Possible better UV filter stack?

[Stefano]

I am trying to find a better UV filter stack, that has OD 5 blocking while maintaining 70%+ of peak (internal) transmission and blocking violet above 400 nm. Is this possible?

 

The solution I (hopefully) found is to use thin Hoya U-340, taking care of the green leak of this filter.

 

So, try to calculate the transmissionof this stack:

 

Hoya U-340 (1 mm)

S8612 (1 mm)

BG3 (2 mm)

 

The BG3 would suppress any green leak in the 550 nm region, while barely affecting the UV transmission. It would also help a bit with the IR leakage.

If it doesn't reach OD5, we can always play a little with the thicknesses of the components.

 

I am very interested with the BG3 filter used to improve UV-only filter stacks.

 

Cadmium, I'm waiting for you.

[ulf]

Maybe I can help.

 

The stack you suggest suffer from problems in NIR.

Normally the main difficulty with designing filter stacks with UV-pass filter glass like UG11, UG1, H-340... is to surpress the secondary peak around 700nm, not to suppress any leakage in visual wavelengths.

Most if not all plants with chlorophyll have a very high reflectivity there.

 

Your suggested stack is slightly better than OD 3 at 710nm.

[Stefano]

So, I should be fine with the same stack, but with both Hoya U-340 and S8612 at 1.5 mm thickness. What if I keep the U-340 at 1 mm and make the S8612 2 mm thick?

[Stefano]

A 2 mm BG3 adds ~2 OD points at 690 nm and 1 OD point at 700 nm. An Hoya U-360 can move the IR leak towards longer wavelengths, making it much easier to block.

[Stefano]

Just out of curiosity, try this recipe:

Hoya U 340 (1 mm)

Hoya U 360 (1 mm)

S8612 (1 mm)

BG3 (1 mm)

 

I don't know if it will work, but I am sure that a 70% OD 5 filter under 5 mm can be made.

 

If it doesn't work, increase slightly the S8612 and/or the U-360 (like 1.25 mm).

[dabateman]

I actually think the BaaderU fits your criteria.

Or the Astrodon UV filter, but I have heard it leaks and would need to be stacked with S8612.

https://www.astroshop.eu/planetary-comet-filters/astrodon-filters-uv-venus-filter-1-25-/p,14692

[Stefano]

I actually think the BaaderU fits your criteria.

Or the Astrodon UV filter, but I have heard it leaks and would need to be stacked with S8612.

https://www.astroshop.eu/planetary-comet-filters/astrodon-filters-uv-venus-filter-1-25-/p,14692

I know that the baader U has an excellent UV transmission, but I want to make it with ionic colored glass.

[ulf]

Just out of curiosity, try this recipe:

Hoya U 340 (1 mm)

Hoya U 360 (1 mm)

S8612 (1 mm)

BG3 (1 mm)

 

I don't know if it will work, but I am sure that a 70% OD 5 filter under 5 mm can be made.

 

A 4-component stack will be expensive and difficult to make.

Each glass disc has to be ground and polished before assembly.

At each glue layer there is risks for air-bubbles and dust-inclusions.

Each layer has likely to be glued and cured separately.

 

"a 70% OD 5 filter under 5 mm can be made."

No, I don't think that is possible without dichroic technology.

At each air to glass transition you loose ca 4.5% in surface-reflections.

The transmission values you can read in the data-sheets are for internal transmission excluding the surface-losses.

 

The Baader U is slightly above 70%, but has an OD of slightly better than OD 4, not OD5.

 

[Stefano]

This should really work.

 

Hoya U-340 (1.5 mm)

S8612 (1.5 mm)

BG3 (2 mm)

 

It should have (maybe?) OD 5 at least below 700 nm, and have around 70% peak transmission.

 

I am referring to a glued stack, so you can reduce losses between layers.

 

If it blocks IR more than OD 5, I can reduce the thickness of the BG3.

 

I have almost surely (needs checking) OD 5 if I use 1.75 mm for both the S8612 and the Hoya U-340. If I put one of them at 1.5 mm and leave the other at 1.75 mm, I could nail the OD 5 blocking while still maintaining a 70%+ transmission.

 

U-340 1 mm + S8612 1.75 mm would have less than 70% peak transmission.

 

Replacing the U-340 with a U-360 would require less S8612 to block the IR peak, since it is moved towards longer wavelengths.

 

The idea is that I can make a stack blocked up to OD 4 before 700 nm, and the BG3 would bring that down to OD5.

 

This is another stack:

 

Hoya U-360 (1 mm)

S8612 (1.75 mm)

BG3 (2 mm)

 

Or

 

Hoya U-360 (1.25 mm)

S8612 (1.5 mm)

BG3 (2 mm)

 

A 1 mm thick U-360 passes a bit of violet, which I don't like.

[ulf]

This should really work.

 

Hoya U-340 (1.5 mm)

S8612 (1.5 mm)

BG3 (2 mm)

 

It should have (maybe?) OD 5 at least below 700 nm, and have around 70% peak transmission.

 

I am referring to a glued stack, so you can reduce losses between layers.

 

If it blocks IR more than OD 5, I can reduce the thickness of the BG3.

 

I have almost surely (needs checking) OD 5 if I use 1.75 mm for both the S8612 and the Hoya U-340. If I put one of them at 1.5 mm and leave the other at 1.75 mm, I could nail the OD 5 blocking while still maintaining a 70%+ transmission.

 

Transmission including the losses from the two air to glass surfaces maximum ca 57%. Peak at ca 355nm

Almost OD5 at 720nm.

 

An U-360, 1.5mm + S8612, 1.75mm is a few percent better in peak transmission, with a peak at a better 370nm, where the cameras are more sensitive

The suppression is also a tiny bit better in NIR. This filter is cheaper to produce and have better performance.

 

Chasing a few percent improvement in transmission is often quite meaningless as the difference in exposure time is almost not noticeable.

I'll stop here and will not do any more calculations for you.

Earlier I have been doing the same type of search as this without much success.

[Stefano]

Chasing a few percent improvement in transmission is often quite meaningless as the difference in exposure time is almost not noticeable.

Yes, a 10% difference is almost insignificant, but I would like to have the best filter possible, with the highest transmission.

[ulf]

How important is it for you to decrease the exposure time from 10s to 9s or 5s to 4.5s?

What is that effort worth?

[Stefano]

How important is it for you to decrease the exposure time from 10s to 9s or 5s to 4.5s?

What is that effort worth?

Not much. But it's kind of the same reason why we usually use S8612 instead of BG39. Same IR suppression, and a 20% difference in UV. That's less than a stop. But having more sensitivity isn't a bad thing I think.

[Stefano]

So, I think that the best OD 5 ionic filter possible is a Hoya U-360 (1.5 mm) + S8612 (1.75 mm). One good OD 5 filter, U-340-based, would be a U-340 (2 mm) + S8612 (1.75 mm). You probably can not do better than that.

[Cadmium]

I would skip the idea of adding BG3 to suppress the 500nm range visual U-340 1mm leak.

The 1+1+2mm will only suppress slightly below OD3, and that is borderline and you should go for at least OD4, no more than OD5 needed.

Using U-340 1.5mm will solve the 500nm range visual leak.

 

Just use U-340 + S8612. Just decide how much suppression you want.

 

U-340 2mm + S8612 1.75mm = UV 60% + OD5

U-340 2mm + S8612 1.5mm = UV 62% + OD4.5

U-340 1.5mm + S8612 1.5mm = UV 65% + OD4

U-340 1.5mm + S8612 2mm = UV 60% + OD5

 

My favorite stack is U-360 2mm + S8612 1.5 to 2mm (1.75mm = OD5).

Such a stack crosses the 400nm line just slightly above OD3, UV 60% + OD5 (1.75mm S8612).

White balance is the same as using Baader U.

[Stefano]

So, the best OD 5 stacks are:

 

Hoya U-360 (1.5 mm) + S8612 (1.75 mm). Total: 3.25 mm.

 

Hoya U-360 (2 mm) + S8612 (1.75 mm). Total: 3.75 mm.

 

Hoya U-340 (2 mm) + S8612 (1.75 mm). Total: 3.75 mm.

 

Hoya U-340 (1.5 mm) + S8612 (2 mm). Total: 3.5 mm.

 

All at around 60% peak transmission.

Does glued make a difference?

 

Would you sell them as a product? Your 3 pre-made filters (LUV U, La La U, Moon U) should use a different stack formula.

 

Also, I noticed that this section https://www.google.com/url?sa=t&source=web&rct=j&url=https://www.ultravioletphotography.com/content/index.php/topic/1313-filter-transmission-charts/&ved=2ahUKEwjBnOrBsrbnAhWFxosKHQ_1CGUQFjAAegQIBBAB&usg=AOvVaw2GOTdG4Fq59Ja6iWj9a8_v lacks graphs for U-360 stacks. There are a lot with U-340.

[Cadmium]

Glued should make some difference, however two filters screwed togehter work fine, and can be more versatile,

for example, using the same S8612 2mm thick filter for UV only shots (when stacked with U-360, UG1, UG11, U-340), and also sued for UV+Blue+Green shots (AKA: Bee Vision)

when stacked with UG5/U-330, three filters do two things, instead of 4 filters, etc..

The S8612 2mm can be used for many stacks: UG1, UG11, U-360, U-330, UG5, UG2A, BG3, BG25, B-410...

[ulf]

If you want the flexibility for experimenting separate filters screwed together is the best way to go.

Then a single S8612, 2mm is then the ideal BG-type of filter as it has good UV-transmission and enough IR attenuation to work well with most other filters, just as in Cadmium's list above.

 

Two filters screwed together have two more air to glass passages that cause ca 9% more losses.

I think that issn't very important compared to the gained flexibility.

 

One important thing to know about is that many of the UG and BG glass types, when uncoated, are sensitive to humidity.

After some time the polished surfaces can start to matten and eventually become unusable.

That can be avoided by polishing and cleaning the filters when seen needed.

 

They need some attention and care. To do that is not difficult.

[Stefano]

As a last (impractical) solution you can make a stack with Hoya U-340 (1 mm), thin S8612 (to bring the OD down to less than 4, like 3.5), Hoya B-410 (maybe 1.5 or 2 mm thick, to help with the NIR leak), and 1 mm BG3 (to help with the green leak). It could work. This is NOT a good filter regarding price, difficulty to make etc, but it may do what I want (OD 5 without violet >400 nm and 70% peak internal trasmission). If you want to play a little with the transmissions, do that. This is just a very theoretical concept, not meant to be made with actual glass.

 

If someone wants to calculate it for me...

 

Hoya U-340 (1 mm)

S8612 (1 mm or 1.25 mm if needed)

Hoya B-410 (2 mm)

BG3 (1 mm)

[ulf]

As a last (impractical) solution you can make a stack with Hoya U-340 (1 mm), thin S8612 (to bring the OD down to less than 4, like 3.5), Hoya B-410 (maybe 1.5 or 2 mm thick, to help with the NIR leak), and 1 mm BG3 (to help with the green leak). It could work. This is NOT a good filter regarding price, difficulty to make etc, but it may do what I want (OD 5 without violet >400 nm and 70% peak internal trasmission). If you want to play a little with the transmissions, do that. This is just a very theoretical concept, not meant to be made with actual glass.

 

If someone wants to calculate it for me...

 

Hoya U-340 (1 mm)

S8612 (1.25 mm)

Hoya B-410 (2 mm)

BG3 (1 mm)

 

The stack above gives slightly less than OD4 at 720nm and an UV peak of 68%.

 

I think it is time to give up this quest until new glass types appears.

 

The "without violet >400 nm" is something we have debated much here before.

I am of the opinion that passing little light just beyond 400 nm isn't that bad if the transmission of the stack just below 400nm is much higher.

[dabateman]

Just buy an Astrodon Uvenus. Yes I know its dichroic and $465 US for a 50mm unmounted filter.

But researching the latest comments from Astrophotography forum, it seems to be the best with over 90% transmission from 325nm to 385nm. The newest ones post Astrodon purchase by some company are reported to be excellent. The best blocked.

 

If I had the cash, I would buy one to check out. The 1.25" version is $210 US. But still too much for me.

[Stefano]

I think it is time to give up this quest until new glass types appears.

Yes, that was my last, desperate attempt. We simply don't have the right glasses to do this. A glass such as BG3 but with the IR cut-on shifted towards longer wavelengths by as little as 20-30 nm would be "game-changing". In that region, 20 nm more to the right would do A LOT. Maybe Schott will make an improved version of the S8612, allowing for 10% more UV. Until then, we really don't have other choises other than a 65% stack like Hoya U-360 (1.5 mm) + S8612 (2 mm) (we should check it) or other stacks in the 60% region like Hoya U-340 (2 mm) + S8612 (1.75 mm). This is the best we have.

[Stefano]

The "without violet >400 nm" is something we have debated much here before.

I am of the opinion that passing little light just beyond 400 nm isn't that bad if the transmission of the stack just below 400nm is much higher.

A bit of violet (Moon U as an example) changes your false lavenders to a little bluer shade. I like to consider UV as light below 400 nm, and I classify 405 nm (blu-ray GaN lasers) as violet. Then I like the U-340 filters because they cut a little deeper, making objects appear slightly differently (I made a topic recently about this).

[Stefano]

Just buy an Astrodon Uvenus. Yes I know its dichroic and $465 US for a 50mm unmounted filter.

That's very expensive! Even though the quality should be very high based on what you said.

[Cadmium]

Here is the graph, don't get dizzy.

Works rather well, except that is a lot of glass (5.25mm thick). Also, the BG3 1mm is maybe not so easy to find.

If you glue, that will be a big job, and if you mess up any of the glued layers then you have to un-glue all of it and start over in a few days or more.

You can screw them all together instead of gluing, but with all those layers it will reduce the amplitude of your UV, so you will loose much or more of what you gained.

If you want to use U-340, I suggest using 1.5mm, then adjust your S8612 according to the OD you demand, which should be OD5 or OD4 at least. OD3 is where you WILL see a leak.

If you want to use UG11 then you can use 1mm or 1.2mm.

My preference is U-360 2mm + S8612 1.75 (for perfect OD5), but 2mm will be more versatile for use in other stacks.

If I were going to pick out 3 filters, I would get these:

U-360 2mm

U-330 1.5mm

S8612 2mm

With those you can do reflected UV and also UV+Blue+Green (AKA: Bee Vision).

UV+Blue+Green is rather attractive. It ads a whole other color to the UV pattern mix.

Reflected UV = Yellow, Violet, Black and White.

UV+Blue+Green = Yellow, Green, Violet, Black and White.

(yellow, violet, green being dependent of certain materials and/or type of flowers present in the photo)

There are a lot of other filters you and stack with S8612 2mm, any of the U filters, and a lot more.

If you wanted to do UV+Blue+Green+Red (AKA: Bird Vision, or would that be Butterfly Vision? Where is my wall chart?), then you could use UG5 or U-330 1mm,

that will show all the same UV patterns, looks essentially the same as the U-330 1.5mm stack, except red flowers will look red instead of black.