Article: "Monochrome Camera Conversion: Effect on Sensitivity for Multispectral Imaging (Ultraviolet, Visible, and Infrared)"

[JMC]

I've just had a paper come out in the MDPI Journal of Imaging which looks at the effects of monochrome conversion on some of my cameras, with an emphasis on trying to understand how the sensor type influences sensitivity. The article is available open access from here - https://www.mdpi.com/2313-433X/8/3/54

 

Abstract - Conversion of standard cameras to enable them to capture images in the ultraviolet (UV) and infrared (IR) spectral regions has applications ranging from purely artistic to science and research. Taking the modification of the camera a step further and removing the color filter array (CFA) results in the formation of a monochrome camera. The spectral sensitivities of a range of cameras with different sensors which were converted to monochrome were measured and compared with standard multispectral camera conversions, with an emphasis on their behavior from the UV through to the IR regions.

 

This article has been a long time in the works. I originally wrote it about 2 years ago. I then submitted it to a journal who sat on it for 18 months (the journals editor left and it took them ages to find a new one). After that they contacted me and said 'yeah, not really suitable for us', even though it was a follow up to a previous article I had published with them. I decided to go Open Access as my experience has been that the reviews tend to be done in a much more timely manner. This time it was about a month and a half, but I had to pay for the privilege. However this does mean that it isn't stuck behind a paywall and anyone can access it.

 

It has just come out, and hopefully there aren't too many typos in there (I get a bit 'print' blind when proofreading these things).

[colinbm]

Congratulations Jonathan.

[dabateman]

Great work Jonathan. Glad its out.

I was wondering if you could partner with Dan from MaxMax and ask for a stack of coverglass from the monochrome camera conversions that he does.

Then you could show us the evolution and which models block UVB and below at the point of the coverglass. 

 

[Stefano]

24 minutes ago, dabateman said:

Great work Jonathan. Glad its out.

I was wondering if you could partner with Dan from MaxMax and ask for a stack of coverglass from the monochrome camera conversions that he does.

Then you could show us the evolution and which models block UVB and below at the point of the coverglass. 

 

That would be very interesting. I would like to know the transmission of the coverglass of my Canon EOS M for example.

[JMC]

Cheers all. 

David, yes I've mentioned that to Dan before, but nothing ever came of it unfortunately.

[Andrea B.]

Jonathan, thank you for sharing all your work with us here at UVP. We have all enjoyed your investigations.

 

 

(I get a bit 'print' blind when proofreading these things).

This happens to me also. It is difficult to edit oneself!

[JMC]

On 2/26/2022 at 4:58 PM, Andrea B. said:

Jonathan, thank you for sharing all your work with us here at UVP. We have all enjoyed your investigations.

 

 

(I get a bit 'print' blind when proofreading these things).

This happens to me also. It is difficult to edit oneself!

Cheers Andrea.

[photoni]

@JMCJonathan, I downloaded your review, as far as I can understand I really like it, and I find the difference between the D800 and D850 amazing.
I think that the monochrome modification is very useful to have the maximum sensitivity; removing the bayer filters allows you to use the narrow filters in a more precise way ... since the three channels are much more coherent and not with strange curves (for the normal logic of standard digital photography)

Antonio

[JMC]

Thank Antonio.

[nfoto]

An enjoyable read. I have added the paper to my digital reference archive.

[JMC]

Cheers Birna.

[KhanhDam]

wow what an great article! Thank you for writing it. I made a list of all current BSI Mirrorless sensor cameras below (excluding the top end very expensive cameras like Z9, A1,R3, etc) with approximate used market price in USD. Since the warranty is void I am fine with a used camera body. 
I want to get one converted to remove Bayer filter so it is 100% B&W mono conversion. Looks like I will get about 3 stops more light vs a full spectrum conversion. I already have a full spectrum converted camera that I use for IR (very bad uv sensitivity, I think cover glass issue) , but I want to try UV photography now. I want a mirrorless camera because the focusing should be better and easier to adapt wierd lenses to it. 

 

Used for pretty much artistic photography landscapes and flowers will be 90% of what I photograph. Maybe some portraits to show freckles and skin damage. 

 

I am leaning towards the Nikon Z50 since it is the least expensive. Would there be any benefit to getting the Z6 which has image stabilization? Seems like slow shutter speeds  are common and would benefit from IBIS.  I realize the z6 has an IR shutter led, but seems like that only effects images that are iso 3200+ and very long shutter speeds. Maybe a mono conversion negates that issue?

 

also curious if there are any autofocus solutions ? Techart PRO Leica M – Sony E Autofocus Adapter ver II (LM-EA9) – TECHART Adapters
Is there enough UV light so that the techart AF adapter could be used? If so that would make the  sony A7rII appealing.  Wish they had a nikon version, as I'm more familiar with Nikon cameras. 

 

I have a venus UV filter, several el Nikkors and of course a 35mm 3.5 lens. Will probabaly get a cheap 21mm 3.8. I wish there were faster lenses that are uv capable. 1.8 would be nice. 

 

thanks for any help! First post so go easy on this newbie to UV.
 

List of Mirrorless cameras with BSI sensors. All prices are approximate for used gear in USD$.

nikon z30 20.9 $690 (new price) 
Nikon z50 20.9mp $650 
nikkon zfc 20.9mp $800
nikon z6 20.5 $1150
nikon z7 45.7mp $1900

sony a7r2 42.4mp $1050
sony a7s3 12mp   $3400
sony a7 iv 33mp $2200
sony a7 III 24.2mp  $1420
sony a7C 24.2mp   $1560

 

Fuji X-t4 26.1mp $1500
Fujifilm X-S10 26.1MP $950
Fuji X-T3 26mp $900

 

sigma Fp 24mp $1400

 

Panasonic LUMIX S5 24.2MP $1400

[Andy Perrin]

I would not get any camera converted that has a shutter IR LED. You WILL want long exposure times at some point. It is the nature of working with UV that we deal with very dim light. 

 

Definitely get something with IBIS if you can. It's made my life so much better with the A7III. 

 

Also, while removing the Bayer will get you faster exposures, it's nice to have color (even if it's a bit limited) especially as a beginner because detecting IR contamination is easier. Also, given the enormous expense of converting to mono, buying a $650 camera seems ridiculous. If you are going to do that, may as well get a Raspberry Pi mono conversion.

[KhanhDam]

 

watch around 16 minutes in where the laser strips off the bayer layer, I think he also sells debayered pi cams. 

I am not familiar with sony cameras. the 2 least expensive ones the A7III one you have and the A7R2 with almost double the resolution would be contenders. Why did you choose the more expensive lower resolution A7III over the A7r2?

 

[Andy Perrin]

Because the low light performance of the A7III is the best of all Sony’s cameras, even the A7IV.  Lower resolution means bigger pixels with more light per pixel, and in actual tests the A7III still wins. But honestly it’s a very near thing — the A7R series are very good, even in low light. But I also started adding up the storage costs for higher resolution photos and that also made me want to stick with 24MPix. 

[photoni]

I have a Sony A7 fs (from 150 €) with the sensor scratched ...

I would still get an A7, (on many reviews it is considered the best sensor)
One question: I've always wondered why nobody has a monochrome Leica

(apart from the cost that I couldn't afford) it should be the easiest to convert and the most sensitive,

since it keeps the micro lenses in front of the sensor, not like the machines born with the Bayer.

[KhanhDam]

thanks but A7 does not have a BSI sensor so I wouldn't be interested. I think cost is the main factor with the Leica. One can get a modified camera for less.

 

[ulf]

Another good thing beginning with the Sony A7 III is that IBIS is working with any lens.

You can tell the camera the FL manually and do not need a lens with electronic communication.

It also uses a bigger battery that gives more on time.

 

With some patience you'll get an used A7 II to a better price than  $1420

 

So far I have not found any other camera that for my purposes will be better.

 

My conversion is a full spectrum type where all filters and dust shaker are removed and not replaced with any window.

The sensor's cover window is not replaced, so I have the wavelength limitation from that surface coating.

My type of UV photography is not into deep UV-B, so that is not a problem.

[dabateman]

The Nikons Z-mount cameras seem to still have a banding problem,  that Birna showed with her Z6 and gets sometimes with the Z5. 

But there are the Megadap AF adapter,  the techart Af adapter and a pronto AF adapters that are Leica M-mount and can use a second adapter to most any other lens to mount to the Nikon Z-mount. For Sony E-mount you have your choice of the Pronto or the techart. For Fuji X-mount only the Pronto is available.  These are the only mounts with AF adapters. 

However,  if you send your camera to be monochrome converted, then you are stripped off the color filter area and the phase dection pixels.  So none of the AF adapters will work after monochrome conversion,  as they require on sensor phase detection pixels. After monochrome conversion you are left with contrast AF only.

 

[KhanhDam]

1 hour ago, ulf said:

My conversion is a full spectrum type where all filters and dust shaker are removed and not replaced with any window.

The sensor's cover window is not replaced, so I have the wavelength limitation from that surface coating.

what about all the super thin micro gold wires that go from the sensor to the outer edges of the frame? how are those protected? Or are you just extremely careful to never ever expose those?

 

thanks for the info on IBIS. I assumed it worked with any lens, now I know to look out for that limitation.

[KhanhDam]

1 hour ago, dabateman said:

However,  if you send your camera to be monochrome converted, then you are stripped off the color filter area and the phase dection pixels.  So none of the AF adapters will work after monochrome conversion,  as they require on sensor phase detection pixels. After monochrome conversion you are left with contrast AF only.

Geeze i didn't realize that. And I thought my idea was clever, lol. Gotta learn to be humble I guess. 

1 hour ago, dabateman said:

 

[ulf]

28 minutes ago, KhanhDam said:

what about all the super thin micro gold wires that go from the sensor to the outer edges of the frame? how are those protected? Or are you just extremely careful to never ever expose those?

 

thanks for the info on IBIS. I assumed it worked with any lens, now I know to look out for that limitation.

The gold wires are not exposed.

AFAIK backside illuminated sensors like the ones Sony are using have their connections on the rear side.

Furthermore removing the BG-filter and dust shaker filter would not expose the wires anyhow.

 

There is a window cemented on the sensor itself behind those filters.

It is only when you want to monochrome convert a sensor by removing micro-lenses and bayer matrix that the sensor cover window has to be removed.

 

JMC had the window replaced on one camera without doing the monochrome conversion, to improve the UV-reach.

That is a rather rare modification, that might be a bit expensive, compared to just do a normal full spectrum conversion.

 

The advantage with the filter removal with no replacement glass-method is that some quality lenses designed for thin analog film will perform better, especially in the corners, as a thicker glass behind the lens was never expected when the lens formula was designed.

The camera must be realigned to compensate for the changed optical path length, when missing the filters.

 

The disadvantage is that any dust on the window will be closer to the sensor, than with filters or replacement glass present.

The dust will be more visible then, but there is no possibility of any trapped dust inside the replacement glass.

[KhanhDam]

22 minutes ago, ulf said:

JMC had the window replaced on one camera without doing the monochrome conversion, to improve the UV-reach.

That is a rather rare modification, that might be a bit expensive, compared to just do a normal full spectrum conversion.

I've removed that window once. Using heat is best way to get cover glass off. Use a thermal tape like kapton over the cover glass so it if does crack it wont' fall on sensor.
Does't take long but risk of damaging the sensor go way up so companies have to charge more incase they mess up. 

[dabateman]

1 hour ago, KhanhDam said:

Geeze i didn't realize that. And I thought my idea was clever, lol. Gotta learn to be humble I guess. 

Yes I like auto focus, even more so on cameras with in-camera focus bracketing.  I may have looked at every possible combination. 

There are AF adapters without glass for Leica M-mount to Z-mount, E-mount and X-mount. But these adapters require on sensor phase detection.  Most of the cameras I like don't have that or Sony doesn't have in camera focus bracketing.

 

There is the Nikon 1.6x teleconverter which will AF manual lenses, but it has glass inside to do the focusing.  I have this adapter and its great, but doesn't allow a lot of UV through it. Its F-mount to F-mount. 

 

So knowing that most cameras now have an EF-mount AF adapter,  I looked at Canon lenses that use a simple forward and back focus mechanism.  The 2 best candidate lenses are the 50mm f1.8 ii lens and the 50mm f2.8 macro lens.  I bought cheap fungus versions, actually the 50mm macro I could have saved as was only dust. But I drilled through the optical groups, as they are glued into these lenses. The Canon 50mm macro isn't as good as its only focus by wire,  you can't manually adjust the focus. The 50mm f1.8 ii lens you can actually manually focus, the newer 50mm stm you can't. 

I then glued in a 75mm fused silica element into the 50mm ii lens.  I have another with rings, as well as the 50mm macro with rings. This allows me to add the Nikkor 80mm or 105mm EL lenses for auto focus or other fused silica assembies. 

So with a custom EF mount lens I now have auto focus for nearly any camera.

 

 

[KhanhDam]

12 minutes ago, dabateman said:

I then glued in a 75mm fused silica element into the 50mm ii lens.  I have another with rings, as well as the 50mm macro with rings. This allows me to add the Nikkor 80mm or 105mm EL lenses for auto focus or other fused silica assembies. 

So with a custom EF mount lens I now have auto focus for nearly any camera.

 

sounds like a cool set up do you have any photos of this adapted lens technique?