Very quick question - IR 1um+ filter recommendation

[Lou Jost]

Regarding the original topic, this might be a good place to consider the amount of incident daytime IR radiation on the earth, as a function of wavelength. Infrared emission from the sun decreases steadily with increasing wavelength, but there is also a very strong absorption (by atmospheric water vapor) around 930-970nm. In addition, silicon detectors become transparent at about 1100nm. So beyond about 930nm you will have very limited light, and an 850nm filter will probably act somewhat  like a bandpass filter (850-930nm).

[Andy Perrin]

Regarding the atmosphere, that is pretty far off the truth. There is quite a lot of IR, especially in the shortwave range. There are a few holes but the NIR and SWIR situation is not remotely comparable to UV-C. I can easily image (using my germanium camera) outdoors in sunlight. 
 

The silicon sensor is a different issue — it does indeed go down pretty fast as you go past 950nm or so, although I can take pictures out to 1200nm (as Stefano showed me when we attached a solar panel to batteries and visualized the glow). 

[Lou Jost]

3 minutes ago, Andy Perrin said:

This is pretty far off the truth. There is quite a lot of IR, especially in the shortwave range. There are a few holes but the NIR and SWIR situation is not remotely comparable to UV-C. I can easily image (using my germanium camera) outdoors in sunlight. 
 

The sensor is a different issue — it does indeed go down pretty fast as you go past 950nm or so, although I can take pictures out to 1200nm (as Stefano showed me when we attached a solar panel to batteries and visualized the glow). 

 

[Lou Jost]

I didn't say it was comparable to UV-C. But there really is an important absorption window there as I stated. You just might not have noticed it.

https://commons.wikimedia.org/wiki/File:Solar_Spectrum.png

 

Couple that with the sensor issue (which you agree with), and you get the situation I described.

 

[Andy Perrin]

I see that picture but it does not match my observations very well? I have a lot of bandpass filters but I do have a gap in my collection at 1300nm, which could be where the “hole” is. I don’t think that hole will affect silicon sensors much, as the gain is truly zero there. 
 

The other hole near the 900’s is not a factor. As said above, I specifically hunted down a bandpass to take advantage of the water absorption peak at 976nm and there was plenty of sun for photography purposes. 
 

I think what is misleading is that the graph is a linear graph and losing a stop or two is just not very noticeable usually. It’s like the same amount of darkening as if the sun went behind a cloud. Plotting this on a log chart would be more meaningful from a photographer’s point of view. 

[Lou Jost]

1 minute ago, Andy Perrin said:

I see that picture but it does not match my observations very well? I have a lot of bandpass filters but I do have a gap in my collection at 1300nm, which could be where the “hole” is. I don’t think that hole will affect silicon sensors much. 

No, the hole is where I said it was. There is actually a well-known window, not a hole, at 1300nm.

 

Here is a very high resolution graph of atmospheric absorption (this must be convoluted with the solar emission graph to get intensity):

https://link.springer.com/article/10.12942/lrsp-2014-2

 

[Andy Perrin]

I am not contesting the location of the hole, I’m contesting the SIGNIFICANCE of the 900’s hole for photography. You are not giving enough time for me to edit my posts before replying I think. 

[Lou Jost]

In the context of long-pass filter behavior, I think the hole I mentioned is significant. Sure, you can take photos in that hole. But when using a long-pass filter whose range includes a lot of wavelengths outside that hole, the image will be dominated by that set of wavelengths outside the hole.

[ulf]

6 minutes ago, Lou Jost said:

In the context of long-pass filter behavior, I think the hole I mentioned is significant. Sure, you can take photos in that hole. But when using a long-pass filter whose range includes a lot of wavelengths outside that hole, the image will be dominated by that set of wavelengths outside the hole.

I think this is the filter Andy used:

https://www.thorlabs.com/thorproduct.cfm?partnumber=FB980-10

That will not include a lot of wavelengths outside that hole.

https://www.ultravioletphotography.com/content/index.php?/topic/3505-dark-water-and-the-nir-absorption-peak/#comment-30339

 

This topic has been discussed before here.

[Andy Perrin]

The point I wanted to make is that the 900’s hole is not as deep as Lou is thinking. It’s 1-2 stops perhaps. The sensor sensitivity it the dominant problem. 
 

ulf, agreed, in the bandpass situation I do not get much on either side, especially with the Thorlabs filter. Lou wanted to discuss long pass behavior though. 
 

The longpass has a lot more light to work with in general though and the hole is not that deep. 

[Lou Jost]

6 minutes ago, ulf said:

I think this is the filter Andy used:

https://www.thorlabs.com/thorproduct.cfm?partnumber=FB980-10

That will not include a lot of wavelengths outside that hole.

https://www.ultravioletphotography.com/content/index.php?/topic/3505-dark-water-and-the-nir-absorption-peak/#comment-30339

 

This topic has been discussed before here.

Sure, that's why I said "In the context of long-pass filters..."   What I said definitely does not apply to Andy's narrow bandpass filters.

[Lou Jost]

5 minutes ago, Andy Perrin said:

The point I wanted to make is that the 900’s hole is not as deep as Lou is thinking. It’s 1-2 stops perhaps. The sensor sensitivity it the dominant problem. 

 I agree with that.

[ulf]

5 minutes ago, Lou Jost said:

Sure, that's why I said "In the context of long-pass filters..."   What I said definitely does not apply to Andy's narrow bandpass filters.

Sorry I missed that.

May I blame my minor dyslexia?

 

I just pulled out my Noflexar 35mm with the stack of the very cheap BP 980nm that Fandyous found plus a 1mm 850nm and took a shot at my bathroom sink with some water.

https://www.ebay.com/sch/i.html?_from=R40&_trksid=p2334524.m570.l1313&_nkw=980nm+bandpass&_sacat=0&LH_TitleDesc=0&_odkw=970nm+bandpass&_osacat=0

Not sunshine, but illuminated with two 15W halogen bulbs. ISO1600 F/3.5 and 1/30s för a proper exposure.

 

The summer before last I also did some outdoor shooting too of fountains, without any problem handholding.

 

That tiny BP-filter is really interesting when stacked with a 850nm longpass filter. You just have to find a lens with a small enough front- or rear-element that is reasonably good for IR.