filter question, cheap alternative to (UV pass near IR block) midopt sp635?

[Johan]

Hi,

 

Looking for a cheap alternative to this material, a (hopefully) magenta cast killer -

 

http://midopt.com/filters/sp635/

 

Anyone here know of such a thing (to cover flash, not pure optical)?

 

Thank you

 

-Johan

[Andrea B.]

Key question: what kind of IR blocking are you looking for, if any?

 

The filter you show looks like Schott BG 38. So I would contact Cadmium via PM to see if he has any larger pieces of BG38, BG39, BG40 or S8612 that could be cut to fit a flash. (Choice from those 4 glasses depends on your IR blocking needs.) Schott glass would be the wisest choice for quality reasons. However, it will not be coated like the MidOpt filter you show. Thus you will have to be diligent about maintaining it by using cerium oxide polish or a hydrogen peroxide bath regularly to keep oxidation at bay.

 

There is really cheap Chinese ZBW blue-green glass on Ebay, but transmission is usually not equivalent to Schott in the same thicknesses. And IR transmission is usually not equivalent either. It is hard to predict what you would get. Z glass would also be uncoated and need maintenance.

[Cadmium]

Johan, Indeed, Andrea has a good eye, that filter is close to BG38, although it doesn't transmit UV as deeply, and it transmits slightly more IR.

Like BG38, it doesn't completely suppress Red/IR, as you can see they both have lingering transmission all the way from 700nm to 1200nm,

so these are not filters to use for Red/IR suppression in UV stacking.

BG38 is often very good for visual photography with a full spectrum camera. BG38 is my personal choice for visual shots with my Nikon cameras, and BG40 is also very good.

So the SP635 might work OK for visual shots, however it would look more red than BG38 and have a little more IR.

Returning to your question, I am not exactly sure what you are looking for when you say, "magenta cast killer".

As we have discussed before, 'Magenta' is a mix of high red and low violet, I will let Andrea define 'Magenta', but magenta doesn't really exist at any one point on a graph,

it is more of a mixed reflected color, not a defined color of light, therefor I am not sure what nm you need to block.

There may be other more common options for flash, if you are using it for visual shots.

If you want to use the filter for removing red from the light, and make it more blue, then BG38 or SP635 would not work for that,

you would want to use something that cuts off more red, such as S8612, and thicker S8612 cuts even more red.

Here is a comparison to Schott BG38 (on the lower linear graph).

 

Schott BG type glass compared.

[Johan]

Thank you, this is tremendously useful and informative.

 

My problem is that I converted two flashes to full spectrum for UVIFL but the images are awfully magenta-"ey".

 

I figured the excess red contributed to this, so I'm looking for that red blocker that also still passes UV.

 

I just did inventory of the bits I have from my non skint days, and I do have an 8612,phew.

 

(moved to a new house last year so very skint, just trying to make most of the bits I have already)

 

So maybe 8612 over one and baader-u over the other, that might help.

 

 

But ultimately the difference between the flashes and the nichia are probably also the broader wavelength spread produced by the flash?

[Andy Perrin]

Wait, is the camera blocking UV well enough? UV looks magenta when un-white balanced.

[Cadmium]

You probably already know all this, but for a full spectrum camera, generally, first we filter that flash with UV only filtration, typically U-340 2mm + S8612 2mm, which only transmits UV, and particularly the 365nm peak which is best for inducing visual fluorescence.

With a full spectrum camera, we also need to filter the lens with a visual range filter, such as BG38, Baader UV/IR Cut, BG40, or the like,

and stack that with a 420nmm or 435nm longpass filter to remove any reflected UV from being seen by the camera.

 

'Magenta' may mean something different to you than it does to me.

 

Huge difference between the flash and Nichia,

but either one should have the U-340 2mm and the flash needs the S8612 also.

If you use the Baader U, then you don't need to use anything else stacked with it, but you may want to use something on the camera lens to remove reflected UV light from the target.

[Johan]

Again, thanks!

 

I don't have a full spectrum camera, just full spectrum flashes.

 

UVIFL is perfectly achievable with the Nichias + filters I have, but the flashes don't yield the same results, too magenta

 

For various reasons I want to correct this at the light output end, without resorting to white balance or RAW fiddling

[JMC]

I inherited a couple of filters from a lab when it shut down. These were Kopp 5860, from here;

 

http://www.grayglass...s.cfm/conid/203

 

There were 6 inch by 6 inch and used for flash units for a fluorescence rig. The original purchase cost in 2013 was $335 for each one (at 6" square). Ok they don't have the IR blocking that some filters have, and UV transmission isn't as high as 6812, but they were at last available in large sizes for flash units.

[Johan]

I inherited a couple of filters from a lab when it shut down. These were Kopp 5860, from here;

 

http://www.grayglass...s.cfm/conid/203

 

There were 6 inch by 6 inch and used for flash units for a fluorescence rig. The original purchase cost in 2013 was $335 for each one (at 6" square). Ok they don't have the IR blocking that some filters have, and UV transmission isn't as high as 6812, but they were at last available in large sizes for flash units.

 

Yes I have some of those, thank you you-know-who, and they're good for the flash filter.

 

But, they do eat light, I also have enough U340 to cover a flash and it transmits a good extra stop which is v. handy

[Andrea B.]

So, the problem is that the visible fluorescence has a magenta cast when it is induced by UV-flash rather than by a UV-LED torch.

 

Did I get that correctly?

 

Some magenta cast might be from recording some visible violet output from the flash. Visible violet is recorded in various ways by our cameras but often appears to be the blue + red mix we call magenta. Depending on the camera the mix can lean more towards the blue side or more towards the red/pink side. But the confounding factor is that very high UV can also recorded in magenta/violet/purple tints.

 

If the colour you are seeing is more pink than magenta, then it might be from high red or low IR. In low saturations, it is difficult to know whether you have magenta or pink unless you sample the raw colour.

 

To repeat what is needed for proper visible fluorescence reording:

• The flash is filtered with a UV-pass filter so that only UV light is inducing the visible fluorescence. In particular you must block any violet output and IR output.
  

• The lens is filtered with a UV/IR blocker so that only visible light is being recorded. A filtered lens blocks the high UV which can get recorded as magenta/violet/purple and also the IR which can cause pink/reddish casts.
  

• An unconverted camera has an internal UV/IR blocker, but it is often not strong enough to overcome flash output. That is why you must filter the lens.
  

 

Now, even with proper filtration on the light and on the lens, white balance in a visible fluorescence photo is a big problem to be solved.

• The only accurate way to obtain proper white balance in a vis fluor photo is by using a standard which fluoresces visible white or visible grey under UV light. Those are expensive.
  

• If you choose not to use such a vis fluor standard, then you must try various white balance settings in your camera to see which one gets the fluorescent colors closest to what you are actually seeing (in the dark before your eye adjusts and moves to scotopic vision). Some folks here on UVP have used Daylight, Auto WB and very high K settings to get closer to what was actually seen.
  

• Some cameras permit white balance fine-tuning. For example, current Nikons have a color grid which would let you move white balance away from magenta and towards whatever other colour would best work to give more accurate white balance in the fluorescent photo.
  

• It is quite often unavoidable that some work must be done in the converter/editor to bring the visible photo closer to the actual visible colours seen while making the photo. If you are not using the double filtration, then adjustment in the converter/editor is unavoidable.
  

An alternative way to obtain proper visible fluorescent colours is to know the wavelength of the emitted vsible colour and use a narrowband filter on the lens to trap that.

 

Quite a lot of the visible fluorescent colours seen in "black light" photos across the internet are not accurate. :D

 

In this topic, Mark shows a cure for pink/magenta cast using Baader UV/IR-cut filter.

http://www.ultravioletphotography.com/content/index.php/topic/2168-uvivf-filters-evaluation/

[Johan]

In this topic, Mark shows a cure for pink/magenta cast using Baader UV/IR-cut filter.

http://www.ultraviol...ers-evaluation/

 

Thanks Andrea for taking the time to write this up, good overview.

 

I do actually have that Baader filter as well, think I bought it in 2016 in response to similar problems ;)

 

I've erm forgotten rather a lot.... onwards and upwards

[Cadmium]

Here is a typical UVIVF full spectrum camera stack I would use. One could easily use whatever BG glass is preferred, or Baader UV/IR Cut if desired,

and you could use GG435, or any other, or Wratten Gel or Schott KV if you can find one still.