practical use of an autocollimator and single mirror for mapping a surface plate

[metalmagpie]

I just brought home a Hilger and Watts autocollimator and a Nikon mirror on a 5" base. I want to use it to map out a 36x36" surface plate.

I have two questions.. First, given that I have drawn my British flag pattern with 7 measurements along an edge (not diagnonal), I believe the autocollimator is set along the line of measurement, with the mirror set at the 4th measurement i.e. along the middle of the line. I believe the autocollimator is zeroes on that mirror position and then the mirror is moved to the other 3 locations along the far end of that first line, taking measurements. Then do I move the autocollimator to the other end of the same line, put the mirror back in the middle position, and zero the autocollimator again and take the readings from the second half of the line along the edge? I believe this is what the Standridge guys did in Oxtool's video. I'm trying to see how data taken from two autocollimator positions can accurately reflect the surface. I hope the question is clear. Thanks for any clarification.

Second, how do I deal with the fact that I cannot get an integer number of intervals along a diameter, not and have the interval be the same as the one used along an edge line?

Thank you. BTW I intend to use the Moody method to calculate actual microinches of error. I have this idea I can maybe implement that in an Excel spreadsheet. My goal is simply to learn how to measure the flatness of a surface plate.

metalmagpie

 

[ballen]

You might find this helpful:

GitHub - ballen4705/Moody-Method: A portable implementation of Moody's method (Union Jack) for surface plate calibration, and accompanying paper

A portable implementation of Moody's method (Union Jack) for surface plate calibration, and accompanying paper - ballen4705/Moody-Method

github.com

it includes code for the Moody worksheets and answers questions about mirror and autocollimator locations, tilt directions, and signs.

 

[Mark Rand]

You can use a single mirror, but you need to be able to mount the autocollimator off the surface plate on a mount that is completely rigid and tied to the surface plate. Otherwise you'll get problems.

I did do three small surface plates (two cast iron and one granite) by putting the auto collimator on the side of a granite square resting on a larger granite surface table and having the plates under test also on the surface table. That meant that I could actually keep the autocollimator in the same place and rotate and move the plates to line up with it for each measuring run. An alternative to that is to clamp a stage off the side of the plate to support the autocollimator. You will need that even if using two mirrors, to mount the stationary mirror so it can reflect light from the AC to the moving mirror without obstruction.

If you do the measurements by splitting the run into two halves, then pay plenty of attention to ^^^ Bruce's paper and make sure that you are confident about the effects of changing the direction of measurement and working towards or away from the AC on the sign of the readings. If neccessary, make rough diagrams to show the effect of the mirror leaning to the left or right and approaching or leaving the autocollimator to get your head around it.

PS: Bruce:- I still haven't got around to putting those changes on Github. Life seems to have got in the way.

 

[Crimi Giuseppe]

Hai un collimatore automatico allineato con uno specchio piano, lo specchio è appoggiato su tre sfere poste su un triangolo isoscele (le sfere sulla base identificano l'asse di rotazione) la sfera sul vertice, sale e scende, appoggiandosi sulla superficie . Questa rotazione è indicata dalla deviazione dei secondi d'arco che si leggono sull'asse verticale. Conoscendo l'altezza del triangolo di base ei secondi d'arco, puoi risalire a quanti micron si è spostato in altezza il vertice del triangolo.
Per misurare la superficie, spostando lo specchio lungo i lati del piano, le diagonali e le mediane, si ottengono sezioni spezzate (in secondi d'arco, poi trasformate in micron) che danno il profilo locale del piano
Tutti questi segmenti devono coincidere al centro del piano (0,0 per costruzione)
In questo modo è possibile visualizzare la forma della superficie che caratterizza il piano.
L'autocollimatore è posizionato su un supporto esterno al piano, preferibilmente posizionato su un tavolo ottico a cuscino d'aria, perché se posizionato su un pavimento, muovendosi su di esso si flette.

 

[TGTool]

Google Translate:

You have an automatic collimator aligned with a flat mirror, the mirror is placed on three spheres placed on an isosceles triangle (the spheres on the base identify the rotation axis) the sphere on the vertex, rises and falls, resting on the surface. This rotation is indicated by the deviation of the arc seconds that are read on the vertical axis. Knowing the height of the base triangle and the seconds of arc, you can find out how many microns the vertex of the triangle has moved in height. To measure the surface, by moving the mirror along the sides of the plane, the diagonals and the medians, we obtain broken sections (in seconds of arc, then transformed into microns) which give the local profile of the plane All these segments must coincide in the center of the plane (0.0 by construction) In this way it is possible to visualize the shape of the surface that characterizes the plane. The autocollimator is positioned on a support outside the top, preferably positioned on an air cushion optical table, because if positioned on a floor, it flexes when moving on it.

 

[Crimi Giuseppe]

I apologize, but to better understand what you are saying, I had enabled the always translate to Italian button. With the help of google translate I wrote my speech, translating it into English.
When I posted it and went to check if there was anything to change, of course I had all the pages in Italian and I didn't realize that my post was in Italian.
I will try not to fall into the same trap.