Here is Bill Marriott's post. I just switched from 220 SiC to 240 Al203, and there is a remarkable difference in surface finish between the two under 15X loop. The Al2O3 leaves a much more regular and smoother surface.
Cheers, Gary Lowell glowell@sgi.com
========================================================================== Hi folks:
As I've been reading this list over the last few weeks I've seen some discussions on grinding mirrors in terms of stock removal for different size grits. I run a small precision optical shop (we specialize in 'flatwork') and use the criteria that in free abrasive grinding the depth of subsurface damage is equal to twice the size of the abrasive particle, i.e., with 9u alum. oxide , the subsurface damage(i.e. pits) is 18u deep (25.4u = .001) or about .0007
To figure stock removal you subtract the subsurface damage (SSD) of the next size grit from the SSD of the previous grit, and the difference equals the necessary stock to be removed by the next grit size.
i.e.: going from 20u to 9u (Typical commercial grind) 20u SSD = 40u 9u SSD = 18u __________________ subtract 22u (almost .001)
i.e. going from 15u to 9u 15u SSD = 30u 9u SSD = 18u ___________________ subtract 12u (almost .0005 '1/2mil'
With high speed machinery, the .001 for 20u to 9u is nothing (5 min.) By hand, you can see the value of intermediate grit sizes as it cuts stock removal in half. With silicon carbide, the equation is slightly different due to the difference in hardness and 3D shape and in my experience may be 50% deeper for as given grit size Alum oxide is plate like in form whereas SiC is much more angular as well as much harder.
i.e. 80 grit SiC = 80/" =.0125 SSD =.025 180 grit Sic =180/' =.0055 SSD = 011 _________________________________________ stock removal =.014 x 1.5(50%deeper)= .021= lots of work!!
Now, as to the value of this for the average ATMer its hard to say. The mirrors I've made here at work I measured stock removal (edge and center) using calipers or micrometers which are not that expensive. Visual inspection with 40X scope confirms evenness of grind.
This formula works for polishing also in that the SSD depth of the final fine grind is the amount of stock to be polished off. i.e.: 9u final grind = 18u SSD, so in theory, when .0007" has been polished off, there should be no pitting
with 5u SSD = 10u, = .0004 or almost 50% less than with a 9u grind. On the high speed polishers I've used this doesn't make much difference as they polish out in 2-3 hours anyway, and the risk of sticking and scratching with 5u may make 9u the best finish for hand polishing also.
Using this criteria, polished surfaces exhibit no pitting visible at 100x.
Anyway, that's my 2cents on the subject.
Bill Marriott Forest Knolls, Ca. btk@ix.netcom.com ========================================================================