Jump to content


Photo

Pros and cons of mirror cell designs

  • Please log in to reply
28 replies to this topic

#1 B9Robot

B9Robot

    Explorer 1

  • -----
  • Posts: 64
  • Joined: 08 Oct 2013

Posted 11 October 2013 - 11:30 PM

I am looking at Reiner Vogel's mirror cell design in which the mirror supports are attached to a frame, and then the whole frame is moves for collimation.

http://www.reinervog...g/3_MBox_e.html

I am wondering about the pros and cons of this design as compared to something more traditional like the Kriege and Barry book.

One advantage is that collimation adjustment is on the front, which is convienient. For those of you who have selected this design - what do you think?

I am also curious about this design has one point of the traigle cell fixed, and then only two adjustment points for collimation. Good? Bad?


thanks!

#2 hbanich

hbanich

    Apollo

  • *****
  • Posts: 1110
  • Joined: 17 Jun 2005
  • Loc: Portland, Oregon

Posted 12 October 2013 - 01:06 AM

I made a cell much like this for my 28 inch f4 scope that I copied from Dan Gray's 28 inch f4.5, and have been using it for the past nine years. Here's my take:

Pros
1. No worries about the edge support introducing astigmatism when adjusting collimation.
2. Low profile because the main support arms for the back support triangles are almost flush with the mirror cell frame. This can allow for a narrower mirror box.
3. The collimation knobs can be either in the front or the back - mine are in the back - but can be conveniently placed in the corners of the mirror box for easy access. Putting the third point at the bottom and on a pivot works well, but even though it works well it does prevent moving the mirror forward or back in the mirror box to fine tune where the focal plane ends up in the focuser.
4. Easy and inexpensive to make.

Cons
1. Haven't found any.

Attached Files



#3 Pinbout

Pinbout

    Cosmos

  • *****
  • Posts: 7760
  • Joined: 22 Feb 2010
  • Loc: nj

Posted 12 October 2013 - 07:32 AM

I think its an excellent design if you like wiffle trees.

the only con is it add to the eye piece height.

#4 kfrederick

kfrederick

    Gemini

  • *****
  • Posts: 3011
  • Joined: 01 Feb 2008

Posted 12 October 2013 - 07:52 AM

On the chief cell I have the whole plate move[The cross bars for the triangles are mounted under thie plate].I have changed the side supports to Glatter linear bearings

Attached Files



#5 B9Robot

B9Robot

    Explorer 1

  • -----
  • Posts: 64
  • Joined: 08 Oct 2013

Posted 12 October 2013 - 10:37 AM

Thanks for the pictures guys - looks great. I really like this design.

I wonder about collimation adjustment. Because the whole frame moves, one corner of the frame triangle can be at a different height compared to the other corner. Is there a problem with the collimation adjustment screws to bind on the nuts on the frame (because they are at an angle)? Or is the amount of adjustment so small, that it is taken up by any slop in the system.

thanks
Don

#6 hbanich

hbanich

    Apollo

  • *****
  • Posts: 1110
  • Joined: 17 Jun 2005
  • Loc: Portland, Oregon

Posted 13 October 2013 - 02:36 AM

Inside the mirror cell frame at each corner I have a threaded spherical nut made from Delrin that can rotate. Each collimation bolt passes through one of these spherical nuts which eliminates any chance of binding.

I have some close up photos here - scroll about halfway down: http://tinyurl.com/kkrjlh8

#7 kfrederick

kfrederick

    Gemini

  • *****
  • Posts: 3011
  • Joined: 01 Feb 2008

Posted 13 October 2013 - 04:49 AM

Howard I like your design alot

#8 hbanich

hbanich

    Apollo

  • *****
  • Posts: 1110
  • Joined: 17 Jun 2005
  • Loc: Portland, Oregon

Posted 13 October 2013 - 11:12 PM

Howard I like your design alot


Thanks, it's worked out pretty well!

#9 jhayes_tucson

jhayes_tucson

    Viking 1

  • *****
  • Posts: 995
  • Joined: 26 Aug 2012
  • Loc: Bend, OR

Posted 14 October 2013 - 12:17 AM

Good stuff guys...these cells look quite nice. The only thing that I personally don't' like about most ATM mirror cells is that the adjustment screws are often positioned at 120 degrees (like I see here.) I personally don't like having the adjustments coupled when I'm aligning an optical system. So, let me suggest you give some thought to making the adjustments orthogonal by placing the screws positions at right angles. It is actually pretty easy to do in most cases and it will make your alignment process a great deal easier--even if you have to space things so that the X,Y screws have to act at slightly different moment arms. Try it, I think that you will like it.
John

PS Alignment gets really easy if you can do the same thing with the secondary mount AND you make the alignment axis for each of the two mounts coincident.

#10 MitchAlsup

MitchAlsup

    Surveyor 1

  • -----
  • Posts: 1524
  • Joined: 31 Aug 2009

Posted 14 October 2013 - 11:39 AM

Good stuff guys...these cells look quite nice. The only thing that I personally don't' like about most ATM mirror cells is that the adjustment screws are often positioned at 120 degrees (like I see here.) I personally don't like having the adjustments coupled when I'm aligning an optical system. So, let me suggest you give some thought to making the adjustments orthogonal by placing the screws positions at right angles.


When I made my mirror cell (circa 1999), I ran into a patent that describes a way of articulating the cell at right angles.

#11 B9Robot

B9Robot

    Explorer 1

  • -----
  • Posts: 64
  • Joined: 08 Oct 2013

Posted 14 October 2013 - 01:41 PM

Good stuff guys...these cells look quite nice. The only thing that I personally don't' like about most ATM mirror cells is that the adjustment screws are often positioned at 120 degrees (like I see here.) I personally don't like having the adjustments coupled when I'm aligning an optical system. So, let me suggest you give some thought to making the adjustments orthogonal by placing the screws positions at right angles. It is actually pretty easy to do in most cases and it will make your alignment process a great deal easier--even if you have to space things so that the X,Y screws have to act at slightly different moment arms. Try it, I think that you will like it.
John

PS Alignment gets really easy if you can do the same thing with the secondary mount AND you make the alignment axis for each of the two mounts coincident.


Can you give me more of an idea of how this will work - maybe a picture, if you have one? I am having a hard time visualizing it.

thanks
Don

#12 Pinbout

Pinbout

    Cosmos

  • *****
  • Posts: 7760
  • Joined: 22 Feb 2010
  • Loc: nj

Posted 14 October 2013 - 05:17 PM

check this out

http://aurorap.com/PDF/MC0002.pdf

#13 B9Robot

B9Robot

    Explorer 1

  • -----
  • Posts: 64
  • Joined: 08 Oct 2013

Posted 14 October 2013 - 07:14 PM

Very cool - thanks for the link. In concept its similar to Reiner's mirror cell - one point fixed; two other points do adjustment. However the use of spherical bearings, and having the thread fixed is a much simpler idea.

#14 davidpitre

davidpitre

    Soyuz

  • *****
  • Posts: 3574
  • Joined: 10 May 2005
  • Loc: Central Texas

Posted 14 October 2013 - 10:14 PM

My cell is an 18 point that is of very similar design. I like it a lot and after 4 years can see no disadvantages. It uses a wiffle-tree (not sure why someone would not like them) that supports the relatively thin mirror very well with no suggestion of astigmatism. Mine allows front collimation which I appreciate. I can see the reflected image of the barlowed laser as I collimate the primary.
Contrary to the assertion, the design allows for a low profile cell that can be shorter than a typical Kriege cell.

#15 jhayes_tucson

jhayes_tucson

    Viking 1

  • *****
  • Posts: 995
  • Joined: 26 Aug 2012
  • Loc: Bend, OR

Posted 16 October 2013 - 12:26 AM

Can you give me more of an idea of how this will work - maybe a picture, if you have one? I am having a hard time visualizing it.

thanks
Don


Don,
I've attached a sketch I was working on for an 8" Cassegrain telescope mirror mount. It is rough and shows just the adjustment plate in the back of the tube. The blue circles are where the adjustment screws go. Note that they are at right angles relative to the pivot screw. The pivot is near the center because this is a perforated mirror and it is center mounted. In principle the pivot could go at the center of a Newtonian but I'd prefer to spread out the mounting points to get a longer moment arm and to make it more stable so I'd keep the pivot off center. You'll have to add the wiffle tree to this plate and have a support plate (as with any design) but I'll leave those details to you.
John

Attached Files



#16 B9Robot

B9Robot

    Explorer 1

  • -----
  • Posts: 64
  • Joined: 08 Oct 2013

Posted 16 October 2013 - 02:24 PM

John - thanks for the clarification and the attachment. I understand what you now mean by having the adjustment screws at 90 degrees to each other in relation to the pivot point. Very interesting!

In my current design the adjustment screws are about 55 degrees apart as compared to the pivot point. The pivot is at the top centre of the frame and the adjustment points are in the lower two corners.

If I understand you correctly, if I have the points at 90 degrees to each other, then collimation wont mess with alignment?

#17 jhayes_tucson

jhayes_tucson

    Viking 1

  • *****
  • Posts: 995
  • Joined: 26 Aug 2012
  • Loc: Bend, OR

Posted 16 October 2013 - 07:04 PM

Don,
Placing the screws at 90 degrees means that when you adjust the angle of the mirror, you will only touch two screws, which tilt the mirror independently. You will find that this is MUCH easier than when you have to adjust three screws that are all coupled. "Coupled" means that when you adjust screw #1, you also have to adjust screw #2 or reset screw #3 iteratively until you get it where you want. When you have both the primary and the secondary mounted with orthogonal adjusters that are coincident, alignment (aka "collimation") is trivial.
John

PS For some reason, in the world of amateur astronomy, the word "collimation" is used to mean alignment. In the world of optics, the word "collimation" means to adjust the spacing between elements to create parallel rays. I'm not out to change anything, but you will never see me refer to the act of aligning a telescope as "collimation" -- unless there is an eyepiece involved and we are also adjusting the spacing between the objective and the eyepiece. ;)

#18 B9Robot

B9Robot

    Explorer 1

  • -----
  • Posts: 64
  • Joined: 08 Oct 2013

Posted 17 October 2013 - 11:58 AM

Hi John, I am thinking for this setup to work, the pivot point must be in the centre of the mirror. Do I understand that correctly?

Don

#19 careysub

careysub

    Surveyor 1

  • *****
  • Posts: 1900
  • Joined: 18 Feb 2011
  • Loc: Rancho Cucamonga, CA

Posted 17 October 2013 - 12:52 PM

Hi John, I am thinking for this setup to work, the pivot point must be in the centre of the mirror. Do I understand that correctly?

Don


No, there is no dependency at all on the pivot point being at, near, or in-line with the center.

Here is a nice write-up on the related principles of kinematic mirror support.

These principles are used to build high precision, highly repeatable mirror adjustment systems and are applicable to both primary and secondary mirror mounts.

Briefly put - kinematic support depends on adjustments that touch the mirror plate at only three places (some force-generating mechanism must hold the plate against the supports typically).

Each point of adjustment contact has a particular geometry.
The first point consists of a sphere seated in a conical cavity, this allows the sphere to rotate, but not move around - it is captured by the cavity.
The second point consists of a sphere seated in a vee-shaped trench pointed directly at the first point. This allows the sphere to slide radially with respect to the first point.
The third point is a sphere resting on a flat plate. It is free to move however it likes.

This system uniquely and non-redundantly defines the position of the mirror plate in 3D space, without generating any stresses that might distort the plate.

The use of a captive ball, or spherical bearing pivot, is the implementation of the ball-in-cone contact point in the basic model.

If you look at a picture of a kinematic mount:
http://www.newport.c...4/1033/info....
you will see the right-angle support point arrangement that John Hayes describes.

And if you look at this mount:
http://www.newport.c.../1033/info.aspx
you will see that they omitted one of the adjustment screws - instead it uses the fixed pivot, which is located at the corner of mirror support system, not in the center. This is the same layout, scaled down, that John has been describing.

#20 B9Robot

B9Robot

    Explorer 1

  • -----
  • Posts: 64
  • Joined: 08 Oct 2013

Posted 18 October 2013 - 12:18 PM

Thanks for the detailed reply. Good information!

So to get the complete advantage of this system would both the primary and the secondary need to be in the same relative position? E.g. both at the centre or both in the upper left corner, etc?

#21 jhayes_tucson

jhayes_tucson

    Viking 1

  • *****
  • Posts: 995
  • Joined: 26 Aug 2012
  • Loc: Bend, OR

Posted 18 October 2013 - 02:20 PM

Thanks Carrey for jumping in with a great answer. Yes, co-aligning the tilt axis for each mount will make telescope alignment very easy. Let us know how it works out when you get it done.
John

#22 Starman1

Starman1

    Vendor (EyepiecesEtc.com)

  • *****
  • Vendors
  • Posts: 22871
  • Joined: 23 Jun 2003
  • Loc: Los Angeles

Posted 22 October 2013 - 12:50 PM

John:
col·li·mate [kol-uh-meyt]
verb (used with object), col·li·mat·ed, col·li·mat·ing.
1.
to bring into line; make parallel.
2.
to adjust accurately the line of sight of (a telescope).

Both definitions apply to reflective optics, since two axes (focuser and primary mirror) are being made coincident.

#23 Pinbout

Pinbout

    Cosmos

  • *****
  • Posts: 7760
  • Joined: 22 Feb 2010
  • Loc: nj

Posted 22 October 2013 - 05:04 PM

Don, excuse me, but you forgot the etymology of collimate.

The word "collimate" comes from the Latin verb collimare, which originated in a misreading of collineare, "to direct in a straight line".

#24 jhayes_tucson

jhayes_tucson

    Viking 1

  • *****
  • Posts: 995
  • Joined: 26 Aug 2012
  • Loc: Bend, OR

Posted 23 October 2013 - 12:37 AM

John:
col·li·mate [kol-uh-meyt]
verb (used with object), col·li·mat·ed, col·li·mat·ing.
1.
to bring into line; make parallel.
2.
to adjust accurately the line of sight of (a telescope).

Both definitions apply to reflective optics, since two axes (focuser and primary mirror) are being made coincident.


Don,
As I said, I'm not out to change anything. Just understand that if you are talking to a professional optical engineer who is not also an amateur astronomer, you will get a confused look when you talk about colllimating optical components,by simply tilting them. Optical engineers collimate a beam of light by rendering the rays parallel--usually by changing component spacings and they align mirrors by tilting them. I was an amateur telescope builder long before I was a professional engineer but I still have trouble even thinking about collimating a telescope by solely tilting the mirrors. No big deal...I can translate. I just want you guys to know the difference.
John

#25 Jon Isaacs

Jon Isaacs

    ISS

  • *****
  • Posts: 43894
  • Joined: 16 Jun 2004
  • Loc: San Diego and Boulevard, CA

Posted 23 October 2013 - 06:21 AM

Placing the screws at 90 degrees means that when you adjust the angle of the mirror, you will only touch two screws, which tilt the mirror independently.



Using 4 screws means that the cell is "over constrained" and that it is possible/probable to induce bending in the mirror cell. A secondary mount this is commonly done but secondary mounts are very stiff. I am not so sure about the wisdom of over constraining a relatively flexible/compliant mirror cell.

Myself, I am fine with the 3 point support. I just leave one alone and adjust the other two..

Jon






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics