Telescope stepper motors ... or?
CounterRotatingProps
Posts: 1,132
Hi,
kinda new to town ... hope it's ok to post this here.
I'd like·to·position a telescope using a Parallax MC. Yes, there many telescopes·with·motor controls, but mine is not one of them - nor is a retrofit available.
Have read a bit on stepper and servo controls and have tried·some experiments with the Stamp and SX·MC's - wonderful stuff!
My problem is that I'm new to both astronomy and steppers/servos, but have a fair amount of hardware and coding experience. So I'm considering the·fun of rolling my own.·· But I'm near clueless as to what kind of motors will work. "Work" meaning, smooth motion, 12 vdc or less, and a reasonable amount of torque --- I don't know what kind of motors the telescope manufacturers use... still researching that.
This is actually a very interesting robotics problem (well, at least I think so
)
any ideas?
thanks much,
Howard
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Buzz Whirr Click Repeat
kinda new to town ... hope it's ok to post this here.
I'd like·to·position a telescope using a Parallax MC. Yes, there many telescopes·with·motor controls, but mine is not one of them - nor is a retrofit available.
Have read a bit on stepper and servo controls and have tried·some experiments with the Stamp and SX·MC's - wonderful stuff!
My problem is that I'm new to both astronomy and steppers/servos, but have a fair amount of hardware and coding experience. So I'm considering the·fun of rolling my own.·· But I'm near clueless as to what kind of motors will work. "Work" meaning, smooth motion, 12 vdc or less, and a reasonable amount of torque --- I don't know what kind of motors the telescope manufacturers use... still researching that.
This is actually a very interesting robotics problem (well, at least I think so
)any ideas?
thanks much,
Howard
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Buzz Whirr Click Repeat
Comments
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
- Stephen
Ideally it should be small enough to turn the telescope slowly, to track for photography without jitters. The gearing is something I've got to learn about too. Probably fine to very fine resolution would be best - again a trade between 'slewing' quick to a target, vs. slow motion. I'll probably opt for slewing by hand, letting the motors take over for tracking.
thanks
- H
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Buzz Whirr Click Repeat
Will it be mainly planetary, lunar, or comet and asteroid? Or
perhaps sidereal rates for stellar and deep sky. Long exposure
imaging is quite demanding with CCD and a correction system
is advised. Other factors include the type of mounting. Is it a
Dobsonian, Equatorial, or AltAzimuth? You would use different
rates for each of those. Corrections may also be considered in
declination. Do you have a portable telescope or plan an
observatory? Consider telescope size, weight, force required to
turn the mounting to overcome friction, etc. Then you can plan
better for motor type and drive.
humanoido
thanks for the well considered reply,
The answer is basically "yes" to everything except 'equatorial' and 'observatory'
I have a 10 inch Newtonian reflector on a Dobson mount. It's "push-to" so it has no gears or motors, but is does have an object locator. I do understand that Dobsons are far from Ideal for photography (much less long exposure). However, there are homebrew "dobbies" out there with these abilities.
Basically, I'd like to make it easier to track things just for visual observations first. Something like a couple of knobs and switches for each axis - you know: direction, speed, and slew rate. ((Eventually, I can hack the locator and interface it too. But what would be more interesting would be to datalog the time, place, and knob settings when I've actually locked on to a target for uploading to a PC to recall later.))
My problem is not all this interfacing, its what you and Stephen point out. I think I need the "Stepper Motors and Gears for Dummies" book. I can barely tell you the difference between a Propeller Cog and a Spacely Sprocket (aka The Jetsons [noparse]:)[/noparse]
What book would you suggest for learning the ABCs of gear calculations??
thanks again
- Howard
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Buzz Whirr Click Repeat
I would recommend reviewing back issues of Telescope making, Observatory Techniques, Sky & Telescope, and browsing the web for information about such designs already in place. If I were to do this over again, this time I would use a Basic Stamp and create a debug screen operated telescope control console with a laptop. Keyboard control can set slewing rates, tracking, guiding, and various object rates (sidereal, lunar, and planetary).
You could even go GPS with the Parallax module. With date, time, latitude, and longitude, just calibrate on a star and the rest is automatic, even finding objects in the night sky automatically. The database could be stored in the PC, while the Basic Stamp could communicate or data log the info. Plus, algorithms can be applied, the same as the MMTs, the largest telescopes in the world, that use these kinds of mounts.
If you check the list of published astronomy books, there's one dealing with Celestial programs and the mathematics behind the concepts, that could be useful. Check with Willmann-Bell, a publisher that specializes in astronomy books and software. There's astronomy shareware programs to change the monitor to red to maintain dark adaptation - sort of a red debug screen. For stepper motor information, begin with Parallax and review their offerings. There's a tremendous support base from Parallax. For gearing, whether you use gears or round disks, the principles are the same, for step up and step down. Again, some tutorials on gearing found on the web are helpful.
humanoido
I hope you don't mind that I drill down into your reply in some detail...
> The wonderful equatorial platform for Dobsonian mountings.
An equatorial Dobson?· Is that the simple·wedge-shaped platform which has an angle setting made (manually) equal to your latitude?
> a stepper motor on each of the two existing alt azimuth axes.
Precisely!
> get a long threaded rod from the hardware store and drive it with a stepper motor and a slip clutch.
Yes, just what·I'm considering, but what's a slip clutch? (Remember I need the·Gears for Dummies·book [noparse]:)[/noparse]
> For visual, this is workable.
If the rod thread were fine enough, the step fine enough, and the alt-azi motions programmatically coordinated enough to be the equivalent of equatorial motion, could this be use for photography too?
> I would recommend reviewing [noparse][[/noparse]...] such designs already in place
Indeed, just what I'm working through slowly ... the web is super for stuff like this! Have found some links already to (large) Dobsons with homebrew tracking.
>·I would use a Basic Stamp and create a debug screen operated telescope control console with a laptop. Keyboard control can set slewing rates, tracking, guiding, and various object rates (sidereal, lunar, and planetary).
Thanks, this is a very good idea!·· I was planning on using the SX because the protoboards are so inexpensive and I like having interrupts. Had not thought of using the debug screen this way (I've only had basic experience w/ the Basic Stamp (no pun intended
I bet this is a common question, which leads off topic, but isn't there an easy way to interface to the Stamp without going through the·debug screen, say as Stamp Plot seems to do? Sure, there's SIO ... or ...·now·wait a second, is that really all the debug screen is doing, (I mean just SIO)·but with predefined tokens?· Well, back on track...
> You could even go GPS with the Parallax module.
Bingo!·That's one of the "expansion modules" I have for my phase II ... this is one of the reasons I was asking about the AppMod stuff in another thread ... was planning on using a rs-485 bus for add-ons like GPS, compass, datalogging, etc.
> the same as the MMTs, the largest telescopes in the world, that use these kinds of mounts.
MMT = multi mirror telescope?
> books and software
Thanks too for those suggestions ... the math stuff is actually really cool, but weird at first!· I'm working that (spherical) angle too
> There's a tremendous support base from Parallax.
Yes, this is a wonderful place!
> Stepper motor information
Back to square one a second... am I on the right track in using steppers at all, or would highly geared DC motors be better? Hmmm, I guess this is sort of obvious...
>·some tutorials on gearing found on the web are helpful.
Now there, I've gotta start hitting the books!
What kind of telescope gear do you have?
thanks again,
Howard in Florida
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Buzz Whirr Click Repeat
For more information about the Equatorial Driven Platform, for Dobsonian mounts, consult the web page created by the master himself, www.equatorialplatforms.com/. Tom made this design popular and the web site is a gold mine of information.
Yes, you can use these systems for astro imaging, astrophotography, and CCDs. The threads per inch will be important, but also important are some slew rates for guiding. For SEA, short exposure astro imaging, you need a good drive rate with good accuracy. A well built and adjusted drive can do it. There are many examples of working systems. I think the Basic Stamp will work well for this.
The actual Dobsonian mount is amazing simplicity. On Big Bear Mountain, John Dobson showed us his homebuilt 24-inch reflector. Up on a 10-foot ladder, a good hand push could move the scope. It used a friction slip bearing surface. Teflon is a popular material. The Dobsonian mounting also worked well for the 40" reflector. About the drive for your scope, the threaded rod is very simple - a great advantage. About the only disadvantage is you will have to reset it, once it reaches the end of travel.
As you have surmised, you will be interested in the threads per inch to calculate the drive rates. The length of the threaded rod will determine the amount of tracking time. Some people have bent these rods into near perfect semicircles to obtain a kind of pseudo gear. At the time, I wrote a computer program to calculate parameters, drive rates, TPI, motors, lengths, radii, and tried out different systems inside the computer first. You can do this with a Basic Stamp and the Debug screen.
About the slip clutch - if you use only a direct drive to stepper motors, the telescope will always remain locked in some position. There will always be some night where you just want to move the scope by hand. So a slip clutch will allow telescope movement by hand when the motor is still engaged. It's all about adjusting the amount of friction in the clutch.
I suppose you could use PM motors and gear down. I did it before with the VIP machine. A PM motor will have coasting to deal with, compared with stepper motors. With stepper motors, you will be concerned with the step rotation size.
Interfacing the Basic Stamp has many approaches. It's another thread or two. Read up on Stamp Plot Pro, data logging, some of the near space stamp experiments, and further hone in your astronomical robotic stamp requirements. I suggest you begin with the most simple system first, build a simple prototype, then enlarge it later.
MMT = multiple mirror telescope. I got a lot of ideas when building a 144-servo driven motor mount for the 40-inch primary objective mirror. It could shape and figure the mirror in real time. But the concept for the telescope's alt-az mount is the same as your scope. Remember you can program and design different drive rates. One approach is to construct an analog VFO, variable frequency oscillator. Another is to use code and go all digital.
I'm working with tiny ultra portable telescopes now, using the same 10x approach with CCDs and a 60mm Meade ETX-60 for results like a 24-inch scope. My current research is about solar corona and photosphere resolution with solar filters created from atmospheric pollution and depth. I have attached one result, taken with camera only (no telescope).
humanoido
just wanted to drop a note of thanks for the excellent reply ... mulling over what you said and will respond after moving forward on what you suggested.
cheers
- Howard
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Buzz Whirr Click Repeat
Humanoido, I had to drop this in favor of more pressing matters, but now I'm looking at it again... and have narrowed it down in the mean time.
When you have a moment, I'd appreciate your thoughts.
The threaded rod method seems to be the best option. Yes, the link to equatorial platforms was very helpful. I've decided to DIY down to the controller level for the learning experience.
The scope I want to motorize is a 10 inch Dobbie on the usual teflon mounts. It's an Orion Intelliscope, so I can interface into their position sensors once I crack the way it works. (Looks like magnetic pulses.)
Considering that this 'push-to' scope weighs about 70 pounds, but can be moved easily with just finger nudges when the tensioners are set just right, do you think this stepper motor Parallax offers will be strong enough to rotate the rod? (They are pretty small) :
4-phase / 12 volt Unipolar, 7.5 degree step:
http://www.parallax.com/Store/Accessories/MotorServos/tabid/163/CategoryID/33/List/0/SortField/0/Level/a/ProductID/65/Default.aspx
And was thinking about either of these controllers, but am uncertain as I've not used them yet:
Little Step-U
http://www.parallax.com/Store/Accessories/MotorServoControllers/tabid/160/CategoryID/33/List/0/SortField/0/Level/a/ProductID/68/Default.aspx
probably better would be:
BiStep USB controller
http://www.parallax.com/Store/Accessories/MotorServoControllers/tabid/160/CategoryID/33/List/0/SortField/0/Level/a/ProductID/375/Default.aspx
thanks again,
Howard
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
No matter how much you push the envelope, it'll still be stationery.
humanoido
humanoido
do you think 7.5 deg/step will prove to be too course for photo work ? The rod will probably be only 1/8". (Problem is, there's really no way to tell just how bouncy or jaggy it will be until I try it.)
Have not decided on camera --- perhaps I should wait for your invention to be ready? It sounds very cool - how will it differ from an off-the-shelf CCD w/tracking software?
> a new Parallax sensor ... by simply scanning along one axis while the other is held stationary.
Interesting, if I understand it, this would be like an old raster-scanned T.V. image... one line scan, bump to next line, scan, repeat til full image... right? Getting the lines to, er, "line up" with each other seems like it would be either very easy or very difficult. Easy if the data array was cleanly aligned, difficult if there was a slight shift.
Thanks too for recommending Parallax --- Yes, they are GREAT !
cheers,
- Howard
~~~~~~
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
No matter how much you push the envelope, it'll still be stationery.
>>do you think 7.5 deg/step will prove to be too course for photo work ? The rod will probably be only 1/8". >>(Problem is, there's really no way to tell just how bouncy or jaggy it will be until I try it.)
Yes, I would absolutely use the Parallax unipolar 7.5 deg/step stepper motor. An alternative way to step this motor is called half step control, where we power also two adjacent coils at the same time, to move the motor half-way. This implies a better control on position and speed, at the cost of doubling the pulse ratio at any given speed. It can lead to 3.75 degrees per half-step, a much finer tracking motion. Here are some benefits of this system.
* multiple speeds, to track the sun, stars, moon, deep sky objects, galaxies, nebula
* Fine tuning of all speeds, with permanent memory of the last stored value
* Quick motion to pass to another object, at different speeds (Slewing)
* Pause mode to momentarily stop the motion
* Reverse rotation, both at tracking and quick speeds (no additional parts needed)
* Ease of use
Cold rolled threaded rods are usually quite accurate for this application. I believe you could work up the math for the tangent arm and the length of the rod to determine motion traveled/tracking velocity vs time in operation before reset. Again, consult formula by Osypowski and back issues of Telescope Making, Observatory Techniques, Sky & Telescope, Modern Astronomy, the Celestial Observer, the Astrograph, The Journal of the Association of Lunar and Planetary Observers, and various Astronomy Meetings such as the Telescope Makers Association, San Francisco Sidewalk Astronomers, the Riverside Meet, etc. Here, the web is filled with the information you are looking for, as this type of project was popularized back in the 1970s and repeated many times with newer and newer technology, thus you have numerous sources to draw upon. However, the basic concept remains the same.
>>Have not decided on camera --- perhaps I should wait for your invention to be ready? It sounds very cool >>- how will it differ from an off-the-shelf CCD w/tracking software?
I enjoy using low cost DIY techniques, recycle kitchen parts, Parallax Basic Stamps/sensors.
>>>> a new Parallax sensor ... by simply scanning along one axis while the other is held stationary.
>>Interesting, if I understand it, this would be like an old raster-scanned T.V. image... one line scan,bump >>to next line, scan, repeat til full image... right? Getting the lines to, er, "line up" with each other seems >>like it would be either very easy or very difficult. Easy if the data array was cleanly aligned, difficult if >>there was a slight shift.
With a telescope that can track a dime - i.e. a Marathon runner a mile away holding up a dime traversing a football field, it should be no problem.
I built a similar imager using one of those hand held scanners. Instead of moving the scanner by hand, it was attached to the telescope which provided the motion. It took really fantastic closeup images of the Moon in 256 gray scale. You can still find these scanners on ebay.
With the Parallax sensor, it would be a line by line scan, and a natural drift using the Earth's rotation to impart the stepper motion. So you won't need a servo - just let the Earth do all the work. [noparse]:)[/noparse]
>>Thanks too for recommending Parallax --- Yes, they are GREAT !
They are!
humanoido
I've ordered two of the steppers and the BiStep controller. That conttoller·seems·more sophisticated than the Little StepU. Sure, I could have an SX or Prop control them, but I'd rather concentrate on the high-level. Although my spherical trig. and calculus aren't too shabby, and I have Smart's classic work,· I was thinking of doing a brute force approach to calculating tracking, something similar·to·"letting the earth do the work."
= Crack and install an OTC autotracker / autoguider, not·for tracking, but to log *mis*tracking.· (Or, one of the new line scanners and an SX might make for a cheap guider, eh?)
= Pick a star, any star.
= Set·arbitary speeds on the motors -
= Visual check·tracking until it's reasonable, over a few minutes say --- GOTO step 2a.
1. (re) adjust speeds
2a. Data-log the motor-motion parameters.
2b. Data-log the autoguider's attempts to correct.
2c. let run an hour or more...stop, dump data.
3. Compare the data sets - a simple plot would probably do - and I'll "see" how much/little the motors need to be changed.
Repeat at 1 until 'fine tuned'.
Then save 2a. data for future auto control. (as you mention in your second point, last message).
All with NO math (unless I have to do a Fourier analysis on the dataset that is [noparse]:)[/noparse]
What do you think, is this nuts?
cheers,
Howard
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
No matter how much you push the envelope, it'll still be stationery.
there are lots of people there· that talk about driving steppers and what gear ratios to use and formulas.
my first cnc was using pminmo driver cards ( opensource , sells cards for $8 , or you can get the copper design free)
the advantage of using driver boards for your stepper is much better than the ones that you·mentioned due to the fact that you can increase the voltage to the stepper, it's common to do that by a factor of 20, as long as you keep the current at the steppers limit.( this increases the torque of the motor, due to the current building quicker in the coil , its called a chopper circut)
i've looked at pminmo's boards, and it wouldnt be hard to drive them with a prop. as a matter of fact, if you rearranged the copper on his 4 axis bob, you could probally drop one right in. or you could use it for it's original purpose, to be driven by a computer, and have some sort of image comparison/seeker program going
i use 495 oz/in motors to move a gantry that weighs about 100 pounds at 200 in/min using direct drive screws 5 start 2tpi. my resolution is about .005"
the principle is the same, moving a large load to a prcise point .
you can buy 10tpi screws from enco (1/2" ) for about $6 for 6 feet, nylon nuts for them on ebay for $5 ea.
anyways that kind of info is available at cnczone forums, you just have to apply it·to your case.
dan
·
thanks for the valuable tips. Hanging out a while at the CNC site sounds like it'll teach me some good ideas.
Yes, you're gantry is a good example of how similar these things really are - pretty amazing (to me anyway) that such a mass can be moved at that speed and get 0.005" accuracy.
Do you think I'd really need 1/2" all thread for this light of a load? Was thinking of only 1/8" (... I have a bunch of 3/8" left over from hurricane retrofitting I could use, but thought that too large.)
Oh, DOH! - the 1/8 might flex too much.
Can you clarify one point in your post for me please:
> using direct drive screws 5 start 2tpi.
2tpi = 2 threads per inch, but what's a "5 start" ?
thanks again... see you maybe over in the CNCZone...
cheers
- Howard
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
No matter how much you push the envelope, it'll still be stationery.
An autoguider has a CCD sensor that can see a star and lock onto it.
I presume you are talking about making one, or referring to Stamp.
With a Stamp, just dial in or program from the pc (DEBUG) input a value,
and adjust the stepper by visual tracking inspection at the telescope on
a star over a couple minutes.
My last tracker and autoguider used the SBIG ST4 and a home-built
variable frequency oscillator. The autoguider requires two motors
and the means (connections) to reverse the motions. This could be
accomplished through smart PBASIC software.
(the principles are the same here) Before the ST4, I used
my own hand held control panel. It had a knob for coarse adjust and
one for fine tune. I would center the star, set the coarse knob for
sidereal rate, and watch the star's position in the illuminated reticule,
thereafter making fine adjustments to the rates. With an equatorial
mount, there was a declination screw mount to make fine adjustments
over time, but was minimally necessary as the mounting was so
accurately polar aligned.
(going altazimuth) With the Dobsonian mount, you will get
an altazimuth representation of the night sky and the rate will change.
There is also some field rotation to consider. These points are mulled
over many times in various source documents. A lot of this depends on
your specific intended applications.
humanoido
You've aldready done much of what I've yet to try, but want to. So it's
inspiring to see that this stuff works.
For autoguiding, the OTC stuff is a bit over my budget at this point, so
I will likely DIY - maybe even with the new line sensor Parallax has.
(Did you notice that it has the typical Red filter?
not so good for astro photography, but it might be ok for tracking as
you want to sync to a hot star anyway. Maybe the filter is removable?)
Isn't the SBIG really too big for just tracking alone?
RE Handheld control. Yes this is exactly what I plan to do - ALT AZI
before equatorial... If I can track 3 changes (ALT/AZI/Field Rotation)
then equatorial will be very easy later. I call "Field Rotation" "Slew" as
I've not come across the correct technical term yet - is it simply "feild
rotation" ?
My first handheld design:
4 buttons = on/off, direction (fwd/rev) for ALT/AZI/SLEW;
4 red/green indicator LEDs for the above
3 slider-pots = speed for ALT/AZI/SLEW
My latest version is much simpler: using a hacked up Nintendo NES
controller - it's cheap, easy to interface, and has more controls than
I need.
cheers,
Howard
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
No matter how much you push the envelope, it'll still be stationery.
Hey, that red filter is fantastic! A red filter is the most useful filter in my telescope parts box. It has the ability to darken the sky under city lights, penetrate atmospheric seeing, look through haze and clouds on Mars right down to the surface, act as character selection filter, filter out specific wavelengths for astrometric analysis, maintain dark adaptation, reduce light intensity for Lunar purposes, and numerous other functions. I suggest you keep the red filter!!!
The SBIG ST4 is much smaller than the ST7. It's ideal for tracking, guiding, and even imaging! With the ST4 and a homebuilt 12.5-inch diameter Newtonian, I discovered a star not on the 10-million object data base that the Hubble Space Telescope uses for Star Positioning! You can do great discoveries with this equipment!
Slew is not field rotation. Imagine a circle that you look through at the telescope ocular. Now imagine it turning. This turning is called field rotation. You'll need another way to drive this angular axis for imaging, but only if your imaging is longer than a couple minutes or if you want the finest of high resolution.
humanoido
http://forums.parallax.com/showthread.php?p=814265
humanoido