Jump to content


Photo

Precision Rotary Stage for a RA axis?

  • Please log in to reply
21 replies to this topic

#1 Brian L

Brian L

    Viking 1

  • *****
  • Posts: 560
  • Joined: 17 Nov 2008
  • Loc: The garden paradise of Pittsburgh, PA

Posted 24 February 2010 - 11:03 AM

I've come into possession of a Newport RV240PP rotary motion stage and controller which I think may be adaptable to use as a RA-axis for a serious fork mount. It is a Newport RV240PP, and it is a monster.

http://search.newpor...=sku&q2=RV240PP

Here are some specs:

Travel Range: 360 degrees
Worm Gear ratio: 1:90
Resolution: 0.00005 degrees
Centered load capacity: 4000N (~400 lb)- would be less of course when tilted to proper latitude
Maximum speed: 20 degrees/sec

Detailed specs are attached.

Any thoughts? The one spec that concerns me is the absolute accuracy: 0.010 degrees. This depends on the slope and travel, so I am not sure how this applies to sidereal tracking rates.

#2 JAT Observatory

JAT Observatory

    NOT a Wimp

  • *****
  • Posts: 9431
  • Joined: 20 Feb 2005
  • Loc: In the Primordial Soup

Posted 24 February 2010 - 12:52 PM

Wow very nice.

#3 Brian L

Brian L

    Viking 1

  • *****
  • Posts: 560
  • Joined: 17 Nov 2008
  • Loc: The garden paradise of Pittsburgh, PA

Posted 24 February 2010 - 04:07 PM

The only drawback is that it is a stepper motor design and not a servo. The smallest increment is 0.001 degrees, which is about 3 arcsec. I'm not sure that is ideal for this application. There is another model with a reduction unit that can move in 0.3 arcsec steps.

#4 neo

neo

    Surveyor 1

  • *****
  • Posts: 1644
  • Joined: 12 Feb 2008
  • Loc: Iasi, Romania

Posted 24 February 2010 - 04:27 PM

My goodness...almost $8000?? :bigshock: :bigshock:
What scope do you plan to use with "the monster" :p ?

#5 Brian L

Brian L

    Viking 1

  • *****
  • Posts: 560
  • Joined: 17 Nov 2008
  • Loc: The garden paradise of Pittsburgh, PA

Posted 24 February 2010 - 07:06 PM

The good news is that it cost me next to nothing- I would never drop that kind of dime on such a project.

As far is the scope is concerned, I haven't thought about it really. I think it could comfortably handle 250 lbs. A C-14 binoscope maybe? I don't even know if this specific rotary stage has enough stepper resolution for a proper telescope mount. Ideally, I'd be able to develop an interface for the GCC to control it.

I see this potentially being a multi-year project that ends up costing thousands of dollars and produces nothing particularly useful. There is the added problem of not knowing where to put such a thing if it ever gets completed.

#6 neo

neo

    Surveyor 1

  • *****
  • Posts: 1644
  • Joined: 12 Feb 2008
  • Loc: Iasi, Romania

Posted 24 February 2010 - 08:08 PM

If it turns out to be suitable for a big mount, I guess it should be interesting for a permanent set up.
About the resolution...maybe it could be managed with a proper driver with microstepping.
Why do you say that the stepper motor design would be a drawback?

#7 Brian L

Brian L

    Viking 1

  • *****
  • Posts: 560
  • Joined: 17 Nov 2008
  • Loc: The garden paradise of Pittsburgh, PA

Posted 24 February 2010 - 08:20 PM

I think the issue is that the resolution of the rotary encoder is 0.001 degrees, or about 3 arcsec. I do not know yet if this limitation of the stock system can be overcome.

#8 neo

neo

    Surveyor 1

  • *****
  • Posts: 1644
  • Joined: 12 Feb 2008
  • Loc: Iasi, Romania

Posted 24 February 2010 - 10:28 PM

Hm..any chance to get rid of that encoder? just drive the stepper with a good controller with microstepping.
But still..3 arcsec sounds good. would that mean you get a 3 arcsec precision on guiding? If so, it's winner. :jump:

#9 galacticphoto

galacticphoto

    Vostok 1

  • -----
  • Posts: 190
  • Joined: 12 Nov 2009
  • Loc: Huntsville, AL and Atlanta, Ga

Posted 25 February 2010 - 08:05 AM

I think the issue is that the resolution of the rotary encoder is 0.001 degrees, or about 3 arcsec. I do not know yet if this limitation of the stock system can be overcome.


You will probabaly be imaging in the neighborhood of 1"/pix (or slightly less) w/ the C-14 you mentioned earlier. A minimum step of 3" won't do well. If they make 0.3" unit, I'd go for that, or plan on a different stepping regime (and probable reduced torque).

Robert

#10 Brian L

Brian L

    Viking 1

  • *****
  • Posts: 560
  • Joined: 17 Nov 2008
  • Loc: The garden paradise of Pittsburgh, PA

Posted 25 February 2010 - 10:08 AM

I got this unit for next to nothing- it retails for $8K. The model with the reduction unit giving 0.3 arcsec steps is about $13K. I don't know if the reducer can be retrofitted, but if it can I am sure it is more than I am willing to spend.

It might be possible to strip out the stock motor and encoders, and just use the worm and stage with a different motor and encoder. I certainly don't need the kind of torque this think can produce, nor do I want a C-14 slewing around at 20 degrees a second.

#11 TeaDwarf

TeaDwarf

    Lift Off

  • -----
  • Posts: 11
  • Joined: 20 Feb 2010
  • Loc: Oxon, UK

Posted 25 February 2010 - 10:53 AM

The Newport stages are nice stages -- but as I think you've identified, not really ideal for a telescope drive. What it would make is an absolutely beautiful derotator for an alt-az telescope...

Did you manage to get the controller as well? may be a big job to re-engineer that part... You'll need access to the encoders + reference switches, in addition to driving the stepper motor, to run it properly I suspect.

#12 Mert

Mert

    Skylab

  • *****
  • Posts: 4219
  • Joined: 31 Aug 2005
  • Loc: Spain, Pamplona

Posted 25 February 2010 - 01:39 PM

I think the issue is that the resolution of the rotary encoder is 0.001 degrees, or about 3 arcsec. I do not know yet if this limitation of the stock system can be overcome.

Have you got any specific information on the stepper motor
mounted in this beauty????
Maybe with microstepping you can go down to 1/16th of a
step and have way better resolution???

Just my :penny::penny:

#13 TxStars

TxStars

    Apollo

  • *****
  • Posts: 1328
  • Joined: 01 Oct 2005
  • Loc: Lost In Space

Posted 25 February 2010 - 01:46 PM

My guess this spec "Mini-step drive"
I would think they are talking about micro steps.
So you are already at the minimum step size using that stepper.
As large as this thing is you should be able to change to a servo drive. :question:

#14 Brian L

Brian L

    Viking 1

  • *****
  • Posts: 560
  • Joined: 17 Nov 2008
  • Loc: The garden paradise of Pittsburgh, PA

Posted 25 February 2010 - 09:20 PM

I do have the controller- it is a Parker compumotor Zeta 4. The motor is a Newport 60 VDC UE63PP stepper motor. I think the minimum rotational increment of 0.001 is defined using the minimum step size.

I think it would be possible to use a different motor. The worm and bearing blocks for this thing are seriously heavy duty and ultra precision. I suspect that periodic error would be incredibly low with this thing. 60 VDC is a little inconvenient. However, if the unit could be re-engineered with a 12VDC servo motor it could prove to a heavy duty mount without equal in the amateur astronomy world.

I've got a PDF of the manual if anybody is interested. I fear that reverse engineering this for a telescope mount might be out of my league.

#15 galacticphoto

galacticphoto

    Vostok 1

  • -----
  • Posts: 190
  • Joined: 12 Nov 2009
  • Loc: Huntsville, AL and Atlanta, Ga

Posted 26 February 2010 - 08:39 AM

I certainly don't need the kind of torque this think can produce, nor do I want a C-14 slewing around at 20 degrees a second.


That slew rate would make a great mount for tracking fast moving satellites, airplanes, birds....

Robert

#16 Starhawk

Starhawk

    Space Ranger

  • *****
  • Posts: 5507
  • Joined: 16 Sep 2008
  • Loc: Tucson, Arizona

Posted 26 February 2010 - 10:14 AM

I have used these. You are not going to be able to just slap whatever on there and have it take the weight- you will need to balance it.

The controller is based on classic Proportional Integral Derivative (PID) control theory, largely because the main uses for these are as precision location stages. So, for example, if you have something in your factory which needs to be turned to precise clocking angles, the controller will drive to those locations.

What I haven't seen native for these is what astronomy needs- precision location during movement at a constant rate. Here what you would need to do is be able to use the controller to get to a precision rate and stay there. The mount would then need to be able to move to an arbitrary clocking and stay oriented there.

With that said, I am sure the device is capable of that, and you might be able to call up Newport and ask them if someone has done exactly that with the mount, and if you can have the code. My own enquiries along those lines have characteristically come up with a negative response, but it can only help to try. Otherwise, I hope you like programming.

Alternatively, another way to use this unit might be to set it up as a virtual Dobsonian table. For this case, what you would do is give the mount a slew rate table to process, and input the slew rate to match your desired tracking rate. You can then put a separate Alt-az setup on top of the ring, so it will now be equatorially mounted. If you put something like Orion Intelliscope encoders on this setup, you should be able to have a sort of Goto (well, not actually Goto- drive-to).

With that said, the altaz setup will need to be balanced on the mount. The Newport controller can do some very wrong things if the balance changes significantly from what it was when you set it up (I know, PID shouldn't do that...keep it balanced).

-Rich

#17 Brian L

Brian L

    Viking 1

  • *****
  • Posts: 560
  • Joined: 17 Nov 2008
  • Loc: The garden paradise of Pittsburgh, PA

Posted 27 February 2010 - 11:26 AM

Starhawk-

THanks for your thoughtful response. If you just look at the unit from a mechanical perspective it is really no different than a mount axis. Agreed, the intended applications for these rotary stages are not ideal for astronomical tracking. However, I think those limitations are more or less contained in the existing controller, motor, and encoder. If you strip the stage down to its mechanical components, it seems like it must be adaptable to the task. I'm not sure how much effort it will be to re-engineer it from the bare mechanics.

You are right about the loading and balancing. This stage does not have the load capacity mounted vertically as it does mounted horizontally.

#18 Starhawk

Starhawk

    Space Ranger

  • *****
  • Posts: 5507
  • Joined: 16 Sep 2008
  • Loc: Tucson, Arizona

Posted 27 February 2010 - 12:26 PM

Starhawk-

THanks for your thoughtful response. If you just look at the unit from a mechanical perspective it is really no different than a mount axis. Agreed, the intended applications for these rotary stages are not ideal for astronomical tracking. However, I think those limitations are more or less contained in the existing controller, motor, and encoder. If you strip the stage down to its mechanical components, it seems like it must be adaptable to the task. I'm not sure how much effort it will be to re-engineer it from the bare mechanics.

You are right about the loading and balancing. This stage does not have the load capacity mounted vertically as it does mounted horizontally.


It sounds like you are onto the problem- and the solution. The stage controller really is intended to go to discrete stationary locations, and then point at them with known accuracy. The slew rate is usually chosen for reasons such as safety, and is not actually reported back in real time to get a rate and coordinate location.

With that said, if you look a the stage mechanically, it is really good stuff. The bearings are very high quality, and it is meant to undergo insane numbers of cycles compared to what astronomical use will ever ask from it.

If you can ratio the device so an existing gear system will work with it, then that could be an easy answer. So, for sake of discussion, if you can procure a 288 tooth worm, then hardware for Losmandy's Gemini will work with it (with your worm and gear). The large diameter may make this problematic, though. I would be interested in what you find, since actually, a success in this area would mean you had a technique for making ANY high quality bearing into a mount, which would open up a vast array of possibilities.

The open center and large diameter of the of this type of bearing actually begs the possibility of doing things like installing a fork on top of it (and I mean don't screw around- make it stout- an English yoke arrangement could be supported, which would give you observatory class load carrying capability). Newport builds fork type multi-axis stages for about $100k, and these too need to be balanced and come with all of the peculiar constraints we have talked about already.

Alternatively, if you can get a second one of these (Newport typically stacks these things that way to build up their own multi-axis stages, such as the fork I mentioned above), you could use it as a declination axis and do some neat party tricks like put a Coude focus through the bore of the mount. And as mentioned, Newport will give vastly different numbers for static loading and dynamic accuracy, so balance is key with their controllers.

If you don't mind me asking, where did this come from? Generally, the hardware I have used with them comes as part of an optical table with too many zeroes attached for me to even contemplate buying one (ever).

-Rich

#19 Brian L

Brian L

    Viking 1

  • *****
  • Posts: 560
  • Joined: 17 Nov 2008
  • Loc: The garden paradise of Pittsburgh, PA

Posted 27 February 2010 - 04:04 PM

Starhawk-

A lot to think about. Gemini Control Center (GCC)- the software emulator of the Gemini system has a custom mount feature that allows the user to specify worm and spur gear ratios as well as encoder resolution.

The stage actually came from a micro CT scanner that was in my laboratory. It was obsolete and dismantled 5 or six years ago. The rotary stage held the CT tube and phosphor screen detector. I kept any potentially useful pieces, gave the X-ray source to the Physics dept., and scrapped the rest. Nobody at the university wanted the linear or the rotary stages, controllers, or power supplies. They've sat in a corner for 5+ years.

Now we are cleaning out for major renovations, and a lot of junk has to go. Surplus didn't want them, neither did any surplus equipment dealers. I couldn't bear the thought of seeing them go into the dumpster, so I rescued them with the idea of using them for a mount project.

#20 Starhawk

Starhawk

    Space Ranger

  • *****
  • Posts: 5507
  • Joined: 16 Sep 2008
  • Loc: Tucson, Arizona

Posted 27 February 2010 - 08:11 PM

That's actually a neat idea. I still see Byers 359 tooth worm sets show up from time to time in sizes of 9" or bigger, so that could yield a very nice drive setup using GCC.

I can see how you wouldn't want to set such a piece of equipment aside. But with that said, I'm actually wondering if it might be better as a Dec axis.

-Rich

#21 Brian L

Brian L

    Viking 1

  • *****
  • Posts: 560
  • Joined: 17 Nov 2008
  • Loc: The garden paradise of Pittsburgh, PA

Posted 28 February 2010 - 09:07 AM

I'm not sure exactly how I would use a Byers worm set with the Newport stage. I've been afraid to take it apart, but it looks like the teeth for the existing worm are integral...

#22 Starhawk

Starhawk

    Space Ranger

  • *****
  • Posts: 5507
  • Joined: 16 Sep 2008
  • Loc: Tucson, Arizona

Posted 01 March 2010 - 09:35 AM

I wouldn't take it apart, either. The model I was thinking of was to attach to the mounting face on the side- essentially it would be externally driven with the internal drive unused and its gearhead removed. And that was what started me thinking it might make a lot more sense to use this as a Dec drive, where constant rate isn't a requirement. In that case, you can use another bearing for the RA, whether you go GEM or fork.

Taking one of these apart is a big deal- just changing the location of the limit switches on one has meant calling in riggers to pick an entire 3 axis stage up and put it on a pallet and ship it back to Newport (surprise! Even minor mods aren't minor).

So leaving well enough alone and using it as a Dec Axis is making a lot of sense to me at the moment.

-Rich






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics