Jump to content


Photo

Open Source Equatorial Mount

  • Please log in to reply
53 replies to this topic

#26 Adam Brunette

Adam Brunette

    Sputnik

  • -----
  • Posts: 28
  • Joined: 13 Jan 2013
  • Loc: Charlton, Massachusetts

Posted 21 January 2013 - 11:41 AM

you can get the gears from Aeroquest:

http://www.aeroquest...com/pricing.php

smaller one for the DEC. or, if goto is not necessary, use a tangent-arm DEC. it would be simpler, cheaper, and suffer less from backlash and stiction.

i would vote for a Raspberry Pi (or Chipkit32 Arduino) for the Goto controller, if Goto is desired.

the 10 Micron mounts use an embedded Linux SBC, not the cheesy low-RAM microcontrollers that are generally used in mounts.


Hello orlyandico,

Those are some nice gears!! Looks like top quality machining. I will keep them in mind, but they are super expensive.

This will be a GOTO mount.

The electronics will be AVR based which will make it Arduino compatible. My actual passion and schooling is electronic design. Machining and engineering is just a hobby.

This little guy I built a few months ago for work is a hand held network tester and cable checker complete with 2.4" color touchscreen: https://lh4.googleus...AAF3s/nhOLF8...
https://lh4.googleus...AAF28/nwWoxE...

I should know a little about the OpenSource licensing since I
follow alot of OS projects, but I don't. Adam can join the G+
community "Makers, hackers, artists, engineers" community run
by adafruit.com and get the answers there.

I like this open source astro equipment trend. I have(just got)
a 3d printer. If the cad files are in a Blender friendly format
I can get them to my printer and print them to see how they fit together.
If the files can be exported to a '.stl' file, even better. If proto-typing in plastic will help I can be of
service.


BoriSpider,

YES!! all the files can be exported in numerous 3D formats. When the time comes, this might be an excellent way of building prototype pieces instead of aluminum!

A lot of nice looking work can be done using pipe fittings bored to suit bearing inserts etc. Clearly you want to get past the "sawmill" look on your mount.


ccaissie,

That is a great idea! I will keep it in mind for a smaller project. This one Im sticking with pieces designed for the mount specifically and easy machining on a lathe.

-Adam

#27 Adam Brunette

Adam Brunette

    Sputnik

  • -----
  • Posts: 28
  • Joined: 13 Jan 2013
  • Loc: Charlton, Massachusetts

Posted 21 January 2013 - 12:22 PM

Alright, Sorry for the delay on the cutaway and dimensions.

First, the cutaway:
https://lh6.googleus...AAHa0/JjdttC...

As you can see, The gear end of the 2 axes are supported by 6218-2rs deep groove bearings. At the far end of the housing, there is another bearing with a bearing pre-load nut. This is tightened down to take all the backlash out of the bearings and provide a fluid axis. The RA axis will be set up the same way.

You can see at the top of the DEC that I have been playing with a clutch design. This is based on the clutch in the Losmandy G11. The tan layer is a piece of cork used as a slip/friction surface for the clutch pad. This would have a secondary shaft through the center of the main shaft and a hand wheel on the end to adjust clutch tension. The thing that worries me is if the end plate and shaft are not rigid enough, they could flex when the clutch is loose and cause inaccuracy. I think bushings would be sufficient for holding the clutch shaft concentric with the axis shaft.

If this is acceptable, I will replicate it on the RA axis.

Here is a rough dimensional drawing. The ones marked "True" are radiuses.

All measurements are in Inch's.

https://lh5.googleus...AAHa4/EaOkyF...


Im still looking for suggestions on the range of latitude the mount should be capable of.

-Adam

#28 ahopp

ahopp

    Messenger

  • -----
  • Posts: 402
  • Joined: 24 May 2012

Posted 21 January 2013 - 02:04 PM

Mine has an 8.5' by 10' observing room, retractible fabric roof. The front has an 8.5'x10' plus v-nose, for computer room and living quarters. I use a 3K Honda generator with a 200 AH battery for power. Custom 12vDC control panel to power all the gear. A/C plus baseboard and gas heat.

Tony

#29 BoriSpider

BoriSpider

    Vanguard

  • -----
  • Posts: 2082
  • Joined: 10 Aug 2004
  • Loc: S.W.FLA

Posted 21 January 2013 - 05:48 PM

Looking good so far Adam. I love the 'lil tester/checker thing.

#30 neo

neo

    Surveyor 1

  • *****
  • Posts: 1645
  • Joined: 12 Feb 2008
  • Loc: Iasi, Romania

Posted 22 January 2013 - 06:20 AM

Nice design Adam!
How do you plan to mount the worm gears on the shafts( if that is what you are going to use)?
It would be a good idea though to use roller bearings instead of ball bearings for a better precision and as a sophistication you could use taper roller bearings on the back end of the shaft, especially if you want to go with the preloading idea. Also it would be a more efficient way to deal with the axial and radial forces when the mount points in different positions. I believe Takahashi uses this set up in their mounts.
If you want to go Losmandy way then you should consider mounting the worm wheel on a bearing on the Ra (or DEC) housing itself making it independent to the shaft and avoid the small errors due to flexing or bearing runout. The most important thing I see with this set up is that the worm and worm wheel keep the same contact which will help accuracy. In this way the shaft is driven only by the friction between the worm wheel and the flange of the shaft.
http://www.astro.uni...xperiences.html

http://www.astromast...intenance_e.htm

Here's also a new ukrainian series of mounts using pretty much the same design with very good accuracy, not very hard to build on a lathe.
http://www.youtube.c...low=grid&view=0

#31 m. allan noah

m. allan noah

    Viking 1

  • -----
  • Posts: 752
  • Joined: 14 Aug 2009
  • Loc: Virginia, USA

Posted 22 January 2013 - 08:24 AM

You need to add a flange or washer that covers the seals on the exposed side of the bearings. Eventually during transport, one of those will get hit and damaged.

allan

#32 Adam Brunette

Adam Brunette

    Sputnik

  • -----
  • Posts: 28
  • Joined: 13 Jan 2013
  • Loc: Charlton, Massachusetts

Posted 22 January 2013 - 10:34 AM

Nice design Adam!
How do you plan to mount the worm gears on the shafts( if that is what you are going to use)?
It would be a good idea though to use roller bearings instead of ball bearings for a better precision and as a sophistication you could use taper roller bearings on the back end of the shaft, especially if you want to go with the preloading idea. Also it would be a more efficient way to deal with the axial and radial forces when the mount points in different positions. I believe Takahashi uses this set up in their mounts.
If you want to go Losmandy way then you should consider mounting the worm wheel on a bearing on the Ra (or DEC) housing itself making it independent to the shaft and avoid the small errors due to flexing or bearing runout. The most important thing I see with this set up is that the worm and worm wheel keep the same contact which will help accuracy. In this way the shaft is driven only by the friction between the worm wheel and the flange of the shaft.
http://www.astro.uni...xperiences.html

http://www.astromast...intenance_e.htm

Here's also a new ukrainian series of mounts using pretty much the same design with very good accuracy, not very hard to build on a lathe.
http://www.youtube.c...low=grid&view=0


The large cylinder that mounts on the large bearing is to be the worm gear.

Taper bearings are better for preloading, but with every advantage comes drawbacks. Taper bearings are never sealed units (that I have seen). This means measures have to be taken to hold the grease in the bearings.

One major issue to overcome in the homemade/diy category is in order to use needle and direct roller bearings is the shafts need to be hardened and ground as they serve as the inner race. Without this, you will have very poor results. From my past research, it is perfectly acceptable to preload deep groove bearings. A common preload for these is about 4 percent. 5% is 300lbs of preload.

Thanks for the links. Lots of good breakdown pictures to look at.

The Ukrainian mounts are really nice! They look very well built. I cant understand anything they are saying, but I can tell they have has the shafts of the axes hardened and ground. In the 240 mount I made out what I think is cork for the friction pad between the RA gear and end mount.

You need to add a flange or washer that covers the seals on the exposed side of the bearings. Eventually during transport, one of those will get hit and damaged.

allan


Allan,

The small end of each axis will be covered by the knob that tightens the clutch. The large bearing will need some kind of cover, but this will also serve as a gear cover.

-Adam

#33 Adam Brunette

Adam Brunette

    Sputnik

  • -----
  • Posts: 28
  • Joined: 13 Jan 2013
  • Loc: Charlton, Massachusetts

Posted 22 January 2013 - 02:58 PM

Does anybody know if the diameter and thread of polar scopes for aligning a mount are a common size? I was looking at some of the Celestron ones, but they don't have any measurement, They just say that they fit X mount.

-Adam

#34 hectar

hectar

    Explorer 1

  • -----
  • Posts: 53
  • Joined: 24 Dec 2012
  • Loc: Ontario, Canada

Posted 22 January 2013 - 06:14 PM

Alright, Sorry for the delay on the cutaway and dimensions.

First, the cutaway:
https://lh6.googleus...AAHa0/JjdttC...

As you can see, The gear end of the 2 axes are supported by 6218-2rs deep groove bearings. At the far end of the housing, there is another bearing with a bearing pre-load nut. This is tightened down to take all the backlash out of the bearings and provide a fluid axis. The RA axis will be set up the same way.

You can see at the top of the DEC that I have been playing with a clutch design. This is based on the clutch in the Losmandy G11. The tan layer is a piece of cork used as a slip/friction surface for the clutch pad. This would have a secondary shaft through the center of the main shaft and a hand wheel on the end to adjust clutch tension. The thing that worries me is if the end plate and shaft are not rigid enough, they could flex when the clutch is loose and cause inaccuracy. I think bushings would be sufficient for holding the clutch shaft concentric with the axis shaft.

If this is acceptable, I will replicate it on the RA axis.

Here is a rough dimensional drawing. The ones marked "True" are radiuses.

All measurements are in Inch's.

https://lh5.googleus...AAHa4/EaOkyF...


Im still looking for suggestions on the range of latitude the mount should be capable of.

-Adam


Adam,
I am trying to figure out your design. when designing with two bearings, one end is always "locked". ie one bearing should be: #1. locked on the shaft (ie it's inner race locked with a nut) , and, #2. Locked with in the casing. i.e. Its' outer race locked "with in the casing/housing" by means of a cap, leaving only 0.001-0.003" clearance for the thermal expansion of the bearing race itself.
Now, the other (2nd) end/bearing would be be still in the casing and bearing cap should be there but it will not be locked. it would have rather 0.032 to 0.125 or more gap between the outer race and cap.(Gap is necessary for the thermal expansion of the shaft in the axial direction)

Your design has lock nut in the bottom (so inner race is locked up against the shaft shoulder but outer race is not locked (no bearing cap there), what will stop it from falling? which way it expand due to heat or cold?? it also raises the question how/which order you will install those two 6218-2RS bearings.

In my opinion, 6218-2RS is not a thrust bearing. It is rather just a radial bearing. ( single row deep groove ball bearing with two seals )(suffix RS ). (Although, it can be used as a hobby project, but it wont be a perfect design).
Also, are you suggesting preloading it by the nut to somehow eliminate the bearing's own internal clearance?
I am thinking about posting a sketch here, but I would wait: what you and others think about this?
Btw, lattitude could be 0-90 degrees. if you end up with 70-80, it's ok

Hectar

#35 hectar

hectar

    Explorer 1

  • -----
  • Posts: 53
  • Joined: 24 Dec 2012
  • Loc: Ontario, Canada

Posted 22 January 2013 - 06:23 PM

Nice design Adam!
How do you plan to mount the worm gears on the shafts( if that is what you are going to use)?
It would be a good idea though to use roller bearings instead of ball bearings for a better precision and as a sophistication you could use taper roller bearings on the back end of the shaft, especially if you want to go with the preloading idea. Also it would be a more efficient way to deal with the axial and radial forces when the mount points in different positions. I believe Takahashi uses this set up in their mounts.
If you want to go Losmandy way then you should consider mounting the worm wheel on a bearing on the Ra (or DEC) housing itself making it independent to the shaft and avoid the small errors due to flexing or bearing runout. The most important thing I see with this set up is that the worm and worm wheel keep the same contact which will help accuracy. In this way the shaft is driven only by the friction between the worm wheel and the flange of the shaft.
http://www.astro.uni...xperiences.html

http://www.astromast...intenance_e.htm

Here's also a new ukrainian series of mounts using pretty much the same design with very good accuracy, not very hard to build on a lathe.
http://www.youtube.c...low=grid&view=0

I like your idea better. If someone can translate video, it would be great.

#36 Adam Brunette

Adam Brunette

    Sputnik

  • -----
  • Posts: 28
  • Joined: 13 Jan 2013
  • Loc: Charlton, Massachusetts

Posted 22 January 2013 - 08:14 PM

Hectar,

Some gaps and reliefs need to be updated. I have done extensive work on the design today, and hope to post a new picture some time tonight. I think it will clear up some questions.

I understand about the bearing concern. The 7218 angular contact ball bearing is a direct fit for the deep groove bearing, but the cost is alot higher. It is a good fit for the task though. There is also the 30218 which is a taper roller bearing.

There are options for the far end bearing as well. The part I don't like still is these type bearings need lubrication.

-Adam

#37 m. allan noah

m. allan noah

    Viking 1

  • -----
  • Posts: 752
  • Joined: 14 Aug 2009
  • Loc: Virginia, USA

Posted 22 January 2013 - 08:21 PM

In my opinion, 6218-2RS is not a thrust bearing. It is rather just a radial bearing. ( single row deep groove ball bearing with two seals )(suffix RS ). (Although, it can be used as a hobby project, but it wont be a perfect design).
Also, are you suggesting preloading it by the nut to somehow eliminate the bearing's own internal clearance?


Don't get caught up in the name 'thrust' vs 'radial' a deep groove ball bearing is not all that far removed from an angular contact bearing. I routinely use them in this type of service with a bit of preload.

However, I would say that the 6218 is really much bigger than needed. In fact, I would be inclined to forgo the rolling bearing entirely on the big end of each axis. I am partial to something more like the Discmount uses, either a disk or a cone of plane bearing.

allan

#38 m. allan noah

m. allan noah

    Viking 1

  • -----
  • Posts: 752
  • Joined: 14 Aug 2009
  • Loc: Virginia, USA

Posted 22 January 2013 - 08:55 PM

Oh, and perhaps this will be of some use to you:

http://translate.goo...efox-a&depth...

#39 hectar

hectar

    Explorer 1

  • -----
  • Posts: 53
  • Joined: 24 Dec 2012
  • Loc: Ontario, Canada

Posted 22 January 2013 - 09:29 PM

Re: 6218-2RS, I almost forgot the cost. since they can take some axial load in both directions, they may be ok.
Adam, you could use/accomodate a cheap automotive lip seal at either one/both bearing caps, which would contain the grease, thus allowing you to use roller/taper/cone bearing at one end. They do come with both races (as suggested above as well). You have to search for a cheaper one though.

Thanks for posting link Allan.

#40 Adam Brunette

Adam Brunette

    Sputnik

  • -----
  • Posts: 28
  • Joined: 13 Jan 2013
  • Loc: Charlton, Massachusetts

Posted 22 January 2013 - 10:15 PM

Allan,

Do you have an example of this Diskmount method?

I still think the 6218-2rs is be best low cost solution. It will never spin more then 1 RPM. They can be had new on ebay for $30 each. Im thinking I will start with these bearings and if they are not up to the job, there are 2 other options that are a direct fit.

Thanks for the link to that! Im looking through it now.

Hectar,

That's a good idea about the lip seals. I never thought of those.

-Adam

#41 Adam Brunette

Adam Brunette

    Sputnik

  • -----
  • Posts: 28
  • Joined: 13 Jan 2013
  • Loc: Charlton, Massachusetts

Posted 22 January 2013 - 11:42 PM

Alright, As I promised, here is an updated cut away view. This is incorporating the cork friction disks I mentioned earlier on. RA shaft is hollow to accept a polar scope. Counterweight shaft screws into the DEC shaft for modular approach.

Next part I plan to work on is the fork mount and base.

Enjoy!

Posted Image

Link for full rez version. https://docs.google....0Z4Ykhpdjg/edit

-Adam

#42 orlyandico

orlyandico

    Fly Me to the Moon

  • *****
  • Posts: 5357
  • Joined: 10 Aug 2009
  • Loc: Singapore

Posted 23 January 2013 - 01:15 AM

sounds like it will need a lathe to build...

this is the cheapest source of gears I've seen...
http://www.gototeles...s.com/gears.htm

although Edward Byers (seller ID erbyers on ebay) sometimes sells his fabulous gears on ebay. They are more expensive than the Aeroquest gears though.. but are a known good quantity. The earliest Paramount GT mounts used Byers gears.

#43 Startraffic

Startraffic

    Viking 1

  • *****
  • Posts: 744
  • Joined: 12 Feb 2006
  • Loc: Lat. 39.143345, Long. -77.174802

Posted 23 January 2013 - 06:56 AM

Ahopp,
Mine will be a 8'x16' bumper pull Obs with a 1kw gen, solar & batteries. The mount will be lowered through the floor to sit in the ground & isolated from the trailer which will be raised on jacks to level. It'll make for a nice stable portable obs. The 8' dome will be either a Skyshed POD or an automated 8' Home dome & will be towards the front of the trailer for balance & ease of mount operation. The electrical will be towards the back to counterbalance the dome, mount & scope. I hope to have a small warm room on the back but I haven't figured that out yet. I don't want to hijack this thread any further though.

Clear Dark Skies
Startraffic
39.138274 -77.168898

#44 neo

neo

    Surveyor 1

  • *****
  • Posts: 1645
  • Joined: 12 Feb 2008
  • Loc: Iasi, Romania

Posted 23 January 2013 - 07:46 AM

Oh, and perhaps this will be of some use to you:

http://translate.goo...efox-a&depth...


I like that for a home made mount!
That's basically the same setup used in the WS mounts in the Youtube videos, and what I was on about although the author used only needle bearings.

Adam you are right about the needle bearings but you can get ones which already have the inner and outer racers. Check the Timken catalog. I can't find the link right now :foreheadslap:

By the way very ingenious design, now that I see it more complete! But you have to admit it's not going to be easy to build without proper tooling. Al the shafts would need high precision machining to get good coaxiality.
Also, I'm not sure I might be wrong, but it looks like the worm wheels will take quite a bit of load, the outer/inner axles resting directly on them. Probably this could be much better judjed with a 'live' prototype and see how it tracks.
All in all a very good effort and quite different design from what most of us are used to, so can't wait for further refinements :waytogo:

#45 m. allan noah

m. allan noah

    Viking 1

  • -----
  • Posts: 752
  • Joined: 14 Aug 2009
  • Loc: Virginia, USA

Posted 23 January 2013 - 08:30 AM

It appears that you want to use the worm wheel as the container for the outer race. In use, there will be some outward force from the bearing. I think your thin-walled gear will be harder to make, and more likely to distort in that situation. I think you need to consider making the design more adaptable to gears from different sources, since that is the most likely highest cost component. Builders might want to scrounge a little, or might already own a gearset from another mount.

allan

#46 Adam Brunette

Adam Brunette

    Sputnik

  • -----
  • Posts: 28
  • Joined: 13 Jan 2013
  • Loc: Charlton, Massachusetts

Posted 13 August 2013 - 02:48 PM

Hello Everyone,

It has been awhile since I have posted anything related to this. However, This does not mean progress has stopped on it!

Allen's last post got me thinking and I decided to change a few things.

I have redesigned how the gears mount to the upper bearings. Instead of being cut with the axis hub, they will be separate gears, loc-tight'ed to the bearing. The hub will also be loc-tight'ed to the bearing. This allows anybody with a large enough gear to bore to 160mm (6.299") can fit them to the mount.

In the past few months, I have been sourcing parts to build the mount. I happened to get the chance to get a set of 9" Byers gears and worms directly from Ed Byers. I talked to him on the phone a few times and he said he retired 2 years ago and was just selling off his left over inventory. He does not have much left now and what is left is on his Ebay page.

Anyways, I was able to get 2 sets of 9" gears with worms mounted on 2 bearing blocks with custom bore for $650. While this is more then I wanted to spend on gears, I could not pass up the opportunity of using Byers gears in my mount.

I have mostly made refinements to the design I posted a few months ago. Shafts the main shafts of each axis is a standard diameter for linear bearing rails. These are very accurate in roundness and also are usually chrome plated. This is perfect for needle roller bearings. The RA clutch shaft is 40mm OD with a 22mm ID. This allows some amount of cables to pass through the RA. Its better then nothing I guess.

The azimuth adjustment is done with 2 precision set screws. These are 2" long and have 80 threads per inch. The push against a block in the base to twist the mount. When your in position, there is 6 locking levers that clamp the base plate with the azimuth plate. These levers are re-position-able so once locked, you can release them and rotate them out of the way. There is +/- 7.5 degrees of azimuth adjustment.

The altitude has been modified a good bit from the original design. The adjustment screw is also a 80 thread setscrew allowing for super accurate adjustments. The block that mounts the set screw is adjustable to 4 positions. The altitude ranges are 10-25, 25 - 40, 40-55, 55-70. I'm in the middle at 42 degrees. The main pivot of the RA axis is on 2 - 3/8" shoulder bolts, and is locked in place with 2.25" 5 lobe knobs.

The biggest changes are the gear box's / gear covers. The lower half of the gear cover is actually the structural mount of the worm gear and servo mount. These are probably some of the most expensive pieces of aluminum. Its about $360 for all the material for both of the gear box's. The lower part of the gear box is 1.5" thick milled to a shell with a 1/2" bottom. Its alot of material to remove, but no real better way to accomplish a tight cover that looks good. The lower cover is bored to fit nicely around the main tube of the axis and is secured to a mounting ring with 5 - 5/16 stainless steel bolts in a circle pattern. The mounting ring is also machined to fit the main tube snugly and is secured to the tube with 5 more 5/16 bolts. Both axes are set up this way.

The upper gear cover is also milled from solid aluminum with a whole bored to fit the hub of the axis. The hub rids on the main bearing. The worm wheel is loc-tighted to the bearing and secured to the hub with 5 screws.

Half of the upper cover is removable to allow easy access to the worm gear and servo components.

The worm gear and servo are mounted on a common pivot point and are held down with shoulder screws. This allows the whole worm/servo block to pivot to maintain constant pressure of engagement between the worm gear and wheel. This pressure is applied with 2 set screws with springs. These are accessible from outside of the gear casing.

The counter weight shaft is an off the shelf piece from Astro Phsyics. It is a 1.875" shaft. I liked the look of the shaft and counter weights and wanted to use these. The shaft can be re sized easily as long as it can thread into the DEC tube.

The mount can be taken apart at the intersection of the RA and DEC for easier transport. If wanted, the RA tube can be removed from the support arms by removing the 2 hand knobs and the 2 shoulder bolts. This would break it into 3 sections roughly about 30lbs each.

Here is a few older renders of the mount with a model of my Meade LX200 12" EMC on it as well as 2 - 18lb counter weights.

I have begun building this now and will post what I have done.

-Adam

Mount minus counter weight shaft:
https://lh3.googleus...AAM-M/2f1rPc...

Base of mount:
https://lh4.googleus...AAM-U/7EoGVD...

Meade 12" Tracking:
https://lh3.googleus...AALuc/1YRRsl...

Meade 12" almost home:
https://lh4.googleus...AALuo/r6nr9Z...

#47 Adam Brunette

Adam Brunette

    Sputnik

  • -----
  • Posts: 28
  • Joined: 13 Jan 2013
  • Loc: Charlton, Massachusetts

Posted 13 August 2013 - 03:04 PM

Here is a link to the latest render that just finished. You can see the servo/worm gear mount on the DEC axis.

https://lh5.googleus...AAM-w/duSZZN...

#48 Pinbout

Pinbout

    Cosmos

  • *****
  • Posts: 7596
  • Joined: 22 Feb 2010
  • Loc: nj

Posted 13 August 2013 - 05:33 PM

again this is open source atm'ing

I see your very nice renderings, but I haven't seen access to your model.

and strock puts up a pdf of the cut parts and hand drawings for open source

#49 BoriSpider

BoriSpider

    Vanguard

  • -----
  • Posts: 2082
  • Joined: 10 Aug 2004
  • Loc: S.W.FLA

Posted 14 August 2013 - 09:44 AM

Yes, access to the models would be nice. That way others can contribute.
You can always put them up on thingiverse.com or Thingtracker which is a github.com based deal(since MakerBot is no longer open-source-ish).

#50 Adam Brunette

Adam Brunette

    Sputnik

  • -----
  • Posts: 28
  • Joined: 13 Jan 2013
  • Loc: Charlton, Massachusetts

Posted 14 August 2013 - 09:55 AM

Can do! I will look into it. Currently it is in Solidworks.

-Adam






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics