Possible better UV filter stack?

[Stefano]

I downloaded the Schott 2017 filter calculator, and played with it for a bit. I took the data for the U-360 from there: https://www.google.com/url?sa=t&source=web&rct=j&url=https://www.sydor.com/wp-content/uploads/2019/06/Hoya-U-360-Ultraviolet-Transmitting-Visible-Absorbing.pdf&ved=2ahUKEwj4i6yklbnnAhVlxIsKHZVaD9MQFjAOegQIAhAB&usg=AOvVaw1lh8_oJUkUB88s3SdoXa_E, and I didn't get a perfect OD 5 with Hoya U-360 (2 mm) + S8612 (1.75 mm) but I got OD 5.22 at 710 nm. In that point I have 5.72E-2 for the U-360 and 1.05E-4 for the S8612, bringing the total transmission to 5.99E-6, or OD 5.22.

 

I couldn't find better U-340 stacks than yours, and I found an almost perfect one with a U-360, if you don't dislike violet:

 

Hoya U-360 (1 mm)

S8612 (1.5 mm)

B-410 (1.75 mm)

[Cadmium]

You don't need to be using B-410. You can if you want, but I would just use two filters.

 

You need to enter the Hoya Ti data into the shot program, it runs off Ti data.

What data numbers do you have at 300nm, 400nm, 700nm, 710nm, 720nm, 730nm, 740nm, 750nm, 800nm?

You have the 0.88 Reflection factor P and 2.50 Reference thickness in mm entered?

 

U-360 2mm + S8612 1.75mm is a perfect OD5 (1E-05) using Diabatic Internal Transmission graph.

 

If you use U-360 1mm, then use S8612 2mm with that, about 405nm at OD3, about 67% at 365nm, 1/3 stop faster than a Baader U

(U-360 2mm + S8612 2mm = 1/3 Stop slower than Baader U).

You will loose transmission amplitude the more surfaces you add. Unless you glue it, and then it will get more expensive, you might as well buy a Baader U then.

If you glue it then it is not versatile, and you need to duplicate all the same filters for other stacks...

 

I priced an Asahi U filter today, 49mm circle to fit 52mm ring, over $520.00 with the shipping and 'bank charge' (basically a charge for not buying $1,000.00 or more).

Of course a Baader U is less expensive that the Asahi.

[Stefano]

I entered the Ti data (bottom line), and I put the thickness at 2.5 mm. At the moment I can not tell you the data I have at those specific wavelengths, but you can guess it.

 

At 710 nm, I should have 2 8% (0.028), and at 2 mm it becomes 0.028^(2/2.5) = 0.0572.

[Cadmium]

For the Hoya U-360 filter, 400nm should be 0.002, 700nm should be 0.0022, 710nm should be 0.028 (as you said).

Reference thickness should be 2.5mm.

Reflection factor should be 0.88.

 

In the Data Input, at the bottom, what is your Peff (effective reflection factor of the combination)?

 

Try this:

Load only your U-360 (2mm, lets say) into Ti Diabatic, then do a screen shot of your graph, and post it here.

Let's see what it looks like. Sounds good? Let's start with a simple single U-360 Ti Diabatic plot.

[ulf]

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).

 

How much attenuation would you like to have in the band 400-405nm?

I don't think you want OD5 there too as for IR.

[Stefano]

For the Hoya U-360 filter, 400nm should be 0.002, 700nm should be 0.0022, 710nm should be 0.028 (as you said).

Reference thickness should be 2.5mm.

Reflection factor should be 0.88.

 

In the Data Input, at the bottom, what is your Peff (effective reflection factor of the combination)?

 

Try this:

Load only your U-360 (2mm, lets say) into Ti Diabatic, then do a screen shot of your graph, and post it here.

Let's see what it looks like. Sounds good? Let's start with a simple single U-360 Ti Diabatic plot.

As soon as I can I will try this.

[Stefano]

 

 

How much attenuation would you like to have in the band 400-405nm?

I don't think you want OD5 there too as for IR.

At least OD 2, maybe OD 3. Any more than that and I would need to make the U-360 too thick.

[ulf]

At least OD 2, maybe OD 3. Any more than that and I would need to make the U-360 too thick.

 

Then both Baader U and the U-360/S8612 2mm-stacks are borderline cases.

They both transmit ca 1% at 400nm.

 

In my experience about photo, it is normally enough with an intensity difference of 2-3 stops between light sources, for a stronger light source, to completely dominate a weaker one. (greyscale)

An attenuation of OD2 is more than 6.5 stops difference. That is quite an overkill margin.

 

This cannot be directly compared to why up to OD5 is needed to cancel IR-contamination.

[Stefano]

I also found this source for the transmission of the U-360: https://www.google.com/url?sa=t&source=web&rct=j&url=https://www.edmundoptics.com/p/u-360-uv-12mm-dia-colored-glass-bandpass-filter/6568/&ved=2ahUKEwin-aeJtLrnAhVEIMUKHeCMDLsQFjACegQIBxAB&usg=AOvVaw0JZYqF9PpW6A_f5T2Vxvw-&cshid=1580905680938

 

The data is slightly different. Judging from the graph the data refers to the total transmission, including reflection losses.

 

Which source is the most reliable?

[Stefano]

My data for Hoya U-360 (internal transmission), 2.5 mm:

 

300 nm: 0.0031

400 nm: 0.002

700 nm: 0.0022

710 nm: 0.028

720 nm: 0.067

730 nm: 0.093

740 nm: 0.112

750 nm: 0.111

800 nm: 0.016

 

I now noticed that I didn't have the reflection factor in it (now added).

 

I put 0.0000001 when the transmission is not defined.

 

Hoya U-360 2 mm alone (internal transmission):

 

[Cadmium]

I think the Peff is needed for the T graphs...

Here is the same graph I made, looks the same as yours to me, right?

 

[Cadmium]

I also found this source for the transmission of the U-360: https://www.google.c...d=1580905680938

 

The data is slightly different. Judging from the graph the data refers to the total transmission, including reflection losses.

 

Which source is the most reliable?

 

I have encountered several U-360 data sheets, not all are the same, I don't know which to use, just try to find the one that is most recent direct from HOYA.

When you do, let me know so I can check against the data I have entered.

[Cadmium]

When it comes to the 400nm OD, it seems to me there are two things to consider, the OOB OD and the % at 400nm.

Some filters might start to cross the 400nm line at as much higher % than others, that is important.

Lets say we have two filters,

1) crosses the 400nm line at 30%, is OD3 at 405nm.

2) crosses the 400nm line at 5%, is OD3 at 405nm.

Obviously #2 will not have as much OOB.

So it is just as important, if not more, what % the transmission crosses the 400nm line.

[Stefano]

When you do, let me know so I can check against the data I have entered.

I didn't understand what I have to do.

[Stefano]

I think the Peff is needed for the T graphs...

Here is the same graph I made, looks the same as yours to me, right?

 

They too look the same for me.

[Stefano]

Hoya U-360 2 mm + S8612 1.75 mm.

 

First column: wavelength (nm), second column: S8612, third column: Hoya U-360, fourth column: total internal transmission. Hopefully I didn't mess up.

[dabateman]

Back to main filter question. I know you don't want a dichroic filter. But why?

 

Also did you consider the 330WB80 improved filter? Its cheap but small at 24.5mm diameter. But has 70% transmission and works well with m43rds cameras.

[Stefano]

Back to main filter question. I know you don't want a dichroic filter. But why?

 

Also did you consider the 330WB80 improved filter? Its cheap but small at 24.5mm diameter. But has 70% transmission and works well with m43rds cameras.

Dichroic filters aren’t bad, but they may be (not necessairly) angle-dependent and may not block IR enough. This doesn’t mean that a Baader U is not a good filter, but it is less “stable” regarding angle etc. Ionic filters are customizable, can be made to have extreme IR blocking if you need, and are not very angle-dependent (at high angles they are thicker, and this should change the transmission a bit). The last reason is purely aesthetic: I like a filter that absorbs OOB light and appears black, like you would intuitively expect.

 

The 330WB80 is a filter I have to try, and I will try it in the future, adding it to my collection.

[Cadmium]

I didn't understand what I have to do.

When you find the most up to date and official data sheet from Hoya for U-360. Best to find it from them on their site.

[Stefano]

Did you enter the same data I entered?

 

Auto Answer: Yes, it seems so.

[Cadmium]

Changing the Peff should change the T diabatic combination plots

I think the Peff (with a glued stack) would be the sum of the outside reflective factor, which I think would be back-outside RF/2 + front-outside glass RF/2? Ask Andi.

[Cadmium]

Hoya U-360 2 mm + S8612 1.75 mm.

 

First column: wavelength (nm), second column: S8612, third column: Hoya U-360, fourth column: total internal transmission. Hopefully I didn't mess up.

 

That is what I have except for 690nm I don't have data there. The rest is the same.

[Stefano]

That is what I have except for 690nm I don't have data there. The rest is the same.

The maximum leak is at 710 nm, with 5.99E-6, which is OD 5.22. Where is the perfect OD 5?

[Cadmium]

There are several 330WB80 filter versions listed on eBay.

The reason I might not recommend one of the 330WB80 filters to someone with less than a UV-Nikkor lens is because those filters cut off the top part of the UVA, so not as much violet UV below 400nm.

So the color palette will be more monochrome, and the exposure time will be drastically reduced for most entry level 'accidental' UV capable lenses.

You will only get a thin slice of the 330WB80 transmission curve when using most lenses, and it will have less false UV color.

So for someone who doesn't have some special expensive lens, I would not recommend that filter.

Also, for me, I prefer 52mm filters out front for my equipment. Makes things much easier for me.

 

As far as the Baader U, I have not encountered any angle dependent problems that I can think of, other than when 'pushed' to extremes that I never use in real life photos.

Baader U is more expensive. It should have a slightly faster exposure than a U-360 2mm + S8612 2mm stack (about 1/3 stop I think).

I think it is fairly well agreed upon, that if you are going to get only one UV only dichroic filter that the Baader U is the one to get.

 

I recently priced several other more uncommon 48/52mm UV only dichroic filters, DUG11, UG11-IRBII, and Asahi, and they were all over $500 each.

The 330WB80 is inexpensive, but for me, too small, and for most it lacks a valuable part of the UVA transmission just under 400nm which less UV able lenses will benefit from.

 

Baader U pushed to an unnatural extreme.

[Stefano]

The reason I might not recommend one of the 330WB80 filters to someone with less than a UV-Nikkor lens is because those filters cut off the top part of the UVA, so not as much violet UV below 400nm.

So the color palette will be more monochrome, and the exposure time will be drastically reduced for most entry level 'accidental' UV capable lenses.

You will only get a thin slice of the 330WB80 transmission curve when using most lenses, and it will have less false UV color.

My camera is a compact point-and-shoot type camera, so the lens are small (and thin). They may transmit UV better than the average "big" lenses you use on a reflex or a more professional camera.