Started Making Grinders for First Lenses (Lens H and Lens J). Pitched a Round Tile Onto

APF

1/16/03

First lenses should be coming at the end of this month…that’s the Dewar window and two of the small optics. On Tuesday we’ll start going through the elements and specifying tools for Jeff to build.

3/15/04

Started making grinders for first lenses (Lens H and Lens J). Pitched a round tile onto each of two pucks (one for H and one for J). Also pitched a round tile onto a flat post for the CX grinding tool for lens H. Then used wax to hold the tile-and-puck onto another post for generating the CC grinder.

For lens H CX grinder (19.2 mm target radius)…Set generator angle to 39.23 deg at first but that was too steep…ended up with 36.2 deg angle, last adjustment directions were shallower on the angle and towards operator on centering, so when we go to do the lens itself we should use these same adjustments). In the process of getting the angle right we took off too much material, so we popped that tile off and will replace it with a fresh one. We did that and the second one measured 19.1mm radius. We will also create a corrector for lens J using the same settings and technique.

3/16/04

Since there was already pitch on the post after popping off the second identical grinder for H, we put a new tile on for lens J’s corrector by just heating the tile and pressing it into the pitch. Even though the pitch wasn’t pre-heated, the tile melted it and I saw some glossy pitch ooze.

Made two CX grinders for J…one 25.94 mm and one 25.79 mm (target was 25.74mm). Last machine adjustments were: centering-towards crank operator and angle:getting shallower. Angle was 24.5 deg instead of 23.9 deg calculated.

Made another pair of grinders for H…

3/17/04

Today we glued hexagonal tiles to the grinder tools for lens H and the Dewar window. We used Devcon #14265 5-minute epoxy, dispensed with a gun onto cardboard and then mixed with tongue depressors. The gun had no instructions, but it was easy enough to figure out. A black plastic clip flipped down, and with the handle squeezed and a metal ratchet held down this allowed the gray plastic plunger to be inserted. Then the glue cylinders could be put through the plastic clip and the clip rotated back into position and snapped in place.

We spread the glue in a thick layer on the metal surface of the concave tool then put a layer of tile in place. Next we put down a layer of foil to prevent the two tools from binding to each other, and then another layer of tile, followed by the convex metal tool, slathered with glue. One cylinder of glue was enough to do four tools (two 3” diameter and two 5” diameter) with a little left over.

Before During

Done

Similarly, we glued tiles to the Dewar window grinders:

During gluing

After gluing tiles

Next we made laps for the tiny lenses (G, H and J)…we used #73 Gugolz pitch, melted with the heat lamp, and then used the grinders we made yesterday to shape the laps. Mostly we used hot water to soften the laps, and lubricated the grinder tile with Orvus and water, before rubbing the two together.

3/22/04

Last Friday I started correcting tools by working the grinders and correctors against each other with some 30mm grit. The larger tools (for F and the Dewar) were pretty straight-forward. The tool on the spindle would tend to go convex, so it was just a matter of putting the right one down. We got the radius values close to the specified lens parameters, and then calculated the sag over the whole tool to see if the deviation from exact radius amounted to very much actual material. We used a high-speed turntable and worked the piece on top by hand.

With the smaller tools (G, H and J) the same principals applied, but they were harder to hold onto and since they needed to be pitched to tool holders, it was not real convenient to flip one or the other to the spindle position. If the tool was left on the one on top, it made for a strange contact with the other grinder.

6/3/04

We’ve spent the last few months working on another project, and during that time the engineers re-designed the optics based on the melt data provided by the manufacturer. The result is that all the grinders and laps we made for the three small optics (G, H and J) are no longer right and have to be re-machined.

Not only that, but in several cases the curve generator now can’t reach the angle needed, either for the glass itself or the grinders that go along with them. We’ll probably ask the machine shop to make cast iron grinders, as they did for lens G originally. We’ve also been looking into getting smaller diamond wheels (1/2” or 5/8” OD), which would be able to create the radius with a reasonable generator angle. Even though those sizes appear in a catalog from 1980, when I called Strasbaugh they said their wheels start at 1” diameter. Gerard, who used to work at Lick, is supposed to call Dave back with more info.

6/7/04

The machine shop has put new radii on several of the lap tools. Now I need to put the pitch back on. We had shaped the pitch last time around using the grinders, and now those also need new radii in order to shape the laps. That requires setting up the curve generator, and right now I’m tied to a polishing machine so I can’t do it. In fact, some of the grinders were changed to cast iron ones, and the machine shop hasn’t delivered those yet, so on those I can’t go forward.

6/11/04

Started grinding lenses!!! Began with Lens H. First checked the radius of the tool for the steeper concave side (-14.967 mm) of the lens and found it to be 14.6mm (too steep). I ground the CC and CX tools together with the CC one on the spindle until the CX one was almost 14.9mm radius, and then began grinding the lens.

After about an hour and 20 minutes the lens had been ground out to the edge.

I measured both the lens and the tool on the spherometer, and found that the lens was shallower than the tool. It was still too steep from the spec, so I corrected the tool but stopped short of going as shallow as the spec, figuring that the lens would again be shallower than the tool. It proved to be the case when I corrected it with 30mm grit and measured both tool and lens. By stopping the tool correction steeper than spec, the grit caused the lens to be just shallow enough to hit spec.

I switched to 12mm grit and found that I could control the radius of curvature by the amount of pressure I applied to the CC lens on the CX tool. By holding it at the edges and not pushing in the center I could make it go shallower as needed.

I switched to 5mm grit and found I couldn’t control the radius as easily, but I got it to 14.95mm or so and Dave looked at how much sag difference there was between what we had and the target radius (14.967mm). He decided we were close enough (.001mm sag difference equated to a radius difference of about .008mm radius difference). This was found by comparing adjacent spherometer measurements.

The grinding motion for 30mm and the first part of 12mm grits was linear, stroking the lens across one half of the grinding tool (usually the far half). For the second half of the 12mm and most of the 5mm grinding I kept the lens near the center so that the rim would be ground more, by the faster-turning outer circumference of the grinder. I was trying to make

6/15/04

Tried to take out wedge…screwed up but was able to generate it flat with 3.33mm thickness (goal is 3mm finished). Because we were tight on excess material, Dave took over and ground CX side with 12mm grit. At one point it was dropped and a small chip appeared in the sharp edge. If we can get CX side with any excess material we’ll try to hit the CC side.

After getting rid of generator pits it was 3.27mm thick. At that point Dave corrected the tools to near the exact radius goal.

Ended up with chips that made the first lens not so good…started back-up blank with 30mm grit.

6/16/04

Started today with 12mm grit on CC side of second lens H. Got near finished and Dave beveled both sides using a CC oversized iron grinder. Then I corrected tool and got 14.97mm radius with 12mm. Then I cleaned it up with 5mm.

Checked wedge (not easy).

Thickness = 4.32mm.

Waxed it onto a post to generate off some thickness on CX side (currently flat). Will use depth gauge to measure overall height of lens and post before and during process to monitor thickness.

Goal is 3mm thick at thinnest point, and it’s currently 4.3mm, so take off 0.8mm.

Used curve generator to take off 0.8mm. The digital readout of the generator actually showed 0.7mm, but the depth gauge, which I read three different times (generating a little more between each) said it was 0.82mm. I was careful to always read the depth gauge at the same marked location on the tool. When I removed the lens and measured with the electronic linear gauge it was 3.6mm, so the digital readout of the generator was the more accurate.

Ground CX side with 30. Checked wedge and corrected with grinder NOT turning (changes fast). Switched to 12um and kept working on wedge. Getting close in thickness, so Dave took over and dialed in the wedge.

Ready for polish with: wedge=.0004” thickness=3.04mm radius 381.28mm

6/17/04

Started polishing CC side with #73 pitch on a rounded post tool. Dave cut a single cross in the pitch, off-center. We have no weights on it (just the gravity-type arm itself). We put padding on the flange of the tool and bubble-wrap on the table around the spindle to protect the lens if it comes off during the process.

First polish is 15 minutes, then we’ll look at it. We’ll be using a 0.65 f# transmission sphere to check figure and radius.

The result of the first look using the Zygo was about ¼ wave (1/2 fringe) bump, and a roll (edge was beveled). Focal length was 14.95mm by using the slide feature of the interferometer. That’s the distance between two sets of straight fringes, one at the focal length and one farther away, separated by the radius of curvature distance.

Under high-intensity lamps we could see even pitting across the diameter, so we knew the radius wasn’t changing too much.

We ran it another 50 minutes in two stretches. Six accurate radius measurements said 14.95mm. That was going from focus of Zygo, back to next straight fringe place and reading the difference. We also went from cat’s eye position (focus) and moved it back the exact target radius of curvature and then counted the number of fringes to see how far off we were from target.

Note that we could go flatter to 15.04mm….this comes from adding a .003” gap (.07mm) to the target radius of 14.967mm.

The figure was 1/5 fringe (1/10 wave) but with a slight hole in center.

We tried to improve it a little with another 15 minutes of polishing but the hole got worse. Going another 15 with a faster spindle and shorter? stroke.

Still worse…we relieved the center and will go 10 more min.

Ouch! Worse yet! Now a hole taking up half the diameter, with a sharp edge. Turns out that the lap was too thin in the center, so there wasn’t enough give to it.

6/18/04

We started with a deep hole today in lens H. We remade the lap thicker and Dave cut 1 line offcenter on it. We then ran it for 20 minutes and it improved considerably, but the radius is getting long (15.01mm). Possibly we could use it for the oil gap, which takes the original 14.967mm radius and adds .003” (.07mm) to make it 15.04mm radius. So we’re not as bad off as all that.

Anyway, Dave lengthened the stroke for another 15 minutes to try to make it go more CC.

After a couple more variations we got <1/10 wave, with R=15.01 (allows for oil gap).

Started polishing CX side. We pitched the lens onto a holder, after protecting it with Chemical 2253 stripcoat. Then we found that the radius that the Zygo reported was considerably steeper than the spherometer measurement. Dave ground the tools together to correct them, and used grease pencil on the surfaces to check the radius with the Zygo. Then he ground the lens to a close value and we’re back to polishing.

Originally we had used a lap made on a 19mm tool, but now that we’ve restarted polishing we’re using a 16mm lap.

The 16mm lap was on a flat tool, and the pitch was very thin. For these reasons it seemed to go flat in the 35 minutes that we polished it. Dave had to re-grind with 5mm grit, putting the grinding tile on a table on a towel and hand polishing the lens on top of that. He did it in small increments, and so far it’s heading the right direction.

After this we’ll go back to the 19mm lap, with a thicker pitch layer on a cut-radius tool.

Just can’t hit desired radius…Zygo readings not consistent. Grinding manually with 5mm grit the radius went flatter, flatter, flatter, steeper?

Grinding eventually caused thickness to go too thin (2.920mm…target 3.0mm)

6/21/04

Turns out that the grinder wasn’t spherical, and the depression in the center caused the lens to grind steeper when it was passed over that area. Tolerance for thickness was 100 mm, and this is just at that edge. Designer can adjust the next lens to compensate, we hope.