“Milling” grinding fused silica optic

Hi all, so I’ve got a small (~ 6x6mm) square fused silica (pretty much aka fused quartz) optical component I need to “mill” a relief into the flat bottom of, such as to leave three small square bosses or pads out on the 6x6mm periphery of the part (for stable 3-point registration of the optic on a mating flat surface). The relief can be very shallow, just a few thou deep will do fine (I.e., the three “pads” are only a few thou tall). I’ve got a fine old B&S surface grinder, that seems like a sensible enough way to go about trying this. I‘ve got “blocking“ waxes and such to hold the optic, etc, so that doesn’t worry me too much. I could design the pads such that a total of five simple linear “passes” across the piece with two different-width abrasive wheels, will “kill off” all the material in-between the three pads, not too crazy seems like. My question for the gurus here is, what sort of grinding wheels am I talking about here? Diamond, I reckon? Where do you source such wheels, like for use on my surface grinder? Any other thoughts about this? Thanks—Charley in Colorado

Let’s start by saying that I’m coming from extremely limited experience with this. I have had to grind a few different glasses in optical systems, but we’re talking qty<5 in most cases.
1. You’re not going to have an optical quality finish when you’re done.
2. Diamond will do the trick.
3. Extremely gentle passes with a generic wheels of coarse or medium grits have worked for me in the past, with flood coolant mind you. Expect some edge chipping in the <1 mm range.

Better to keep the optic plano and put the features into the mating part, but I’m guessing that ship has already sailed.

Hi jcaa, I appreciate the comments. Yep, it’s ok (in fact, good in this case) that the ground surface isn’t optical quality, “fine grind” sort of surface character is a-ok. I hear you on the chipping, I’ll just have to see if that’s trouble or not I reckon. I’m only doing a handful of these, too, for sure. I hear you on making the pads on the mating part, too, and I have been thinking about that. The mating part will be invar36 low-CTE alloy, and is certainly machinable by mortals like me, no problem (nobody’s favorite matl to machine, but easy enough). Where do you get wheels of this sort? The B&S grinder “standard” wheel is 7” dia and 1-1/4” thru hole, I believe. Thanks again—Charley

I recently purchased an "inexpensive" diamond wheel from McMaster for applying a 40-minute wedge to 4" diameter fused silica flats, using a surface grinder machine. The wheel I bought is 1" face width, 6" diameter, 180 grit, and consists of a high-density polymer body with a metal band fixed (molded on) to the wheel face. It is a wheel with discrete "dots" of diamond plating, somewhere in the .125-.15" diameter range. It was about $150. It tested well for the purpose. I made repetitive passes taking up to .005 depth of grind, with flood of plain water, almost zero chipping. If there is an edge break on your parts, that will help eliminate chips. Ref p/n 8725A82

In reviewing the original post, I realized that 6mm square parts aren't going to work well with a monster wheel;-) You probably want more of a saw-type wheel. There are many suppliers of this type of stuff. One I know of is Diacut, Inc. They have a website under that name.

For diamond wheels we just picked one off of MSC. Something like PN 03570447, though we went with something wider for our application. It's basically an aluminum disc with a thin (1/8") ring of the diamond material on the perimeter.
If you do get a diamond wheel, consider how you're going to dress it. For a relief only application, centered with an indicator (protected by some shim stock) is probably close enough. Edge chipping with a diamond wheel may well be only a couple hundred microns. Don't worry if it 's a 6" vs. 7", etc. so long as it's big enough to touch the part and no larger than whatever your grinder was designed for.

Edit,
Given your small size, you may need to do a vertical plunge into the part next to the foot locations rather than a full sweep across.

Thanks for these great thoughts on this! I‘ve rattled down to an MSC wheel, 03571429, that is 0.063“ wide, 6” dia., $178 I believe it is, haven't bought one yet but I believe I will. I can make a series of passes across the 6x6mm piece that will end up with three ~ 0.04” square pads. jcca’s note about using an indicator to center up a wheel like this, is a great one, thanks, I’ll definitely do that. A helpful guy at MSC weighed in with a comment about how grinding “glass” with a Diamond wheel can be prone to “loading up” the wheel and to watch out for that. That said, I know fused silica is WAY harder than any glass I know of; maybe that’ll make it less prone to “load up” the wheel? specfab’s experience with nearly no chipping, is encouraging. I’m starting to think I ought to be able to do this! I’ll let you all know as I get closer. Thanks again—

Since your optic is so small, you might want to consider using a small, diamond plated thick wall core drill on the order of 1.5 mm diameter. Chuck it up tight, run your spindle up as high as you can, flood the cutter. With a 180/220 grit tool your DOC will be about 12.5 microns. Use just like a conventional end mill with a feedrate of 2 minutes per 25 mm. Your DOC should be no more than the "gullet" depth between diamonds. If you can afford to wait for a custom tool get it with a chamfer on the OD. Run it around the periphery first and you will have virtually no chips. Continental Diamond can help with a tool. Hollow tool or at least center relieved. You want a diamond PLATED tool. Second best is a brazed tool. A metal bond, resin bond and vitreous bond tools will disappoint with this approach.

Hi all,
well, I’ve continued on with this, in the time available. I’ve now bought several very thin (0.025”) Diamond dicing blades, as I also want to be able to dice and size these fused silica pieces as well as “machine” them. In fact, the dicing comes first really, so. I just finished an “arbor reducer adapter” for the grinder spindle (the dicing blades have 5/8” dia holes) and have under a thou of run-out on the blade diameter (it’s 7” dia by the way). Tomorrow I’ll set up a simple little fountain pump to put a little coolant on the blade continuously, and give some test pieces a try. Charley
—by the way, also bought a 1/16”-wide MSC diamond wheel (it’s got a 1-1/4” hole, no adapter necessary) that’ll show up in a few days.

I do not do and have never done glass but I do "slice and dice" a lot of carbide.
We use a 6 inch .040 wide resin bond at feed rates of 7-12 IPM on .1875 or .250 thick carbide (single pass cut with coolant).
On ceramics we are down to 3-5 IPM.
The core is only .032 wide so use of 4 to 5 inch wheel flanges is a must.
Rather then buying flange adapters we just take old 1/4-3/8 thick used diamond wheels and cut them down on the lathe.
Bob

If you have milling capacity, and you really aren't strapped for money, you might want to check out the company "6C". They manufacture micro sized PCD tools...very impressive. I have sent them samples of some of the ceramic materials that we produce, I was amazed with what they could do.

A diamond wheel run wet and creep-feeding long travel or down travel works very well with little wear to the wheel. You will need a rolling dresser to dress, and pre-spin the wheel, don't let the wheel start the roller dresser. ..and best to indicate the fresh new wheel to .001 (or so) on the diameter so at the end of wheel life you don't waste wheel.

Small-width wheels need a double wrench tightening as trying to hand-hold the wheel can break a wheel.. Good wheel mounts have a place for a wrench at the rear/back side of the wheel mount.
Depending on your surface finish needs 120 to 320 gt might suit your needs.

I like to set my roller dresser dead center to .03 to the left of the wheel center, might set your table long travel stops if that is easy. I spin the dresser wheel with a pencil eraser end and cross to the lightning of sound. not going way off the dresser.