Cutting dichroic mirrors with hole saw - help needed!

[viktor j nilsson]

In an earlier thread, Hobbys46 reported success in cutting dichroic mirrors with a diamond hole saw:
viewtopic.php?p=72470&sid=76f1142f5f6aa ... 01a#p72470

I have a rather large rectangular Chroma Chroma T400LP beamsplitter (25.5x36mm) that I wish to fit into a Zeiss Fluoro condenser filter cube (accepts 22mm round filters).

I have a 25mm diamond hole saw, which has an inner diameter of quite exactly 22mm. I believe it would work.

What are my chances that I can make a nice, clean cut without destroying this valuable and delicate filter?

What steps should I take? What I believe will work is the following:

- Drill press (have to use a friends')
- Very slow speed
- Submerged in water
- Resting on top of another piece of glass (I am thinking a microscope slide).

What should I do from preventing the filter from rotating?

Any feedback appreciated!

[Hobbyst46]

Viktor,
I apologize for saying so - but I there is no guarantee in glass works !
I drilled many holes in many glass squares (simple glass, absorption filters, reflective ND filters and interference filters). And glazed ceramic plates. Only failed when the drill bit was worn.
The thickest one I ever drilled was 3mm thick.
As a base for the filter during drilling, I used a flat 3-5mm thick glass plate, not a slide. This base should be considerably wider than the filter, so it rests securely (by its weight) within the water tray. A slide would be too thin IMO.
I did it like this: submerge the base and filter in the water tray on the drill press stage, underneath the diamond bit. Hold the filters with fingers - just to have it stay stable. Turn on the drill and slowly lower the bit with the other hand. Apply only slight pressure. If the pressure is light, the drill bit will slowly cut into the glass, without rotation. Rotation signifies overly high pressure. Think "sawing" the glass, not "pushing" through it.
Wear eye protection - goggles, at least.
I did not don any gloves, because I am concerned about glove getting caught in the machine.
I suggest you first practice on 1-2 pieces of plain glass.
Do it carefully, and at your own risk. I take no responsibility.
Good luck!

[viktor j nilsson]

Thank you! I now have a much clearer image in my mind of what I need to do. I'm in no hurry, but I am starting to feel confident that this is what I'll end up doing. I'll definately practice on some scrap camera filters first.

I recently started a thread over at photomacrography.net where I chronicle my efforts to turn my Wild M20 into an epi-fluorescence microscope using a Zeiss fluoro condenser. Please do check it out if you are interested!
Link here: https://www.photomacrography.net/forum/ ... hp?t=41817

best,
VIktor

[Hobbyst46]

The function of water here is the same as of cutting oil on a lathe. No need for a deep water layer - just to supply a coating of water to the drilled filter surface.

[Hobbyst46]

I recently started a thread over at photomacrography.net where I chronicle my efforts to turn my Wild M20 into an epi-fluorescence microscope using a Zeiss fluoro condenser. Please do check it out if you are interested! Link here: https://www.photomacrography.net/forum/ ... hp?t=41817

Will look, sure, thanks !

[Hobbyst46]

Viktor,
So I quickly surfed through your post on the photomacrography.net. Interesting.
As Pau comments, your set of interference filters is a steal. Such filters, each one and new, are expensive.
I like it that you got both long pass and shortpass filters. Very useful !
The fried filters must have been exposed to the HBO for hours - weird, since they are on the "other" side of the cube. Who knows. LEDs will get rid of frying and of heat filters.
At the time of the old Zeiss scopes, filters were almost exclusively absorption filters, and were made by Schott, Germany (Schott and Zeiss were in close relations anyway). They are named BGxx, UGxx, GGxx, such like. Of course the transmittance spectra are less attractive... broad lines, shallow transition from 0 to 100% etc. Possibly, the dark blue filter is one of those.
At least some modern cubes accept rectangular or square beam splitters (dichroics), not round. But that Zeiss is old...
About LEDs: one could use a white LED with filters, but I suspect that the intensity over the band will be low, relative to a dedicated light LED. And the green dip in the spectrum is not good.
A 10W monochromatic LED will likely yield bright excitation. The commercial CoolLEDs are perhaps 15W, not much more.

[viktor j nilsson]

Viktor,
So I quickly surfed through your post on the photomacrography.net. Interesting.
As Pau comments, your set of interference filters is a steal. Such filters, each one and new, are expensive.
I like it that you got both long pass and shortpass filters. Very useful !
The fried filters must have been exposed to the HBO for hours - weird, since they are on the "other" side of the cube. Who knows. LEDs will get rid of frying and of heat filters.
At the time of the old Zeiss scopes, filters were almost exclusively absorption filters, and were made by Schott, Germany (Schott and Zeiss were in close relations anyway). They are named BGxx, UGxx, GGxx, such like. Of course the transmittance spectra are less attractive... broad lines, shallow transition from 0 to 100% etc. Possibly, the dark blue filter is one of those.
At least some modern cubes accept rectangular or square beam splitters (dichroics), not round. But that Zeiss is old...
About LEDs: one could use a white LED with filters, but I suspect that the intensity over the band will be low, relative to a dedicated light LED. And the green dip in the spectrum is not good.
A 10W monochromatic LED will likely yield bright excitation. The commercial CoolLEDs are perhaps 15W, not much more.

Thanks for your comments. I am not quite sure what happened to the filters. But it doesn't really matter - I knew the filters were gonna be bad, and I was mostly keen to get the filter cube as a convenient holder for my own filters. But it does seems like the 585nm LP dichroic mirror might turn out to be useful (although it does have a scratch).

Regarding LEDs, I've bought a couple of "monochromatic" LEDs to play with: a 395nm UV, 430nm UV, Royalblue and Green from this seller: https://www.ebay.com/itm/173380992656 and a generic chinese 365nm UV LED. Getting a good Nichia UV LED would be a natural upgrade if everything else is working well. Still need to figure out how to construct the collector lens + LED holder.

[viktor j nilsson]

One more question to you, Hobbyst46. How did you dry and clean the dichroic mirror after cutting? If air drying, am sure that there will be at least some streaking from the water? And in general, what is a safe method to clean DMs?

[Hobbyst46]

One more question to you, Hobbyst46. How did you dry and clean the dichroic mirror after cutting? If air drying, am sure that there will be at least some streaking from the water? And in general, what is a safe method to clean DMs?

I just hold it by the rims (wearing gloves), rinse it with distilled water, or first with an ordinary soap solution, then distilled water. If the surface is oily for any reason, rinse it with IPA, then with soap water, then DW. Wipe it clean lightly, softly, with a KimWipe or lens tissue, or just leave it to dry out. Distilled water does not leave stains.
Although quality dichroic mirrors (say, from Omega, Oriel) live and function for several years at least, after (say) 10 years, the coating becomes hazy, even if the mirror lies unused in a drawer.

[viktor j nilsson]

Thanks again!

[MicroBob]

Hi Viktor,
instead of drilling with a hole saw you could remove the corners by grinding from the outside. In my view this would introduce less strain in the glass.
I have bought the same set of dichroic filters for my Zeiss RS III epic fluo condenser.

Bob

[viktor j nilsson]

Hi Viktor,
instead of drilling with a hole saw you could remove the corners by grinding from the outside. In my view this would introduce less strain in the glass.
I have bought the same set of dichroic filters for my Zeiss RS III epic fluo condenser.

Bob

Yes, I am grateful that you pointed these out! It set me off on this project.

Grinding is a possibility, but getting it down from 25.5x36mm to 22mm diameter means that a lot of material needs to be removed. I think that I prefer the hole saw.

[Hobbyst46]

There is of course another option - cut off the longer side, that is, shorten the rectangle by about 13mm (from 36 to 23). And grind the rest. Success depends on the experience with glass cutting. In principle, one draws a straight line, with a glass cutter or with the tip of a triangular needle file, from end to end. Then moisten the line slightly with water (real professionals do it with saliva , grip each end and tear apart, in a downward bending motion. Or alternatively, lay the longer side flat on a table, so the line coincides with the rim of the table plate, press on the long part and pull downwards the strip, to tear it apart. Easier done than described... can work if the filter is not too thick. 1-2.5mm is OK. 3 mm - maybe. Thicker than that - doubtful.
Professionals finish the cut by means of a hot-end glass rod. They melt the end of the rod in the gas flame, attach it to the strip and pull. Seen it, never done it.
Whatever the method of cutting and grinding - use eye protection.

[MicroBob]

Whith glass cutting there is an interesting effect: The impact of the scoring settles over time. So scoring and breaking should take place without a pause inbetween.

[viktor j nilsson]

I've never been able to make good, clean cuts with the glass cutter. I didn't know about the settling time - maybe I've been working too slow. In any case, I feel more comfortable with the hole saw approach. Grinding could still be an option though, especially if I end up needing to reduce a 22mm diameter filter down to 18mm. I have a double-sided diamond sharpening stone (coarse/fine) which might work well for this.

[microb]

I guess I don't understand, the beam splitter mirror is round? It's not elliptical. Unitron's had small mirrors but they are elongated, because of the 45 degree slant. But this older Zeiss it's round?

[viktor j nilsson]

Yes, the old Zeiss fluoro condenser uses 22mm round dichroic mirrors, and 18mm round excitation and emission filters.