Infrared block, broad UV and visible pass filter?

[ulf]

Speaking of spectrometer maybe prism can be used test UV transmission? It should be possible to calculate spectrum by refraction angle.

Older types of manual spectrometers sometimes used prisms to divide the different wavelengths.

All or at least most modern spectrometers use gratings instead.

[Avalon]

I contacted optical filter manufacturer in my country and he recommended me KG1. KG3 and KG5 filters. KG3 spectrum looks best - good UV transmission to 350nm but there might be infrared leak just red spectrum.

[dabateman]

Wait do you mean Scott glass?

If so I would write down the name of the person whom recommends this and never ask for their advice ever again.

Scott KG1 lets in 10% at 900nm, totally useless. Scott kg2 never drops below 10%, even more useless and Scott kg3 hits 10% at 800nm, not as bad as kg1 but totally useless. Kg5 hits 10% at 780nm, 5% at 800nm. Again useless, so useless you may be able to use these glass types for a full spectrum conversion and not really notice the slight loss in IR sensitivity. Our camera sensors are really sensitive to IR.

Infrared leak is an understatement.

 

Added:

Just looked up the user notes on Scott KG glass. I would not recommend this for full spectrum conversion as may solarize under uv light. You will lose uv sensitivity over time of use.

[Andy Perrin]

Yeah, that is not a good glass to use for our purposes.

[Cadmium]

The data sheet note mentioned on the KG# data sheets:

"Transmission changes are possible under the action of intense ultraviolet radiation."

That is true of many filter glass types, for example, UG1, UG2A, UG5, UG11, all have the same note, and other types also.

This note would apply to equivalent glass types of other brands, U-360, U-330, U-340... etc..

In most cases, I don't think we are using extreme UV in an ongoing way enough to experience such problems.

 

Which looks best?

[Andy Perrin]

I don't believe anyone has ever done better than S8612 for an IR blocker that also passes UV well. I don't see a reason to use any other glass type for UV reflectance filter stack purposes. (For other purposes, other BG glasses may have good uses.)

 

It is very common for people who haven't done UV imaging much to underestimate how easily we can get infrared contamination. Transmitting 10 or 20% is not anything like sufficient. You need to block everything past 680nm to OD3+, which is 0.1%!

[Avalon]

Ok, I got point. Still KG glass seems to pass little more UV as well infrared, seems it's problem with most filters that good UV tranmission means higher IR transmission as well. Then I have stick for now to QB21.

[Andrea B.]

I usually go for OD4, when possible.

 

IR leakage is supressed in commercially manufactured filters. Worth their price.

Good stacks can also suppress IR very well. We always prefer S8612 as an IR blocker when used with Hoya or Schott dual bandpass glass. Again S8612 worth its price.

[dabateman]

Avalon,

Look into the spectroscope section, where Dmitry has shown lens data filtered with ZWB glass and has tried to use a QB21 filter to cut the IR. If you actually look you will see the IR is not cut out.

I don't think you should waist your money on a QB cut filter.

[Cadmium]

I made this graph to illustrate the difference between S8612 and other possible BG type glass options that would be acceptable.

Points:

 

1) Any stack has to be calculated for optimum aspects of maximum UV peak with acceptable Red/IR OD suppression ratio.

The two types of glass used in a stack are important, but the thickness of each glass in a stack is chosen specifically, not blindfolded, not just "one of these, and one of those glass types", but specific optimally matched pairs of two glass thicknesses.

Don't buy either glass type without knowing the thicknesses, and carefully matching the thicknesses for optimal stack transmission, the formula/recipe needs to be thought out first.

 

2) Basic rule of thumb: S8612 works almost exactly the same as twice as thick BG40, so S8612 1mm = BG40 2mm, S8612 2mm = BG40 4mm.

You will see on the graph below, I am comparing both of those examples, and you see very little difference, except some very minimal extended UV depth, and minimal reduced Red/IR using S8612. So S8612 is preferred even for transmission, and thickness is reduced by half using S8612.

 

3) I have included the BG39 2mm plot to remind people of the comparison between BG39 and S8612, same suppression, but quite reduced UV transmission.

So I would not use BG39 for UV stacking. You have much more efficient options.

 

So in my opinion you have two choices for good Red/IR suppression glass for UV only stacks, S8612 and BG40 (keeping in mind the thickness difference).

BG38 type is not acceptable for UV-only stacking, a poor choice at any thickness.

 

[Cadmium]

I usually go for OD4, when possible.

 

IR leakage is supressed in commercially manufactured filters. Worth their price.

Good stacks can also suppress IR very well. We always prefer S8612 as an IR blocker when used with Hoya or Schott dual bandpass glass. Again S8612 worth its price.

 

Yes, OD4 is best to use. OD3.5 may work but OD3 and above is where you will see a leak for sure.

Anything stronger than OD5 is usually useless, and only cuts down on UV efficiency with no benefit.

[ulf]

Yes, OD4 is best to use. OD3.5 may work but OD3 and above is where you will see a leak for sure.

Anything stronger than OD5 is usually useless, and only cuts down on UV efficiency with no benefit.

Sometimes when the UV-levels are low and VIS+NIR levels still are strong I see hints of leakage even at OD4.

This has occurred in normal sun-lit macro scenarios for highly UV-absorbing areas of flowers that should be black in the image.

So far OD5 has never failed for me.

 

IMHO this rules of thumb is valid only for normal UV-A images.

Down at the UV-AB border

At shorter wavelengths below 350nm I would expect that even more attenuation is needed, due to the lower sensor sensitivity down there.

[Cadmium]

UVB, here we go again... :D

The idea of UVB has popped up a few times recently, but not good putting expectations of UVB into a beginner's head.

You will not be shooting UVB with a stack. The 'rule of thumb' I was applying to stacking, and more normal expectations, not for UVB which is nothing you will do with a stack or a Baader U.

 

A possible more likely scenario might be night time UV photos, but... still not a normal scenario, and you would need to test, prove, etc., first.

[ulf]

UVB, here we go again... :-)

The idea of UVB has popped up a few times recently, but not good putting expectations of UVB into a beginner's head.

You will not be shooting UVB with a stack. The 'rule of thumb' I was applying to stacking, and more normal expectations, not for UVB which is nothing you will do with a stack or a Baader U.

 

A possible more likely scenario might be night time UV photos, but... still not a normal scenario, and you would need to test, prove, etc., first.

Naturally I should have written shorter wavelengths below 350nm or instead of UV-AB.

Thank you the comment. I will edit my entry above to make it more clear.

 

Nice that you understood that I had problems with slight IR-contamination even with the new version of Baader U.

I have seen the same thing with the 330W80 and 330W70-filters from bjomejag too.

 

Never any problems with my stacks with my S8612, 2mm

[dabateman]

The 330WB80 improved is a good filter. I am not sure about the other versions.

You will need to stack for UVb, but with two specialized filters.

 

Cadmium advice about calculating the thickness and working with your UV filter and your IR blocking filter is the best. Unless you just want a one filter solution. Then your options are straight egde mk2 or Baader venus U filter. The 330WB80 improved can work but not really for a novice. You will need to sand it down slightly to fit into a 25mm filter. Or you can buy a 28mm filter and 25mm to 28mm step up ring. Screw the retaining ring out of the 28mm filter and use it to hold the filter in the stepup ring.

Then it will only cover a M43rds camera. So still the Baader or SEU would be better options.

 

For UVb you need a good light source, the sun is not a great UVb source.

[Andy Perrin]

Those 330WB80s can be mounted on the rear of the lens with poster putty (the stuff you use to hang posters in dorm rooms that have “no nails” rules). That also reduces dichroic artifacts.

[ulf]

The 330WB80 improved is a good filter. I am not sure about the other versions.

The 330WB80 improved is the best of the three filters of that type I have, but it also leaks a bit sometimes.

I have the 330WB80 improved, the 330WB80 and the 330WB70

 

Hera are some quick screenshots from my RAW-viewer.

Still I can see the difference on my monitor in the Post Preview.

 

U-360 + S8612, both 2mm. No leakage

 

Baader U, here with weak signs of leakage

 

330WB80 improved

 

BG40? 2mm

 

I think the flower is a Pilosella peleteriana, (Hieracium macrolepideum)

or a Pilosella officinarum, (Hieracium pilosella)

 

The petals are quite UV-absorbing.

[Andy Perrin]

Yeah, definitely consider rear mounting your 330wb80, Ulf.

[dabateman]

There is a lot of glow in your 330WB80 improved shot, UlfW. Mine does not look like that. But there will be variability in these filters as they are production run offs sold at a discount.

I will have to do a similar test to see if I see the same in direct comparison images. I have a 2mm U340, 1 mm and 2mm UG1 filters that I could stack with my 2mm S8612.

[Andy Perrin]

Dabateman - that glow in ulf’s 330wb80 pic is probably from dichroic artifacts. That’s why I suggested rear mounting it. (Lens hood also helps.)

[Cadmium]

So far I am not convinced that anything I have seen has much UVB in it.

Regardless, given that you need a special lens (UV-Nikkor, etc.) transmitting below 320nm, special light with UVB content, and a filter that cuts off any UVA which would tend to overpower any UVB in the mix...

(think of that like a BG40 filter used for visual, you see visual, you don't see the proportionally weaker UVA part of the BG40 band width).

 

There have been a few 'new' people on here recently with seemingly immediate expectations of UVB. At least one person didn't even have a converted camera yet.

I think it should be made clear to people that UVB(!) and UVC(!!) are rather illusive and out of bounds even with specialized lenses and filters.

Are results really that conclusive yet? How much of it is out of band? How much is IR or UVA? Is a UVB/UVA mix working like a BG40 visual shot, as I mentioned above?

Have I ever seen a UVB-only photo? I really don't know if I have.

I totally appreciate anyone who is trying to isolate a UVB photo, on the other hand I want to be realistic here as not to confuse or over complicate the expectations of the larger crowd.

[rfcurry]

Are we doing ourselves and the OP a disservice by only showing him one mechanism for getting good UV shots? It seems that all of the responses take it as a given that all UV-broad bandpass filters will be based upon a double-band - UV and NIR - filter like the Schott UG series or the corresponding Hoya.

 

The UV Bell curve of the UG filter represents a deficiency, not an advantage. Even the Baader U is crippled by its UG11 substrate. How do the many layers of dichroic coatings benefit that filter, when it still does not adequately block NIR in the 734nm range as some have proven? I would think that a simple stack of UG11 and S8612 would perform as well as the Baader at a lower pricepoint.

 

Coastal Optics has a number of interesting filters under the PixelTeq name. Below is one:

 

http://uvroptics.com/images/UV%20Bandpass%20Filter%20%E2%80%93%20350nm%20FWHM%2060nm_Page_2_800px.jpg

 

The right edge of the UV transmission is not as close to the 400nm range as I would like, and the price of $1417.00 USD for a 52mm is a bit off-putting, but it is a fine example of a precise UV bandpass filter.

 

I am pleased with UVROptics SEU Gen2 in its most recent revision. It presents the 380nm-400nm range that the UG-based filters lack. See below:

 

http://uvroptics.com/images/SEU%20Gen2%20Final%20Linear.jpg

 

So we do have more to present the UV reflection photographer than one approach. Let us continue to think of new ways to capture the invisible.

 

Thanks.

[Andy Perrin]

Reed, the price on that PixelTeq is VERY off-putting! How many Baaders is that, 3 or 4? Your SEU2 is certainly better from a price point of view.

 

For small filters, I think ThorLabs also has a nice narrow bandpass, essentially a "355BP10" blocked to OD5 out of band:

https://www.thorlabs.com/thorproduct.cfm?partnumber=FLH355-10

[dabateman]

Reed,

I don't think any disservice has been made. If you look at the actual question. The answer is a one liner. For UV and visible with IR cut you have 3 options. S8612, Bg39 and BG40. What you decide has trade offs.

We then have taken this simple question and answer out to 4 pages now. With lots of good random but some what on topic discussion.

 

[ulf]

Dabateman - that glow in ulf’s 330wb80 pic is probably from dichroic artifacts. That’s why I suggested rear mounting it. (Lens hood also helps.)

Good advice.

 

I have been a bit lazy sometimes with lens hood usage as my ionic filters seldom need that.

 

Rear mounting is not an option for my setup as I want to switch between many different filters quickly, for each composition or motive.

I also try to avoid exposing the sensor for dust as much as possible and want to keep the lens mounted.

 

I bought the 330W.. -filters just to play with, both on the camera and in the spectrometer.

 

This dichroic artefact phenomenon makes me want to examine the behavior more closely for transiting rays at wider angles.

How is the leakage changing by the angle of entrance?

I have to build some suitable mounting gadget for that.