadsheet from Howard Dutton, which aims to account for the lack of precision inherent in microstepping by applying a modifier that varies dependent upon your level of microstepping. I can see how not specifying that initially is confusing, however, and I appreciate everyone's attempts at making sure my math is accurate. A second, third, or fourth pair of eyes on my math is never a bad idea :)


  • kathyastro likes this

#10 klaussius

klaussius

    Viking 1

  • *****
  • Posts: 834
  • Joined: 02 Apr 2019
  • Loc: Buenos Aires

Posted 21 November 2020 - 11:07 AM

Math:

1 step = 1.8 degrees.
32 microsteps = 1.8 degrees.
1 microstep = 1.8 / 32 = 0.05625 degrees.
1 microstep = 0.05625 * 3600 = 202.5 arcseconds.

A 100:1 gearbox gives you 202.5 / 100 = 2.025 arcseconds per microstep.

Kids, don't try this at home.


Ah, yes, brainfart. I used 1° steps not sure why. Anyway the point is that it's far from the necessary accuracy without reduction.

On the other hand the worm gear will act as reduction, so maybew it's ok.
  • kathyastro and galacticinsomnia like this

#11 galacticinsomnia

galacticinsomnia

    Ranger 4

  • *****
  • Posts: 303
  • Joined: 14 Aug 2020
  • Loc: Pacific Northwest - Oregon

Posted 21 November 2020 - 06:45 PM

Math:

 

1 step = 1.8 degrees.

32 microsteps = 1.8 degrees.

1 microstep = 1.8 / 32 = 0.05625 degrees.

1 microstep = 0.05625 * 3600 = 202.5 arcseconds.

 

A 100:1 gearbox gives you 202.5 / 100 = 2.025 arcseconds per microstep.

 

Kids, don't try this at home.

Thank you, I thought I was correct in some of my calculations, but I do sometimes get confused.  (full disclaimer) :)
 



#12 Pauls72

Pauls72

    Mercury-Atlas

  • *****
  • Posts: 2,898
  • Joined: 28 Oct 2007
  • Loc: LaPorte, IN USA

Posted 22 November 2020 - 01:15 AM

I am working on a project myself where I want to have a mount that is completely automated. I have been looking at all alternatives because I do not want to spend 3k on a mount only to leave it outside all 24x7 365 days a year in the weather. So I started looking on ebay. I found not one but 2 older alt az robotic camera mounts for get this $100 plus shipping. They are rated at 60 lbs. All i am saying is look around and find something used perhaps. I have to write my own code to get this thing tracking but I assume you also already know how to do that or else you would not be playing with steppers which are not entirely easy for  a newb.

 

a few photos of my project. these things are built like a tank, no plastic parts whatsoever at all.

Those TSM VS-300 heads use Servo motors, not Stepper motors. Totally different concept in controlling them. Second these are configured as Alt/Az type heads. It will be much harder to control tracking with them, than it would with an EQ style mount.




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