BG25 filter garden shots

[JMC]

Recently got a couple of 75mm filters from ebay that were advertised as BG25 2mm, and BG18 2mm. Longest delivery time ever - I think it was over 3 months from buying them to having them arrive. Given they were cheap, I thought I'd give them a go as they fit in 77mm filter rings.

 

Here's the transmission of the one which was advertised as BG25 2mm.

 

The spectra is sort of what I'd expect for 2mm BG25, although overall transmission is slightly lower than it should be for 2mm, especially in the IR region.

 

Some photos from my garden using the BG25. Images taken on a multispectral EOS 5DSR, using a 17-40mm lens. White balanced in Darktable and reduced in size for sharing. No change in the hue of the image during processing, although contrast has been boosted slightly.

 

 

 

The white flowers are Ox-eye daisies (white with yellow centers in the visible spectrum).

[colinbm]

Wow nice the last is popping good.

[Stefano]

Very strong reds, especially in the last one. Nice photos.

[JMC]

Colin, Stefano, interestingly the last one looks more red and less orange to me too. It was shot when the sun went behind a cloud, while the first two were in more direct sunlight. I did separate white balances for each one, which were done with the same lighting that the image was taken in.

[JMC]

What a difference a white balance makes. If I whitebalance the first two images with the whitebalance target shot for the final one (the one which was done with the sun behind a cloud) I get stronger reds, and the sky takes on a more blue/purple hue.

 

[colinbm]

I like the first series best.

[JMC]

So do I Colin, just surprised how much of a difference it made.

[Stefano]

Do you think this has to do with the difference between direct sunlight and sunlight filtered by a cloud?

[JMC]

I guess so Stefano. The ratio of UV/blue to IR will be different between direct sunlight and sunlight through a cloud, so I presume this comes through as a slight colour shift in the white balanced image.

[Andy Perrin]

I’m guessing the IR spikes up after 800nm in the portion of the spectrum off your graph. Or else how to account for those red plants?

[JMC]

Your wish is my command Andy. Will check above 800nm and update tomorrow.

[dabateman]

I’m guessing the IR spikes up after 800nm in the portion of the spectrum off your graph. Or else how to account for those red plants?

 

It might not. Most likely the same flat look.

I would think you have a small bump in the 600s that we can't see in your plot.

 

I think it was my Chinese BG39 2mm thick and 77mm diameter with my 2mm thick, 77mm ZWB3 filter that had a small bump at 580nm, that resulted in orange to red centered dandelions.

 

If you have a very strong Halogen bulb, you can use that to find these leaks with your spectrometer. Our cameras are very sensitive at the 550 to 610 range and there are lots of sunlight photons there to take over the image.

[Andy Perrin]

Well, we shall find out tomorrow!

[dabateman]

Ok I changed my thinking. I still think its constant into IR without a bump. But also has no leaks.

Remembering this post with Jonathan's cfa response for this camera:

https://www.ultravioletphotography.com/content/index.php/topic/2580-build-thread-at-home-measurement-of-camera-uv-spectral-response/page__view__findpost__p__20116

 

That 690nm to 720nm is most likely all the red we see. Also explain why more red with cloudy filtered sunlight than direct sunlight. The color channels should be more even above 850nm.

 

That link is for a normal camera, so you have to think about removing the IR blocking filter on sensor in your mind.

[Fandyus]

Very nice, I have the Chinese equivalent to this (QB29) and it makes everything yellow instead.

[Andy Perrin]

Which tells us that it’s not even close to the same spectrum. That’s gorgeous though.

[colinbm]

I need that filter....

[JMC]

Update to the spectra. I ran the BG25 (and BG18 which was the other one I got at the same time) out to 1100nm and combined the results from the two spectrometers.

 

The BG25 does increase slightly as you go further into the IR, but it isn't a quick increase. David, there is no discernible bump in the 600nm region.

 

I mainly got these two because combined I thought they might make a good 'collodion' emulation filter.

 

These two were from the garden yesterday afternoon. Slightly cloudy by this stage (not direct sunlight). Multispectral EOS 5DSR with 17-40mm f4 lens. Captured as jpeg in the camera, then desaturated in Photoshop and the curve tweaked slightly before resizing.

 

[colinbm]

I like them Jonathan

I am experimenting with this too.

Can you post a colour photo of one of these that has been white balanced please ?

[JMC]

Hi Colin, I didn't take any white balance photos for the BG25+18 combination, as I was just going to make it into black and white anyway.

 

Happy to get a photo today of the garden with BG25+18 and white balance it for you if that would be helpful for you.

[colinbm]

I did some the other day & called it Digital Daguerreotype, one in colour & one converted to mono.

I just wanted to see the colours you got with this combo.

[JMC]

No problem, Colin. Here's three shots from the garden.

 

First, EOS 5DSR multispectral, 17-40mm f4 lens (at 19mm and f8), BG25 and BG18. Whitebalanced in Darktable.

 

Second, EOS 5DSR multispectral, 17-40mm f4 lens (at 19mm and f8), BG25 and BG18. JPEG from camera then desaturated in Photoshop.

 

Finally, camera phone image in colour.

[colinbm]

Thanks Jonathan

You got more of a mix of cool colours then I got ?

[JMC]

Not sure why that would be Colin. The wavelength range being captured here is relatively small, so there wont be a huge range of colours in the image. Perhaps white balance is pushing the range of colours as far as it can. Personally, I'd only use this filter for black and white anyway, so the range of colours isn't something I am worried about.

[colinbm]

Yes I appreciate that these are only intended for B&W, I was just curious, as I am always, to see how we compared.