Jump to content
UltravioletPhotography

Optical density has its limits


Recommended Posts

When makers of UV bandpass filters describe the virtues of their products they commonly extoll the NIR-blocking capacity, usually in terms of OD, optical density. I do it myself, and look for it from others.

 

Recently, a very knowledgeable customer wrote me that one of my filters was not providing the necessary blocking of near-infrared. I – foolishly - assured her that the filter’s NIR blocking had only a few wavelengths that are less than OD4, though still >OD3, according to both theory and testing on a spectrophotometer. More the fool me, the customer was using the filter in solar UV, a condition in which optical density is not a valuable measure.

 

Many UV photographers have control of the intensity of the ultraviolet input for their photography, either through the use of UV-dominant flashes or UV lamps. By insuring that the subject is lit by UV wavelengths greater in irradiance than the ambient NIR, the optical density of the filter is meaningful.

 

However, in conditions in which the lighting is purely solar input, e.g., outdoor scenes, the NIR will, usually, greatly exceed the UV wavelengths in solar irradiance. The relatively low ambient UV will require a significantly longer exposure time to provide the same image as that acquired with artificial lighting. All of that exposure time is open to the entire solar spectrum, NIR, Vis, and UV; the filter acting as the principal gatekeeper.

 

In such a case, an optical density of three for minimum NIR blocking is not always sufficient. If, instead of considering the OD, we look at the corresponding transmission, the reason becomes clear. An OD of 3 is a transmission of 0.1%. That is not a great percentage, but it is, conceivably, a great amount of NIR photons. If the irradiance of the solar NIR is five times that of the solar UV, the NIR blocking of the filter will not be sufficient to prevent some NIR leakage. It would seem then that a filter's optical density does not always, or in all conditions, indicate a UV bandpass filter's capacity for NIR blocking.

 

Regards,

Reed

Link to comment

Too true Reed

Here is a chart I like to remind people with, that shows how little UV arrives to us on Earth's surface & compared to the IR part of the spectrum & what is blocking it & more importantly, life on Earth would not have started in a UVC world.

Our Silicon imaging sensors are capturing between 300nm to 1100nm.

Col

 

post-31-0-75936100-1431485429.jpg

Link to comment

Agree with Reed that we need to consider the illumination source not just the filter properties as such. Plus factor in the response curve of the camera sensor as well, which is biased towards the IR end of the spectrum.

 

However, we shouldn't paint the situation entirely black. Otherwise there would be impossible to take UV photographs outside the controlled environment of the studio and that simply isn't the case. A Baader U2" of the newer generation(s) attenuates visible and IR enough to render quite convincing UV captures.

Link to comment
Compounding the UV/IR response of silicon sensors (assuming it is more or less the same) with the emission spectra of various sources (direct sun, overcast, naked flash, UV LED) would give the level of IR attenuation needed for each situation. I think a filter manufacturer might want to provide this information as a guide to help clients choose the right filter for their application. However, since UV photography is such a small market and most UV filters are designed for astrophotography, I'm not holding my breath.
Link to comment

We shouldn't forget the visible range either. Some filter designs start to transmit already in the bortderline red/NIR, or extend from UV into violet/blue. Both cases can cause issues for the UV photographer and contaminate or wash out UV patterns.

 

Of the two, I think having a bordeline red/NIR leakage is most troublesome, but either should not occur if at all can be prevented.

Link to comment

I am the "she" referred to in Reed's post. :D Thank you Reed for being discreet until I could get back to the forum from recent domestic and business events which have occupied my time.

 

The context of the story is that I have been testing part of my collection of UV-pass filters - the PrecisionU, the AndreaU and the CopperU. But I had not gotten all the results posted yet - only some from the AndreaU and BaaderU.

 

I was puzzled by the PrecisionU results when shooting ourdoors in strong sunlight and wrote to Reed about it. He has given it some thought and has responded above.

 

In Reed's post, important points have been raised. When testing our UV-pass filters (and other filters) we need to take some care to test them in all types of illumination in which we might use them. As we have seen, some UV-pass filters will work just fine under studio conditions with artificial, dedicated UV illumination, but will work quite differently for IR-heavy solar illumination. Let's not forget fluorescence shooting. And then there is the astrophysics useage (which we don't see here on UVP). When shooting my little wildflowers for documentary reasons, I typically carry a dedicated UV-flash to boost the UV and shorten exposure times even in the strongest sunlight. So I guess you would call that a mixed solar+studio shooting situation.

 

Additionally we should test our UV-pass filters with various types of subject matter: close subjects vs. far landscapes, documentary needs vs. artistic vision, technical vs. photographic and so forth.

 

I think we all already basically know all this, but do not always put it into play when testing filters and presenting the results. We also all basically know that there are times when an IR-blocker is needed for a specific UV-pass and times when it is not. I'm thinking that I need to beef up the general remarks in the Filter Sticky about all this. It seems to me that any UV photographer should own some kind of IR-blocker. Many who use the simple glass UV-pass filters cannot shoot UV without one. But there will be times when the OD-3 and OD-4 filters can also make good use of an IR-blocker.

 

I'm going to try to get in some shooting today and add a few tests with my SB-14-mod.

 

***********

 

The red/IR leakage is the troublesome one. It does surely contaminate UV fotos.

 

A tiny bit of violet/blue leakage is rather interesting to me. Unless we are shooting a narrowband UV-pass, we are always getting a bit of that - certainly the violet. And given that there is not always a clear-cut definition of where violet ends and ultraviolet begins (380nm?, 390nm?, 400nm?), I'll just leave this one alone for now.

Link to comment

Optical density is a perfectly valid measure of a filter's out of band rejection properties. The realization here is more to do with how high an OD is actually needed. Many have said for some time that OD4 may be considered minimal.

 

OD is commonly used to specify filters for various microscopy applications. The average OD is not the most useful descriptor, one really needs to know the minimum OD within the blocking range. This is why OD is properly given as ">ODX XXX-XXX nm"

Link to comment
Can you give an example of what all those Xs are? :D
Link to comment

Sorry, only trying to give generic format. A filter could be described as having an out of band blocking of say OD >6.0 with the wavelength range that meets that specification such as 250 - 320 & 383 - 500 as is the case for this filter from Edmund.

 

Now if you look at the plot they provide you see that they are not claiming OD >6 for ~500 to ~750. However they could fairly describe this filter has having out of band blocking OD >5 (or >OD5 if you like) from ~380-1000nm. This one may have a bit of an error on the short wavelength side which looks more like OD>6, 275-320nm.

post-24-0-90249000-1431529928.jpg

Link to comment
oh duh. Those Xs were for the range of the OD blocking. You even said it. Sorry I was being quite optically dense. :D
Link to comment

No, I was perhaps a bit opaque, low transmittance of insufficient illuminance.

 

Picture does help make the point better.

 

Rather interesting but expensive filter.

Link to comment
A Baader U2" of the newer generation(s) attenuate visible and IR enough to render quite convincing UV captures.

 

That is true within certain parameters of the Baader U2", PrecisionU, and other filters as well. Which is what I was describing in my OP. Once you put the Baader U in a low UV, high NIR, situation, it bleeds NIR and some visible light.

 

John, OD is indeed "a valid measure of a filter's out of band rejection properties", but only when considered in relation to the filter's in-band transmittance properties and the available light source(s). For example, some narrow band UV bandpass filters have a peak of 20% transmittance. Even if such a filter has a >OD4 for the entire out-of-band rejection, it will leak when the user attempts to capture a meaaningful quality photo in low UV and high NIR.

Link to comment

We can review what I have so far w.r.t. Red and IR leak in the BaaderU and AndreaU when used in strong sunlight.

 

Neither of them pass much at all in strong sunlight under the B+W 093 IR-pass filter which cuts in around 830nm, IIRC.

 

Both of them pass something in strong sunlight under the B+W 092 Red+IR-pass filter. It is that border between Red and IR which is the bugaboo for UV-pass filters it seems.

 

These are very long exposures, in my opinion, for strong sunlight - longer than we would typically shoot outdoors I think. YMMV, of course!

 

Everything at f/8 and ISO-200.

 

BaaderU 10" for B+W 093 IR-pass in strong sunlight.

The BU passes a little more IR above 830nm than the AU.

Unedited OOC.

baaderU093ooc.jpg

Raw Digger Raw Composite

baaderU093raw.jpg

 

 

 

AndreaU 10" exposure out-of-camera for B+W 093 IR-pass in strong sunlight.

The AU passes slightly less above 830nm than does the BU.

Unedited OOC.

andreaU093ooc.jpg

Raw Digger Raw Composite

andreaU093raw.jpg

 

 

 

BaaderU 10" exposure out-of-camera for B+W 092 Red+IR-pass in strong sunlight.

Unedited OOC.

stack_baaderU+092RedIr_sun_20150428wf_32295.jpg

Raw Digger Raw Composite

baaderU092raw.jpg

 

 

 

AndreaU 10" exposure out-of-camera for B+W 092 Red+IR-pass in strong sunlight.

This ooc set is very interesting because the 2% reflective "black" standard is still very dark after 10 seconds. Yet the 99/75/50% filters are quite light. I don't think I really know what is going on here. Recall in my Dandlion photos using the AndreaU that the center bullseye was dark with a green tinge.

Unedited OOC. (I did not even clone out the horrible dust bunnies and crud.)

stack_andreaU+092RedIr_sun_150428wf_32365OOC.jpg

Raw Digger Raw Composite

andreaU092raw.jpg

Link to comment

Both BU and AU pass some violet and/or violet-blue. The AU more so, as per its design.

 

Everything at f/8 and ISO-200.

 

BU + Baader UVIR-Cut for 6 seconds in strong sunlight.

Raw Digger Raw Composite

stack_baaderU+baaderUvirCut_sun_20150428wf_32275raw02.jpg

 

AU + Baader UVIR-Cut for 5 seconds in strong sunlight.

Raw Digger Raw Composite

stack_andreaU+baaderUvirCut_sun_20150428wf_32349raw01.jpg

Link to comment

John, OD is indeed "a valid measure of a filter's out of band rejection properties", but only when considered in relation to the filter's in-band transmittance properties and the available light source(s). .

 

You are correct to add that stipulation, it is the relative out of band rejection, or blocking OD relative to the OD of the transmittance band.

 

I would add that the bandwidth of the pass band is certainly also an important factor for total throughput.

Link to comment

BTW, I think that Reed in his capacity as UVR Optics offers an IR-blocker for his customers. Reed, would you like to give us the details?

 

(Folks, the UV world is a small one. UVP wants to provide information to members about various items used for UV photography even though as a website we do not accept or permit advertising nor have any affiliation with any vendors.)

Link to comment

Andrea,

 

Thank you.

 

UVR Optics' customers use our filters under an amazing range of conditions. We developed our filters to provide optimum transmittance of ultraviolet wavelengths, based upon the assumption that the photographer will be able to regulate the ambient lighting. However, we understand that certain conditions, e.g., low solar UV and high solar NIR, require greater visible and NIR blocking than is provided with the standard filter. Therefore, UVROptics will be selling a supplemental ionic filter that may be used to increase the NIR blocking of our Andrea U and PrecisionU filters. These ionic filters, the glass varying by main filter type, will be available to purchasers of the Andrea U and PrecisionU filters for an additional ten (10) dollars, USD. That sum does not cover the cost of manufacture, so we cannot offer the ionic filter as a separate item. The ionic filters are available in a black 52mm threaded ring. Note: the CopperU filter already is provided with a supplemental filter at no additional cost.

Thank you.

Link to comment
  • 1 month later...

I have just received one of these ionic filters for the AndreaU filter, from Reed last week.

I have not got around to using the AndreaU filter on the Panasonic G3 yet, so now was a good time to get it out & have a look.

I setup the G3 with the AndreaU filter on the lens & did a diffused CWB with a special PTFE in a filter holder.

I took a photo with the AndreaU. There is something I like about the AndreaU with the G3, the khaki & blues ?

Next I added the 'new' ionic filter in front of the AndreaU.....the results are below.

Cheers

Col

 

post-31-0-35888500-1434868181.jpg

 

Panasonic G3 ISO 200, f8 at 3 secs.

 

post-31-0-90782800-1434868229.jpg

 

Panasonic G3 ISO 200, f8 at 50 secs.

Link to comment
I just received the latest incarnation of the CopperU filter and hope to be able to test it tomorrow.
Link to comment

Col,

 

Wow, a daytime outdoor shot at 50 seconds! No wonder it is showing NIR trees. That is amazing. I must have sent the wrong glass ;) Sorry.

Please check your messages in a minute.

 

Cheers,

Reed

Link to comment

Wow, a daytime outdoor shot at 50 seconds! No wonder it is showing NIR trees. That is amazing. I must have sent the wrong glass.........

 

Just what I was thinking, looking at Andrea's AndreaU 10" exposure out-of-camera for B+W 092 Red+IR-pass in strong sunlight photo above.

Looks like an IR pass filter sneaked in place of an IR block.

Link to comment
  • 3 weeks later...

I have just received the new ionic filter for the AndreaU filter, & I am very impressed at its ability to suppress the IR leak that the AndreaU has in Sunlight.

I setup the Panasonic G3 with the AndreaU & the CopperUplus filters on the lens & did a diffused CWB with a special PTFE in a filter holder.

 

post-31-0-04820200-1436322102.jpg

 

Panasonic G3 ISO 200, f8 at 4 secs.

 

Cheers

Col

Link to comment
Well, like Andrea, I do not think that the Andrea 'U' requires additional filtering in bright sunlight. So, we will no longer be offering a supplemental filter for the Andrea 'U'. We have filled all the paid requests.
Link to comment

Please sign in to comment

You will be able to leave a comment after signing in



Sign In Now
×
×
  • Create New...