Main Menu

ATM
Automated Obs
WebCams
Sports Cars
Music
Froggy
FAQS
Gallery
Links
About Me
SpaceCam
Dobsonian
 

Section

Mirror Grinding
Mirror Figuring
Mirror testing
Mirror Cell
Focuser
Filters
De Rotation
Counter Weight
Drive

 

 

 

 

 

  

The most active and innovative astrophotography forum in the world! 

Join QCUIAG today.

 

 
Mirror Grinding

Up Next

The plan was for a 20" reflector mainly for use with a CCD camera.  I choose to start with a 1" thick plate glass blank, mainly on the grounds of cost.  For the sake of the CCD chip and to keep the size of the scope down I went for a focal ratio of 4.   The sagitta depth required it a horrific 1/3 inch.  Fearing I might spend the rest of my life hogging out I took a look though the ATM list archives and found a few threads and sites detailing the use of diamond cutters for rough grinding.  Ricardo Dunna's and Peter Smith's  sites describe what is possible but the methods require a mirror making machine or a drill press. I have neither. The TRO site is a good description of the low-tech approach of pivoting the cutter at the centre of curvature.  I adopted a similar method but choose to suspend the cutter for the roof of my garage.

This is a schematic of the curve generator.

{short description of image}

So starting at the top.  Here is a very poor picture of the top pivot point high in the roof of the garage.  The pivot is a ball and socket affair while the top of the arm is 1 1/2 " square wood

{short description of image}

Control over the depth of cut is by bolt arrowed.  I welded some washers into the top of this joint to help stiffen things up.

{short description of image}

. 

 

This is the cutter in action.  The point of contact on the glass is directly under the pivot arm.  The brick acts as counter weight (its actually bolted to the horizontal rod, the tape being insurance). 

{short description of image}


 

Another picture of the disc.   The disc is from a tile cutter and cost £10.  The spindle is a bolt and 2 nyloc nuts keep the disc in place.  The rubber backing of a drill sander attachment is used behind the disc to provide support at the edge to the disc and acts as a clutch.  The spindle bolt was pressed into a bearing to provide more support.

{short description of image}

{short description of image}

After ~12 hours worth of cutting this is the result.  The marks on the surface aren't very deep and should soon come out.

{short description of image}

Pros: I am really happy with the results.  The speed with which the cutter removes glass is amazing.  The bulk of the glass was removed in 2 cuts of about 3mm a time. This meant the side of the cutter was doing the work rather than the edge.  Still the diamond disc held up really well and looks good for another 3 or 4 mirrors, even if I'm not.

Cons:  Nothing serious.  Water tends to get everywhere.  Bit of playing around with glass placement required to keep ground dish concentric with blank.

{short description of image}

The IMPORTANT bit.

If you are thinking of doing something similar remember.

i. Use an RCD on the power supply.

ii. Use lots of water to disperse ground glass and prevent dust.

iii. Safety specs and dust masks are a must 

 

 

leaf

The Poor Meadow Dyke Galleries are now open.  

THE GALLERIES

 

 

See the full range of Atik cooled CCD cameras for astronomy on the Atik Website

 


Copyright 2006 Steve Chambers. All Rights  Steve@pmdo.com

  Up Next