Autocollimator question about resolving power
Close
Login to Your Account
Page 1 of 2 12 LastLast
Results 1 to 20 of 25
  1. #1
    Join Date
    Mar 2005
    Country
    UNITED STATES
    State/Province
    Indiana
    Posts
    5,358
    Post Thanks / Like
    Likes (Given)
    1673
    Likes (Received)
    1595

    Default Autocollimator question about resolving power

    So I've been reading about these instruments a bit just seeing how they work. I see talk of 1/10 arc-second resolution for checking surface plates. Looking at specs of one of these I've seen for sale, a K&E 71-2022, it says resolving power is 3.4 arc-seconds. So is resolving power the same as resolution? That seems like it wouldn't be accurate enough if so. If they are not the same thing, what's the difference?

  2. #2
    Join Date
    Oct 2013
    Location
    Rochester, NY, USA
    Posts
    131
    Post Thanks / Like
    Likes (Given)
    4
    Likes (Received)
    50

    Default

    From looking at the manual, it appears that the 71-2022 is a multi-purpose instrument. It functions as an alignment auto-collimator, but not as a measuring auto-collimator. There is no micrometer reticle used to measure the auto-collimator image (which is how they get down to 0.1 arc sec in manual collimators).

  3. #3
    Join Date
    Mar 2005
    Country
    UNITED STATES
    State/Province
    Indiana
    Posts
    5,358
    Post Thanks / Like
    Likes (Given)
    1673
    Likes (Received)
    1595

    Default

    Quote Originally Posted by greif1 View Post
    From looking at the manual, it appears that the 71-2022 is a multi-purpose instrument. It functions as an alignment auto-collimator, but not as a measuring auto-collimator. There is no micrometer reticle used to measure the auto-collimator image (which is how they get down to 0.1 arc sec in manual collimators).
    Does it need a micrometer reticle when there are direct reading micrometer dials to move the crosshair into alignment?

    20200826_085629.jpg

    I am a little confused on why the dials read in thousandths also, I was under the impression that these were supposed to read angular deviation. So in order to figure angular deviation on something like a surface plate with a target that is moved along the surface I guess one would need to take the number of thousandths and then use that in conjunction with the distance to the target to calculate the angular deviation?

  4. #4
    Join Date
    Dec 2008
    Country
    UNITED STATES
    State/Province
    New York
    Posts
    10,394
    Post Thanks / Like
    Likes (Given)
    0
    Likes (Received)
    2703

    Default

    skinny crosshair 4 arc seconds wide you can barely see. then they say if you take 100 readings and average you can get 1/10 arc second readings.
    .
    but in reality if you had a 2 arc second crosshair you would NOT be able to see it that is it would blend in cause too thin. resolution is ability to see a line or is it too small and it blends in cannot see it
    .
    my experience is calibrating you get within 1/4 a cross hair width and if crosshair is 4 arc seconds that 1/4 is 1 arc second you doing good

  5. #5
    Join Date
    Oct 2013
    Location
    Rochester, NY, USA
    Posts
    131
    Post Thanks / Like
    Likes (Given)
    4
    Likes (Received)
    50

    Default

    This being a multi-purpose instrument, the micrometer dials are marked in 0.001" increments for use as an alignment telescope. They are not used for the autocollimator (from what I see in the manual!). Angular deviation in autocollimator mode is measured just by the reticle markings.

  6. Likes eKretz liked this post
  7. #6
    Join Date
    Dec 2008
    Country
    UNITED STATES
    State/Province
    New York
    Posts
    10,394
    Post Thanks / Like
    Likes (Given)
    0
    Likes (Received)
    2703

    Default

    resolving power think 1/100" ruler or scale. you reach a point where markings so thin hard to see or read.
    .
    some optical instruments have a 10 arc second crosshair. how I know 1 arc second is .006" wide at 100feet so 10 arc second is .006" wide at 10 feet. just looking at a 1/100" ruler at 10 feet you can measure width of crosshair
    .
    focused at infinity you dont talk about .001" only arc seconds

  8. Likes eKretz liked this post
  9. #7
    Join Date
    Mar 2005
    Country
    UNITED STATES
    State/Province
    Indiana
    Posts
    5,358
    Post Thanks / Like
    Likes (Given)
    1673
    Likes (Received)
    1595

    Default

    Quote Originally Posted by greif1 View Post
    This being a multi-purpose instrument, the micrometer dials are marked in 0.001" increments for use as an alignment telescope. They are not used for the autocollimator (from what I see in the manual!). Angular deviation in autocollimator mode is measured just by the reticle markings.
    I am not sure that's correct. I think the autocollimators or collimators without micrometer dials are used the way you describe, but when you have micrometer adjustment I don't think a graduated reticle is necessary, just a cross hair. As near as I can tell the autocollimator reticle on this particular instrument is only used for reference to ensure the mirror starts square to the instrument.

    After a little more google searching, it would appear that the linear displacement from this K&E can be converted to angular displacement by use of a formula involving the focal length of the instrument.

  10. #8
    Join Date
    Dec 2008
    Country
    UNITED STATES
    State/Province
    New York
    Posts
    10,394
    Post Thanks / Like
    Likes (Given)
    0
    Likes (Received)
    2703

    Default

    Quote Originally Posted by eKretz View Post
    I am not sure that's correct. I think the autocollimators or collimators without micrometer dials are used the way you describe, but when you have micrometer adjustment I don't think a graduated reticle is necessary, just a cross hair. As near as I can tell the autocollimator reticle on this particular instrument is only used for reference to ensure the mirror starts square to the instrument.

    After a little more google searching, it would appear that the linear displacement from this K&E can be converted to angular displacement by use of a formula involving the focal length of the instrument.
    .
    optical micrometer only works at focus near objects. at infinity it does nothing.
    .
    you can guess arc seconds if you know width of crosshair. with a theodolite you can measure arc seconds by pivoting scope and reading it.
    .
    in my experience most reticles are 3 to 10 arc seconds so (1/10 arc second) how to read 1/30 to 1/100 of something hard to see ?? its salesman talk saying you can measure to 1/10 arc second not mentioning each reading will vary much more. its a average of many readings thing.
    .
    you lucky to calibrate or repeat read to 1/4 the reticle width (4 arc second reticle thats 1 arc second) by the way at 100 feet air shimmers usually alot not unusual to see crosshair move more than a reticle width as the wind changes

  11. #9
    Join Date
    Nov 2014
    Country
    UNITED STATES
    State/Province
    Colorado
    Posts
    3,635
    Post Thanks / Like
    Likes (Given)
    4058
    Likes (Received)
    1058

    Default

    Ok Longer focal length autocollimators and better built / higher quality autocollimators easily read down to sub arc second... Usually the problem then becomes the field of view is too narrow that you need a second scope (shorter focal lengths scope) to find the target properly (depending on application.).

    Optically it's not so much a fat cross hair... Doesn't really work like that.

    Some useful links,



    Vermont Photonics - High Quality Optical Test Equipment

    ^^ General

    Uses for Optical Equipment | Vermont Photonics

    ^^^ Applications,

    https://www.vermontphotonics.com/VT_...ns.pdf#page=32

    Applications hand book ^^^

    https://www.vermontphotonics.com/VT_...ons.pdf#page=1

    Page 1 ^^^

    There are many older and current and still manufactured autocollimators that have very long focal lengths.

    Regular older Hilger and Watts and Brunson- Davison autocollimators are still very good / sought after.

    Buy new from Nikon, ZEISS specialized integrated systems and MÖLLER-WEDEL

    Welcome to MOLLER-WEDEL OPTICAL | Measuring Instruments You Can Trust!


    __________________________________________________ __________________________________________________ _


    * Brunson --- > this might help - https://www.brunson.us/2022-alignment-telescope.html

    K+ E manual http://www.brunson.us/media/assets/p...ManKE_4210.pdf

    ^^^ This autocolimator is seen a lot for surface plates etc.

    K + E more "Modern explanations " and use http://www.brunson.us/media/assets/p...ManKE_2022.pdf


    __________________________________________________ _______________


    Stuff from Davisdon and Trioptic GMBH.

    http://davidsonoptronics.com/pdfs/D-275-AAT-WW.pdf

    http://trioptics-usa.com/automatic-alignment-telescope/

    __________________________________________________ _____________________


    https://www.haag-streit.com/moeller-...s/collimators/

    Moeller wedel again but pages on selection and application and different size and focal length of of regular collimators ^^^

    https://www.haag-streit.com/moeller-...tocollimators/

    ^^^ VS > How autocollimators work and which and how to select.

  12. Likes sfriedberg, eKretz, gregormarwick liked this post
  13. #10
    Join Date
    Nov 2014
    Country
    UNITED STATES
    State/Province
    Colorado
    Posts
    3,635
    Post Thanks / Like
    Likes (Given)
    4058
    Likes (Received)
    1058

    Default

    Quote Originally Posted by eKretz View Post
    So I've been reading about these instruments a bit just seeing how they work. I see talk of 1/10 arc-second resolution for checking surface plates. Looking at specs of one of these I've seen for sale, a K&E 71-2022, it says resolving power is 3.4 arc-seconds. So is resolving power the same as resolution? That seems like it wouldn't be accurate enough if so. If they are not the same thing, what's the difference?
    Resolution "Old school" for telescopes usually point back to what's known as the "Raleigh criterion"

    Angular resolution - Wikipedia

    Basically what is the angular distance you can separately resolve two fuzzy dots usually created through interference patterns/ diffraction gratings.

    rayleigh-fuzzy-blobs.jpg <--- Click to blow up.


    ^^^ Old school resolution.

    More modern resolution concepts use the frequency domain and contrast - essentially a MTF or modulation transfer function *

    I have ref for that in a mo.

    A bit like what Mr Tomb has said you can fudge and align things better but the over all optical design (assuming you have the right instrument) should not be affected by the thickness of the cross hair unless you have a very low resolving power and low focal length inaccurate autocollimator ?


    IS this for correcting old and worn surface plates ? Or you have other uses ?

    __________________________________________________ _________________________________________________

    * Trying to find the free PDF on "Modulation Transfer Function in Optical and Electro-Optical Systems"
    Author(s): Glenn D. Boreman - It's good read well explained even if you are not into calculus and deep physics , it's pretty understandable to a "technical " layman also. Book's intro chapters help delineate 'old school" resolution from (Rayleigh etc. from more modern ways of looking at optical / electro optical systems in their totality.). (the Free PDF is out there somewhere.).

    ++ I think "Foundations of Mechanical Accuracy " (Wayne R Moor) also handle this pretty well also using 1960's instruments/ autocollimators (Hilger Watts feature also If memory serves me right). .

  14. Likes eKretz liked this post
  15. #11
    Join Date
    Nov 2014
    Country
    UNITED STATES
    State/Province
    Colorado
    Posts
    3,635
    Post Thanks / Like
    Likes (Given)
    4058
    Likes (Received)
    1058

    Default

    Quote Originally Posted by eKretz View Post
    So I've been reading about these instruments a bit just seeing how they work. I see talk of 1/10 arc-second resolution for checking surface plates. Looking at specs of one of these I've seen for sale, a K&E 71-2022, it says resolving power is 3.4 arc-seconds. So is resolving power the same as resolution? That seems like it wouldn't be accurate enough if so. If they are not the same thing, what's the difference?
    Sooo

    Visual Autocollimator Type AKW, AKW MD & MDD | 60deg-Viewing | Double Micrometer | Fixed Focus Setting | MOLLER-WEDEL OPTICAL GmbH

    This is one of their most compact autocollimators short tube 300 mm focal length ---> 0.5" arc second resolution (manual optical)

    Another one short tube compact 500 mm focal length ---> 0.2" arc second resolution (manual optical).

    https://www.haag-streit.com/fileadmi...troduction.pdf

    ^^^ Simple calculation of angles.

    scale deviation of the shortest and lowest res scopes / autocollimators - standard deviation on micrometer scale of one division ≈ 2 arc seconds (300 mm focal length)

    and for 500 mm autocollimator one micrometer division ≈ 1 arc second.

    On the other hand a cheap short focal length Autocollimator would or could be of the order of 8" ... Depends on the application.

  16. #12
    Join Date
    Jan 2007
    Location
    Flushing/Flint, Michigan
    Posts
    9,490
    Post Thanks / Like
    Likes (Given)
    521
    Likes (Received)
    7878

    Default

    Rayleigh limits and MTF .. This is all getting so confusing with fancy terms.
    Bob
    (do not bother explaining such, just trying to make a point)

  17. #13
    Join Date
    Nov 2014
    Country
    UNITED STATES
    State/Province
    Colorado
    Posts
    3,635
    Post Thanks / Like
    Likes (Given)
    4058
    Likes (Received)
    1058

    Default

    Quote Originally Posted by CarbideBob View Post
    Rayleigh limits and MTF .. This is all getting so confusing with fancy terms.
    Bob
    (do not bother explaining such, just trying to make a point)
    Are you trying to 'poop" on me Mr Bob

    It is what it is, refs are all there - the easy PDF are there - resolving two fuzzy blobs and the angular distance between the two is all there , and the effect of micrometer adjustment is there too.

    The Hilger Watts type autocollimators commonly used for surface plate " (resurfacing) are more than sufficient.

    BUT just trying to provide ample proof that sub arc second is not a problem. Assuming you have the right scope and eye piece - cross hairs are not the limiting factor unless you have the wrongs scope.

    OP seemed to want to generally browse and research find out more about auto-collimators

    I'll dig up that MTF reference 'cuz I think you Mr CrabideBob would / might like it. It's well written and easy to understand and not too long. (The PDF of that book used to FREE on the SPIE(E) website but seems not to be anymore / harder to find (I have hard copy).

    @eKretz has a brain I assume and he specifically asked about " Resolving power on an autocollimator" vs sensitivity of measurement of small angles and accuracy.

    Best thing to do is get your hands on an autocollimator and try things out.

    Not sure if eKretz is fixin to buy one or not ? Or has an application he's thinking of.

    The key graphic / easy to understand picture dropped out ?

    Maybe I fix that.

    rayleigh-fuzzy-blobs.jpg

    ^^ Click to enlarge

    ^^^ "Resolution" of two distinguishable points (airy disks) separated by a minimum angle to distinguish two fuzzy blobs from one big fuzzy blob that appears to merge into one.

  18. Likes eKretz liked this post
  19. #14
    Join Date
    Dec 2008
    Location
    Vt USA
    Posts
    8,871
    Post Thanks / Like
    Likes (Given)
    1460
    Likes (Received)
    3143

    Default

    I've always wanted to mount my K&E auto collimator on an AR-10...

    What an OPTIC! They are well made!

    Doubt it would handle the shock. I've got the unit, the illuminator, and an infinite target Nice kit! Vee blocks and the 4' x 5' granite makes for some precision optical alignment possibilities ;-)

    I must say I use the Path 5D825 optical level more often. It's a Great spotting scope ;-)

  20. Likes cameraman liked this post
  21. #15
    Join Date
    Nov 2014
    Country
    UNITED STATES
    State/Province
    Colorado
    Posts
    3,635
    Post Thanks / Like
    Likes (Given)
    4058
    Likes (Received)
    1058

    Default

    Quote Originally Posted by CalG View Post
    I've always wanted to mount my K&E auto collimator on an AR-10...

    What an OPTIC! They are well made!

    Doubt it would handle the shock. I've got the unit, the illuminator, and an infinite target Nice kit! Vee blocks and the 4' x 5' granite makes for some precision optical alignment possibilities ;-)

    I must say I use the Path 5D825 optical level more often. It's a Great spotting scope ;-)
    Funny you should mention that I put "sports optics" on some the cameras I have built lol. It can be a cheaper way of getting there / field alignment.

    Typically the barrel of some of those autocolimators are centerless ground so that's where the optics has to align accurately to contacting high precision surfaces as an optical "tool"(or at least can .).

  22. #16
    Join Date
    Dec 2008
    Location
    Vt USA
    Posts
    8,871
    Post Thanks / Like
    Likes (Given)
    1460
    Likes (Received)
    3143

    Default

    Quote Originally Posted by cameraman View Post
    Funny you should mention that I put "sports optics" on some the cameras I have built lol. It can be a cheaper way of getting there / field alignment.

    Typically the barrel of some of those autocolimators are centerless ground so that's where the optics has to align accurately to contacting high precision surfaces as an optical "tool".
    The certerline of the K&E unit I have is within a gnats ass of perfect to the center of the OD of the cylinder.
    There is a number , some zeros right of the decimal, metyric or imperial, makes no difference. gnats are small!

    The Infinite target is fitted with a tapered sleeve* to fit some particular propeller shafting. Not sure what "ship" in particular it's made for, But it's all as good as any machine tool. *(couple -a sizes up from a CAT 50 ;-)

  23. Likes cameraman liked this post
  24. #17
    Join Date
    Nov 2014
    Country
    UNITED STATES
    State/Province
    Colorado
    Posts
    3,635
    Post Thanks / Like
    Likes (Given)
    4058
    Likes (Received)
    1058

    Default

    Quote Originally Posted by CalG View Post
    The certerline of the K&E unit I have is within a gnats ass of perfect to the center of the OD of the cylinder.
    There is a number , some zeros right of the decimal, metyric or imperial, makes no difference. gnats are small!

    The Infinite target is fitted with a tapered sleeve* to fit some particular propeller shafting. Not sure what "ship" in particular it's made for, But it's all as good as any machine tool. *(couple -a sizes up from a CAT 50 ;-)
    LOL

    I come originally from the UK - we used to say gnats "cock" - I guess a gnat's arse could bigger ? Got get this stuff right RIGHT ??? !! It's the Metrology forum.

  25. Likes CalG liked this post
  26. #18
    Join Date
    Mar 2005
    Country
    UNITED STATES
    State/Province
    Indiana
    Posts
    5,358
    Post Thanks / Like
    Likes (Given)
    1673
    Likes (Received)
    1595

    Default

    Yes, was thinking about picking one up if I get a chance at one cheap for playing around with, possibly surface plate checking. And those kinds of links are exactly what I was looking for, thank you cameraman.

  27. #19
    Join Date
    Nov 2014
    Country
    UNITED STATES
    State/Province
    Colorado
    Posts
    3,635
    Post Thanks / Like
    Likes (Given)
    4058
    Likes (Received)
    1058

    Default

    Quote Originally Posted by eKretz View Post
    Yes, was thinking about picking one up if I get a chance at one cheap for playing around with, possibly surface plate checking. And those kinds of links are exactly what I was looking for, thank you cameraman.
    Yeah no worries , very welcome.

    Randomly seeing what's available on line; some folks know what these instruments are worth $$$$$$ +++ , and sometimes you see some pretty good deals on more unusual autocollimators where some work / risk is required.

    Davidson Optronics D-600 Comparison Autocollimator

    This is from BMI surplus* a "Davidson Optronics D-600 Comparison Autocollimator "

    For only $750.00

    davidson-d-600.jpg

    ^^^ Click to enlarge.


    But not officially tested and probably have to build various mounts for it.


    But the basic technical specs and capability as outlined as follows / blurb to give you an idea of sensitivity versus accuracy (500 mm / 508 mm / 20 inch - focal length).

    __________________________________

    "
    "
    This used Davidson D-600 Comparison Autocollimator comes from a working environment but has not been tested. Please call us for more information.

    From Manufacturer Website:

    - Measure angular differences between two reflecting surfaces to 0.1 arc second
    - Readings are not affected by instrument movement
    - Extremely simple to check surface plates or machine tool alignment
    - Measure about either vertical or horizontal axis

    Model D-600 Comparison Autocollimators apply a unique principle of autocollimation to compare and measure small angles with extremely high precision. Specifically, the instrument measures the difference in angle between two reflecting surfaces. A beam of collimated light from the instrument is divided to illuminate a reference mirror and a second mirror whose relative angle is to be determined. The patented design employs a unique arrangement of optical tipping plates coupled to a calibrated micrometer dial, where the relative angle is read when the images are aligned. Differences in angle of the returning beams can be measured with average repeatability of 1/10 arc seconds, either vertically or horizontally. The micrometer dial is graduated in 1/10 arc seconds, with each second numbered. One complete revolution of the dial is equal to 10 arc seconds with a total range of 120 arc seconds. Of utmost importance to precision is the fact that instrument movement during reading will only move the images in the field of view; it will have absolutely no effect on the reading. This fact eliminates the rigid mounting and extreme handling care demanded by other instruments, where any motion is translated directly into the micrometer reading.

    Manufacturer Specifications:

    - Sensitivity: 1/10 arc sec
    - Accuracy: 0.5 arc sec
    - Focal Length: 20 inch (508 mm)
    - Aperture Diameter: 2.375 inch (60 mm)
    - Measuring Range: 120 arc sec
    - Micrometer Scale: Graduation 1/10 sec One revolution equals 10 sec

    - Housing Material: Aluminum
    - Overall Dimensions: (L x W x H) 23 x 5.375 x 5.125 inch (584 x 137 x130 mm)
    - Finish: Ivory enamel
    - Carrying Case: Hardwood

    This unit has not been calibrated. Buyer will need to source their own calibration.

    ____________________________________-


    ^^^ Just to give you a feel of what autocollimators of different vintages and focal lengths are capable of doing.


    __________________________________________________ ____________________________

    * No affiliation.

  28. Likes eKretz liked this post
  29. #20
    Join Date
    Nov 2014
    Country
    UNITED STATES
    State/Province
    Colorado
    Posts
    3,635
    Post Thanks / Like
    Likes (Given)
    4058
    Likes (Received)
    1058

    Default

    hilger-watts-xx.jpg

    ^^^ Click to enlarge,

    ^^ This is a pretty common Hilger and Watts "Microptic" autocollimator

    Those can still be found...

    But generally autocollimators will be of the order of thousands of dollars and in some cases tens of thousands of dollars rather than hundreds of dollars. (New or second hand that seems to be the case.).


    That one / similar appears in "Foundations of Mechanical Accuracy" page 72 and 73 - describes it's use for mapping surface plates and re-lapping. Also shows various plots and the techniques used.

    Also shows how the optics stack up and are designed and laid out. (page 72). The micrometer is connected to a "refracting block" (sort of Brewster prism effect except shifts the image more in parallel.).

    Later on of course they show the same basic autocollimator being used with various other instruments in combination like optical polygons and the like. (pages 230, 239-240 ).

    For example the knarly looking Hilger and watts instrument was used by MOORE Tool Company to calibrate / produce their 1440 "Small Angle Divider" / indexer, that things is / was accurate to +/- 0.5 arc seconds full circle and has approximately 0.1 arc second effective repeatability... (not easy to prove lol )- But gives an indication of how the basic sensitivity and potential accuracy of an autocollimator can be transferred indirectly to physically and practically built things - like a mechanical indexing plate of extreme accuracy.

    There are much more precise autocollimators out there also.

    DMF_TomB is right about the thermal effects over long distances. (Thumbs up).

    __________________________________________________ __________________________________________________ _______


    *** Appears that Keith Rucker no less uploaded the Hilger and Watts "Microptic" - whole manual - scanned really good fidelity / resolution.

    Hilger & Watts - Publication Reprints - Watts Microptic Auto-Collimator - Instructions for Use | VintageMachinery.org

    http://vintagemachinery.org/pubs/11387/21598.pdf

    ^^^ Scanned manual (beautifully illustrated) and shows the operation of the instrument really well and really well how to be used for surface plates (Nice !) + basic calculations to be performed.

    So they state in the manual with practice 0.1 arc second repeatability is possible and if you look at the optical diagrams you can SEE how the cross hairs are part of a second optical assembly that has a magnification of approximately 40X and that itself is picking off a much smaller / narrow field angle from the returned "bundle" of rays. Page 3 This viewing assembly optically is essentially a microscope, with cross hairs towards the eye piece.

  30. Likes eKretz, Ethan Scott liked this post

Tags for this Thread

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •