Looking for some knowledge of K&E specials for NASA

rimcanyon --

From my dealings with K+E in the 1980s, their Corporate Memory was mostly the collective minds of Kurt Gmunder, Walter I-forget-his-last-name, and Ingrid Meyer. If I recall correctly, they all retired while K+E was under either Kratos or Cubic ownership, before Brunson bought K+E's skeleton.

And, if my recollection of the configuration of the K+E "surface table" autocollimator is correct, it is a single-axis instrument that can be rotated in its carrier through 90 degrees, which allows measurement of small-angle rotations either a) in the plane of, or b) out of the plane of the carrier's support feet. The micrometer knob / dial was graduated in 1/5 arcsecond intervals, with a full-revolution range of 30 arcseconds. The number of micromber-knob turns was indicated the the particular notch of a so-called "comb" fixed in the eyepiece's field of view the micrometer reticle line was occupying. I don't remember if the standard 71-4210 micrometer system was graduated from 0 through 20 arcminutes or 0 +/- 10 arcminutes, but I'm pretty sure that the total range was 20 arcminutes, in which case it would match yours.

It's pure speculation on my part, but it looks to me like your eyepiece is to convert an instrument similar to the 71-4210 to a two-axis micrometer instrument, obviating the need to rotate the telescope in its carrier to make measurements in perpendicular planes, with mechanical knob-revolution-counters added in place of the "combs". I suspect that the micrometer nearest the illumination unit moves a real reticle that generates the projected-and-reflected image, while the micrometer nearest the eyepiece lens would move a bi-filar reticle across the reflected image of the first reticle. There would be, of course, a beamsplitter at the intersection of the main and right-angle tubes of the attachment.

I wouldn't be particularly surprised if they eyepiece lens and the illumination attachment can be interchanged, to allow either right-angle or straight-through viewing.

It is perhaps a pedantic distinction, but the eyepiece micrometers of the type used in the K+E 71-4210 and in your attachment do indeed measure a linear distance, not a rotation. BUT the linear unit is not one of the officially-recognized-international-standard unit, but instead derived as the Tangent (knob-graduation interval, in this case 0.2 arcsecond) X focal length of objective lens system. The non-linearity of the sine and tangent functions is way below the "noise floor" of the micrometer screw and knob graduations for an angle as small as the micrometer range.

John

John, thank you for your reply. As always I learned a lot from it. I can shed a bit more light on the question, because I located another similar part partially disassembled and with some additional tags. I think it was originally the single axis projection attachment for the same model instrument. The short threaded section includes the reticle, and that part is not included with the dual axis eyepiece assembly. I assume the light source attached to that short piece. You can see the beamsplitter in one of the photos. It is also interesting that the MIT Instrumentation Lab was the original contractor that supported this instrument, but it was designed and mfd. by Kollsman instruments and it has a K&E tag. Possibly there is someone still around who might know more about it at MIT. I did have some Google hits on kollsman apollo autocollimator, which I am still following up, including this thread on PM: http://www.practicalmachinist.com/vb/metrology/autocollimator-repair-281482/