Jump to content
UltravioletPhotography

Unmodified cameras for UV - a very scientific approach :)


Timber

Recommended Posts

I just did a very scientific* test today :)

So what made me think that for UV I use a filter stack of Hoya U-360 and a Schott BG40 and the BG40 is something like a UV-Weak hotfilter, so I was curious how will it compare to hotfilters.

In the test I've used:

Modified camera: Sony NEX-6 with Spectrosil 2000 replacement

Unmodified cameras: Sony A7ii, Sony NEX-5r

Lens: Meritar 50mm f2.9 EXA version (that's my best UV lens IMHO)

Filters: Hoya U-360 only (in the room there was no IR source, therefore no IR contamination), I've repeated the test on the NEX6 with BG40 on and got the same results.

Light: 36x 360 UV Led light, has some purple light as well.

For the test I also used industrial UV blocking safety goggles.

 

So what I did I put the safety goggles over the led-light that half of the leds were covered, half was not. I tried to get the same exposure with all cameras (based on histogram). On the photo only the ledbulbs were visible, everything else is dark. I used ISO1600 and f2.9 (wide open) in both cases. I photographed the led light directly.

 

The led light has some visible purple as well which was coming through the clear glass of the goggles, so I had some light on that end, but the difference between the covered and non-covered side of the LED is clearly visible (a lot). Surprisingly the unmodified cameras and the modified cameras recorded roughly the same image, so the unmodified cameras are actually usable for UV photos (I assume near UV only). But their transmission seems to be about 5EV less than the unmodified camera (1/4 vs 1/125).

 

* scientific test - I had safety goggles on and a white lab coat (which was black and a bath robe, but hey... to me it was like a white lab coat, okay?)

 

Did anyone else tried unmodified cameras for UV photography?

Link to comment
You didn't mention the thicknesses, but I did computational tests of a large number of different stack combinations and U-360 with BG40 is pretty far from ideal. The unmodified camera will probably not pass much of the shorter wave UV in my experience. (Said experience was with the Nikkon Coolpix 950, so a very old camera, however. But if anything, newer cameras would do worse, because they would have better blockers. With the 950, I don't think I got anything below 380nm, although that was enough to see flower patterns.) Anyway, using a filter that blocks the short waves with a camera that blocks the short waves would have exactly the effect you mentioned -- exposure time is worse, but you get the same end result if you wait long enough.
Link to comment

Actually, U-360 2mm + BG40 2+mm works rather well for a UV-only stack. Better to use S8612, which only needs about have the thickness as BG40 for Red/IR suppression.

Without knowing the transmitting of the stock camera internal cut filter, it is hard to say how much UV is getting through to the sensor, but most are designed to block UV.

Preferably lighting for UV-only, with a good UV transmitting lens and filter stack, should go down to 320nm. So, for example, 365nm LED's would only provide a rather monochrome photo, with some blue depending on the actual upper bandwidth of the LED. With MTE/Nichia the band is rather narrow, so you would get a mostly monochrome photo, once white balanced.

 

Both BG40 and S8612 transmit UV down to about 320nm/300nm range. You may still have a peak transmission of about 360nm, but there is a lot of bandwidth on either side of that.

Link to comment

Cadmium, I know with the Coolpix 950 and 330WB70 filter (which does fine with my NEX-7 and Noflexar), I could not see ANY false UV color, even after white balancing properly, in full sunshine or with UV torch. That camera/built-in lens combo just did not allow for short wavelengths to get through. Even dandelions did not show as yellow.

 

e.g., here is a buttercup after WB (Coolpix 950, 330WB70, Tuofeng UV torch, and a white balance off the plate):

post-94-0-35137800-1485703499.jpg

Link to comment

I did all my UV photos with this filterstack and I think U360-BG40 stack is quite good for UV only photography, but some visible purple is going through for sure. I think my filters are 1mm thick, honestly... I don't remember :)

I attached the test, where you can see that where UV was not blocked the difference is significant. Don't get me wrong, I am not saying unmodified cameras are good for UV photography, but they are having some possibilities.

post-61-0-46422500-1485720707.jpg

Link to comment

Your U-360 and BG40 are both 67mm diameter threads x 2mm thick each. I have ESP. ;-)

It is essential to know the filter thickness even when used alone, but certainly when used in any kind of stack.

A slightly preferable stack might be U-360 2mm + S8612 1.25mm or 1.5mm.

 

The top pic is visual, using stock camera?

The bottom pic is stock camera with U-360 (only) on the lens?

 

Then the test shows that the BG40 is not blocking the LED light, but the the goggles are blocking some amount of the LED light.

BG and S8612 filter glass is suppose to transmit UV-A and Visual. So the fact that you see 'purple' (lavender) through the BG40 is expected, otherwise, a U+BG stack would not work for UV-A only stacks.

 

You block blue and most of the visual range with the U glass filter.

If the U glass is too thin, you will not block as much of the visual range, you can get a blue leak this way, depending on the U glass type, and/or a 550nm+ leak, again, depending on the U glass type.

With your 2mm + 2mm thicknesses you are not seeing a blue leak, however, LED's can be intense enough on direct view to cut through very strong OD, so sometimes not the best test.

The BG is for blocking the Red/IR of the U glass filter.

Link to comment

Your U-360 and BG40 are both 67mm diameter threads x 2mm thick each. I have ESP. ;-)

(snort)

 

For what it's worth, my particular torch may be very narrowband. It's a 365nm, but I snapped this photo of my lentil soup earlier (BG39 2.5mm/UG11 2mm, Telemegor lens, 30" ISO800 F11) WB off the PTFE from you, Cadmium ;)

post-94-0-74801000-1485745722.jpg

Link to comment

The top pic is visual, using stock camera?

The bottom pic is stock camera with U-360 (only) on the lens?

Exactly. Both were taken with the A7ii. Once it's summer here in the UK I will try to do the same experiment with flowers outdoors, if I am free that day. :)

 

What I am trying now is to use "back filters" with an EOS-FE adapter. So far it works well with IR and ND1000 filters (IR fisheye is fun), but to do so I need to buy new filters (30.5mm)... Any advice what's the best filter stack for 340-360nm?

Link to comment

Andrea the problem is I went through all of them. That's why I am confused LOL

I was hoping someone went through a few stacks / options and could share a more personal recommendation.

Link to comment

The best U-pass filter or filter stack is the one which supresses IR leakage to between OD3 - OD4.

 

Examples:

U340 (2mm thickness) + S8612 (1.5mm thickness)

StraightEdgeU, BaaderU, AndreaU

 

And I'm sure there are more examples.

 

Vendor links are in the Filter Stickies in case you want to pursue purchase with either our filter stack vendor or our UV-filter vendor both of whom can provide information but not sales links within posts. Or you can privately PM to either "Cadmium" or "Reed F. Curry" for more information.

Link to comment

Timber, When you say 340nm to 360nm, do you mean JUST that bandwidth, or to include that bandwidth?

Both U-360 (or UG1) and UG11 (or U-340) include that bandwidth. I would use 2mm thick, which is what your U-360 is.

 

Several questions are factors.

1) What is the transmission/suppression of the internal filter. This is something that might not be easy to define.

2) What is the UV transmission limitations of the lens. Some will cut off UV rather high, even some above 360nm, others above 340nm easy... other transmit all the way down to 320/330nm or below if you have a special lens.

 

So using the U-360 2mm thick alone on you lens, with UV-only light that has no Red/IR in it, should probably give you a UV-only image with no Red/IR,

however, the depth (band width) and amplitude of the UV will be defined by the internal filter transmission, which is not something we have defined.

In most cased the internal cut filter is designed to cut UV rather high, higher than 360nm, but this varies.

Link to comment

Cadmium,

The filter stack I want to buy is not for my unmodded camera (but certainly will give it a try :) )

 

I have a Sony NEX-6 converted for Full Spectrum (Spectrosil 2000 as replacement - transmits from 170-1300nm - so that's not a limitation), but I am not aware of the microlens UV cutting abilities. As of Lenses I have some EL-Nikkor I can use and so far the best results seem to come from a Meritar (Exa version). I am thinking about the U-340 this time, but not sure which IR supressor would give the best results. But I am open to a dedicated filter like Baader U or Andrea U as well... but the main issue with those, they are only coming in 1.25" and I have not found any step up/step down rings between 1.25" and metric. Unfortunately I am not in the situation anymore where I can experiment a lot :)

Link to comment
I think there are solutions for the Baader. Have you looked at RafCamera? I got one of their adapters the other day and it seems very high quality. They sell 1.25" to M42 adapter rings.
Link to comment

The filters you mention come in 2" (48mm-52mm) versions. I don't understand about the 1.25" question.

I would use UG11 instead of U-340, unless you are using thick U-340, like 2mm, because it has a visual range leak when thin, which may or may not show up in photos, but it is there.

I actually prefer a U-360 stack.

S8612 is the best IR suppressor and cuts the least UV. The thickness of each glass is a balance. So for example you might get U-360 2mm and stack it with S8612 1.5mm.

With UG11 you can go thinner, and get 1mm to 1.25mm and stack it with S8612 1.75mm+.

If you wanted to use U-340, then use 2mm + S8612 1.5mm+.

Link to comment

I have a Sony NEX-6 converted for Full Spectrum (Spectrosil 2000 as replacement - transmits from 170-1300nm - so that's not a limitation), but I am not aware of the microlens UV cutting abilities.

You are forgetting about sensor cover glass - I does not pass anything below 310 nm.

Link to comment
The Andrea U and the StraightEdgeU are only available in 52mm. The BaaderU is available in 2" (actually 48mm, IIRC) and 1.25".
Link to comment

You are forgetting about sensor cover glass - I does not pass anything below 310 nm.

I thought that's the micro-lenses limitation. But (at the moment) I have no intentions to go lower than 340nm peak, so I think 310nm+ is perfect for me. I also lack the lenses that would go much deeper than that (my best is the E.Ludwig Meritar, and the EL-Nikkor 80mm)

Link to comment
Alex, the Spectrosil 2000 is replacing the sensor cover glass, I think? If this is one of eeassa's cameras it is, anyway.
Link to comment

Alex, the Spectrosil 2000 is replacing the sensor cover glass, I think? If this is one of eeassa's cameras it is, anyway.

 

Please show me the link. If we are talking about the same eeassa from eBay, here is a quote:

 

"The internal IR hot filter has been removed and replaced with a filter made from uncoated Spectosil 2000 Fused Silica"

 

Ripping off sensor cover glass is a dangerous business - it is attached to the frame with strong epoxy that is very resistant to chemical solvents or temperature. Anyone who provides such service will charge pretty penny for that.

 

An additional remark - adding a piece of uncoated glass instead of ICF ("IR hot filter"), even if it is Spectrosil 2000 has one one use - maintain proper focusing, and only if it is of the right thickness. As an added bonus, it reduces overall transmission and increases internal reflections. For mirrorless cameras I would advice against it.

Link to comment

If the cover glass only transmits down to 310nm, then what is the point of Spectrosil, when other glass types transmit lower than 310nm?

How do we know that Spectrosil is being used, if the cover glass prevents anything under 310nm?

Link to comment

1) eeassa has claimed outright that he uses Spectrosil to replace the IR cut in his camera mods

2) I didn't realize that the sensor cover glass and the IR cut were not essentially one piece, which I think is why I misremembered above.

3) You need something to replace the IR cut or the optical path length changes and you can't get infinity focus anymore. When I converted my Coolpix, I found that out the hard way the first time, and I had to buy a piece of quartz to replace it and remodify the camera.

4) Using Spectrosil at least does not further limit your bandwidth if the cover glass cuts at 310. Spectrosil transmits to UVC I believe.

 

Link to comment

1) eeassa has claimed outright that he uses Spectrosil to replace the IR cut in his camera mods

2) I didn't realize that the sensor cover glass and the IR cut were not essentially one piece, which I think is why I misremembered above.

3) You need something to replace the IR cut or the optical path length changes and you can't get infinity focus anymore. When I converted my Coolpix, I found that out the hard way the first time, and I had to buy a piece of quartz to replace it and remodify the camera.

4) Using Spectrosil at least does not further limit your bandwidth if the cover glass cuts at 310. Spectrosil transmits to UVC I believe.

 

A number of other types of glass also transmit below 300nm, and I think all full spectrum conversions probably use glass that transmits below 300nm.

So does it make any difference if it is Spectrosil or some other type? (I ask that question of Alex too).

I have often thought Spectrosil was over rated, for several reasons.

As you have found yourself, in some cases claims are not always true, and such claims become especially hard to prove when evidently there is no way to really test the claim.

 

You might test a camera transmission limit with a 12.5mm narrow band (10nm) -300nm bandpass filter attached to a body cap, but tape it up real good to prevent light leaks, and use a strong UV-B/C light source, and leave the room when the light is on.

Link to comment

Please sign in to comment

You will be able to leave a comment after signing in



Sign In Now
×
×
  • Create New...