Jump to content
UltravioletPhotography

Unmodified Nikon D810 for UV/IR? Well, no.


Andrea B.

Recommended Posts

Note: Another series of D810 UV/IR shots is posted later on.

 

I tried my unmodified Nikon D810 + Novoflex 35/3.5 Noflexar for some UV/IR shots.

 

IR shots with a B+W 092 IR-Pass filter were just possible but needed long exposure times with moderate to very high ISOs at f/5.6.

 

Moving to the stronger B+W 093 IR-Pass filter proved difficult even at ISO-8000 for f/5.6.

 

The UV shots with Baader-U were impossible at ISO-8000 for f/5.6.

 

Maybe with really, really strong UV/IR lighting I could get better UV/IR, but I'm not sure it is reasonable to try this. An unmodified D810 just really shouldn't be used for UV/IR work, imho.

 

Visible Reference Exposure

Baader UV/IR Cut Filter: f5.6 for 1" @ ISO-400 with On-Board Flash.

There was low ambient light, but I was shooting in "light box" conditions so a lot of it was blocked. I did not need a Baader UV/IR cut filter, of course, but I wanted to use the unmodded D810 just as I would use a modded version for this kind of work. The D810's suggested metering of 1" makes this a bit underexposed.

d810testVisFlash_f5.6ss1iso400_20150215wf_3787.jpg

 

 

Visible Alternate

Baader UV/IR Cut Filter: f5.6 for 2" @ ISO-400 with On-Board Flash.

The camera metered 1", but here's a 2" Visible exposure which is more evenly lit because it makes some use of what ambient light there was.

d810testVisFlash_f5.6ss2iso400_20150215wf_3738.jpg

 

 

IR Exposure #1

B+W 092 IR-Pass Filter: f5.6 for 30" @ ISO-8000 with one (1) On-Board Flash.

This is a bit over-exposed. The 092 passes a small amount of high red.

d810testIr092Flash_f5.6ss30iso8000_20150215wf_3775.jpg

 

IR Exposure #2

B+W 092 IR-Pass Filter: f5.6 for 30" @ ISO-1600 with one (1) On-Board Flash.

Now the foto is under-exposed by about a stop.

d810testIr092Flash_f5.6ss30iso1600_20150215wf_3777.jpg

 

 

IR Exposure #3

B+W 093 IR-Pass Filter: f5.6 for 30" @ ISO-8000 with one (1) On-Board Flash.

The 093 passes no Visible light. The sunflower is just barely recorded in pure IR.

d810testiIr093Flash_f5.6ss30iso8000_20150215wf_3771.jpg

 

 

UV Exposure

Baader-U UV-Pass Filter: f5.6 for 30" @ ISO-8000 with six (6) SB-14-UV Flashes held about 2" from the sunflower.

Nada, muchachos !!

d810testIVBaadSB14_f5.6ss30iso8000_20150215wf_3757.jpg

Link to comment

Whoah!

Something doesn't seem right there.

Sure that Novoflex passes UV?

 

I use 3 flashes with everything I have been shooting. Try 3 flashes with that IR. 30s is too long.

Really surprised mostly at that UV shot. There is nothing there.

 

I am going to try my 5D & 30D tomorrow. If they are crap then perhaps Canon realized their lenses are superior at blocking unwanted light and they reduced the coating on the Mark IV and maybe others once they realized it wasn't needed? My theories are just that as I have no background to qualify that.

 

I will post my 093 shots. I was able to get them at 1/60s @iso 1600...

 

See below:

IR: Canon 1D Mark IV Unmodified, EL-Nikkor 80mm f5.6, 3 Modified Vivitar 285 HV's Unfiltered, Halogen light Unfiltered, 1/60 s @ f/5.6 ISO 1600, Schneideroptics 093 Filter , Daylight/Sunny WB, 1 Click Photo Ninja

post-51-0-82069300-1424054130.jpg

 

-D

Link to comment

The Canon 1D is old, isn't it? and the D810 is new and uses much better internal filtration dedicated to cut down on IR (and UV, as collateral damage as it were). So Andrea's experiments just confirm what could be expected from a modern digital camera.

 

I remember making a few test shots ten years or so back on a friend's Canon 1D , just pushing my UV-Nikkor flush against the mount. Apparently it could record, albeit weakly, the same as my D1H at that time in UV.

 

The Noflexar 35 mm f/3.5 has excellent UV transmission, only about -1.3 EV down re the UV-Nikkor and Coastal 60. It might not let through much below 330 nm or so, but should more than adequately suffice for the described test setup.

Link to comment

OK that makes sense except my Canon 1D Mark IV was made in 2011

I guess that could be considered old in digital camera terms. Although possibly only 3 years difference between it and the D810.

 

I think it is interesting to think about. I have never seen a review of a modern digital camera that claims UV or IR contamination right off the bat. It wouldn't sell.

 

Since the 1D Mark IV does pass some UV and IR but shoots at professional quality in the visible range suggest that maybe designing much better filtration is not always necessary or too costly compared to other methods like lens coatings..

Since I found it nigh impractical to shoot with my modern macro lens and 1D Mark IV but possible without too much difficulty to shoot with a UV lens--it seems to me most of the UV/IR blocking goes into the lenses (at least with my testing with this camera/lenses only). I am testing some other Canons today hopefully.

 

-D

Link to comment

The nomenclature of these Canon 1D|DS Mk. models is highly confusing. The camera I tried might have been 1D Mk.1 (or 2 or whatever. Anyway, the top Canon PJ model at that time).

 

It is correct that today's complex optical designs and their advanced multi-coating tend to make the lens impervious to UV. However, it's all in the exposure received by the sensor. Even such a lens as the AFS 14-24 mm f/2.8 Nikkor can capture a UV scene. Not deep into the UV range of course, but enough to give pleasing results.

Link to comment
I went out in the cold, cold, cold weather and made a couple of test shots with the D810 in the sunlight. Did much better at IR. But no better in UV. I'll get the posts up in a mo'.
Link to comment

Cool!

Where did you see sunlight? You intrepid UV fanatic. I feel better knowing that I am not the only freezing my but off trying to get some pics... :lol:

My UV shack looks like the abode of old man winter...

 

-D

Link to comment

We had actual sunlight here until about 3pm when it started to haze over. It had gotten up to about 15°F when I went out - relative warmth compared to the 4°F of this morning. I was worried that the UV-Nikkor might not do well in below freezing weather!! But it seemed to be OK. When I came in I put a down jacket over it temporarily so that it wouldn't warm up too quickly. I have absolutely no idea if that makes sense. :lol:

The D810 battery held up very well and didn't drain as fast as I thought it would.

Link to comment

Unmodified D810 UV/IR Test with UV-Nikkor in Sunlight

All exposure data is on the photo label. Any View NX2 EV changes are given above the photo. I used some exposure slider tweaks to try to make the histograms all similar in each group.

 

D810 Visible Reference Photos at ISO-64/800/1600/3200/6400.

I couldn't go to ISO 12800 in the bright sunlight because the maximum speed on the D810 is 1/8000".

There are no photos between base ISO-64 and ISO-800 because the unmodded D810 could not begin to record any IR or UV until around ISO-800.

The bright snow highlights were pulled back a bit. There's lots of headroom in D810 shots for pullbacks. No other edits.

 

EV 0

d810visSun_uvNikkf8ss1-250iso64_20150216wf_3844.jpg

EV -1.0

d810visSun_uvNikkf8ss1-1600iso800ev-1_20150216wf_3850.jpg

EV -1.0

d810visSun_uvNikkf8ss1-3200iso1600ev-1_20150216wf_3852.jpg

EV -1.0

d810visSun_uvNikkf8ss1-6400iso3200ev-1_20150216wf_3856.jpg

EV -2.0

d810visSun_uvNikkf8ss1-8000iso6400ev-2_20150216wf_3858.jpg

 

D810 Infrared Photos with B+W 093 IR-Pass at ISO-64/800/1600/3200/6400/128000

It would be tough to have to use such long exposure times in outdoor IR shooting. The noise isn't too bad. I do not have any noise suppression turned on in-camera or in-editor.

No edits except for exposure slider EV done in NX2.

 

EV +1.5

d810ir093Sun_uvNikkf8ss30iso64ev+1.5_20150216wf_3796.jpg

EV +1/3

d810ir093Sun_uvNikkf8ss6iso800ev+1-3_20150216wf_3798.jpg

EV +1/3

d810ir093Sun_uvNikkf8ss3iso1600ev+1-3_20150216wf_3800.jpg

EV 0

d810ir093Sun_uvNikkf8ss2iso3200_20150216wf_3802.jpg

EV 0

d810ir093Sun_uvNikkf8ss1iso6400_20150216wf_3805.jpg

EV 0

d810ir093Sun_uvNikkf8ss1-2iso12800_20150216wf_3809.jpg

 

Ultraviolet Photos with Baader-U at ISO-64/800/1600/3200/6400/128000

Painful, very painful. And there is some flare too.

No edits except for exposure slider EV done in View NX2.

 

EV +2, certainly didn't help at ISO-64.

d810uvBaadSun_uvNikkf8ss30iso64ev+2_20150216wf_3812.jpg

EV +2

d810uvBaadSun_uvNikkf8ss30iso800ev+2_20150216wf_3814.jpg

EV +1.5

d810uvBaadSun_uvNikkf8ss30iso1600ev+1.5_20150216wf_3818.jpg

EV 0

d810uvBaadSun_uvNikkf8ss30iso3200ev+.5_20150216wf_3820.jpg

EV 0

d810uvBaadSun_uvNikkf8ss15iso6400ev+.5_20150216wf_3824.jpg

EV 0

d810uvBaadSun_uvNikkf8ss8iso12800ev+.5_20150216wf_3836.jpg

Link to comment

Thanks for this. In IR your description says 092 pass filter but the photos description says 093.

 

I would like to duplicate this scenario with my Mark IV as soon as I see some sunlight when at home.

 

That UV is painful to look at. Mine may look the same without the flashes. B) Fingers crossed!

 

-D

Link to comment

I was using the 093. I fixed the typo - thanks for catching that.

 

When looking at those outdoor IR photos in the editor at 100%, they seem really much more muzzy than a typical IR photograph although that's not apparent in the downsized versions posted above. Muzzy = softer than usual, kind of foggy. The UV-Nikkor may not be all that great as an IR lens? It's not IR-corrected.

Link to comment

ok, noted !! thx.

So it is just that the IR is fighting to get thru the D810 internal filter which contributes to the noize & muzziness.

Link to comment
  • 8 months later...

Well, this explains the non-converted D810's poor performance in UV light: Spectral Transmission of Digital Sensors

The D810 and other digital cameras have an internal filter which cuts from violet 420nm on down. So no UV and no spectal violet can be recorded.

No surprise really that UV cannot be recorded. But I don't understand the reason to cut violet?

Link to comment

Simple. In order to cut UV and not using a dichroic filter (which inside a camera could cause a lot of other issues), the roll-off has to start a little in advance of the true cut-off point.

 

Film cut off more and nobody complained back then.

Link to comment

Of course film users didn't complain. There was no technology back in those days so they were stuck with what they had. Now we know we know these these things can be fixed!

.

.

.

.

.

.

just kiddin'. . . . . . . . . :D :D

Link to comment
The D810 and other digital cameras have an internal filter which cuts from violet 420nm on down. So no UV and no spectal violet can be recorded.

No surprise really that UV cannot be recorded. But I don't understand the reason to cut violet?

 

As you are aware I have always thought it important to understand the properties of the internal cut filter hence I

published my first ICF transmission spectrum in 2006 http://www.beyondvisible.com/BV3a-ICF.html

I then posted here http://www.fotozones.com/live/index.php/topic/14945-the-icf-database-not-all-icf-are-created-equally/ in 2009.

 

My 5 cents worth, a portion of the email I sent Lloyd.

"While it is true that human vision "typically" goes below 400nm, under normal lighting conditions (re intensity) the response is quite weak and in many cases older people with aging (yellowing) eyes lose the ability to observe colours in the violet/blue end of the spectrum. Our visual response is also weak at the red end. The ICF implemented in many cameras seems to somewhat replicate our visual response with perhaps some added benefits.

Limiting the spectral range from 400-700 down to 420-680nm can reduce the effect of CA since most lenses are not well corrected for CA around 400nm. I seem to remember that you even commented once that "violet" fringing was almost universal amongst lenses and that it was due to uncorrected axial CA in the longwave UV to violet range. A simple application of the appropriate UVIR blocking filter should reduce or eliminate the issue. In this case the DSLR ICF 420-680nm."

Link to comment

Please sign in to comment

You will be able to leave a comment after signing in



Sign In Now
×
×
  • Create New...