Infrared with the M system

[Larry Rexley]

I recently had an M200 converted to full-spectrum infrared (IR), and have been experimenting with it.

I've always wanted to try IR photography, back to the film days, but never got around to it. Digital mirrorless cameras are much easier for this! At IR wavelengths, lenses focus slightly differently than at normal visible light wavelengths, but since mirrorless cameras focus directly with the sensor, autofocus works fine with IR-converted mirrorless cameras.

I got a few IR filters secondhand to try out the converted camera. A full-spectrum converted camera is very versatile, but the disadvantage of full-spectrum over a 'dedicated' IR conversion is that you usually need a filter on the lens to achieve the desired 'range' of light for your image.

Here are images of the same scene using an unconverted M6ii, and the full-spectrum-converted M200 with various different bandpass IR filters. The range of images is similar to what you can find at web sites for companies that do conversions, such as lifepixel.com or kolarivision.com, where you can look at a 'matrix' of images for different types of infrared filters and different processing.

All images were shot RAW, white balance was set in post using the gray part of the cloud on the horizon.


NORMAL CAMERA: Unconverted Canon M6ii, EF-M 11-22mm, 11mm, f5.6, 1/320s, ISO 100


VISIBLE + HYDROGEN-ALPHA IR: Full-spectrum Canon M200, Spencer Camera Visible+H-alpha lens filter, EF-M 11-22, 11mm, f5.6, 1/320s, ISO 100. (This filter is similar to the bandpass range of some dedicated astro cameras such as the Canon Ra)


FULL-SPECTRUM: Full-spectrum Canon M200, No filters, EF-M 11-22, 11mm, f5.6, 1/500s, ISO 100


COLOR IR: Full-spectrum Canon M200, Spencer Camera Extreme-color 590 nm lens filter, EF-M 11-22, 11mm, f5.6, 1/160s, ISO 100


COLOR IR - RED & BLUE CHANNELS SWAPPED: COLOR IR: Full-spectrum Canon M200, Spencer Camera Extreme-color 590 nm lens filter, EF-M 11-22, 11mm, f5.6, 1/160s, ISO 100, Red and Blue color channels swapped in Adobe Lightroom (a common color IR processing technique for more 'natural-looking' colors)


STANDARD IR: Full-spectrum Canon M200, Hoya R72 720 nm lens filter, EF-M 11-22, 11mm, f5.6, 1/80s, ISO 100


STANDARD IR - RED & BLUE CHANNELS SWAPPED: Full-spectrum Canon M200, Hoya R72 720 nm lens filter, EF-M 11-22, 11mm, f5.6, 1/80s, ISO 100, Red and Blue color channels swapped in Adobe Lightroom

I don't have a 'hot mirror' lens filter, which would perform the function of the sensor filters that were removed during the full-spectrum conversion. A hot mirror filter makes the camera back into a standard 'visible-light' camera. I didn't spend the money on this filter since I have other unconverted 'M' cameras I can use for normal visible-light photography.

However, I shot the scene with a Visible + Hydrogen-alpha filter which is mostly similar to a hot-mirror filter, except that it passes some infrared light up to about 690 nm. I found that once I corrected for the slight extra 'red' in the white balance with this filter, that I really liked the rich color saturation of the image. It reminds me of the lush old Velvia film, and I'm going to try this combo a lot more. I thought the image was nicer than the 'standard' visible image.

As for the other infrared filters, I can see why swapping the red and blue channels gives a more 'pleasing' result, and why this has become somewhat of a standard technique.

From my results, it looks like the Canon EF-M 11-22 zoom lens does not have any 'hot spots' in IR light as some lenses do. However, perhaps due to being an ultra-wide lens, some of the IR images show softness in the corners of the frame at 11mm. I'd guess this improves as you zoom in, but I'll try that out. it's also possible that stopping the lens down further than f5.6 may help with this, but diffraction would also start kicking in around or past f8.

Interestingly, shooting full-spectrum, the camera seems to collect 2/3 EV more light than a normal camera... this could have some interesting implications for low-light night photography. The 'color' 590 nm IR filter required 1 EV additional exposure over standard exposure, and the deeper IR 720 nm filter required 2 EV additional exposure, both of which are not surprising as the filters are blocking more and more light.

*****

I'm going to experiment with different lenses in the coming weeks to see which M lenses work well with infrared.

I'd love to hear other folks' experiences with IR on M!

 

[Larry Rexley]

I did some comparison testing with all my lenses this morning, for their suitability for use in infrared. I went back to the same location I shot in the first post. It's a good scene with lots of contrast and detail, plus bright lighting with lots of skylight that should show up hot spot issues.

I have all the EF-M native lenses except the EF-M 28mm macro and original kit 18-55mm, so could not test those. I also have the Roki native EF-M 8mm f2.8 and 12mm f2 lenses, the Siggy 16 and 56 f1.4's, and in EF mount I have the EF-S 55-250 IS STM and Roki 135mm f2.0.

I shot the scene using the Hoya R72 (720 nm) filter which is a deeper infrared filter, almost B&W IR, which would show hots better than 'super color' IR filters that allow some visible light such as red and even orange. I shot with the lenses at maximum aperture, and in roughly 1.5 to 2-stop 'steps' also testing each lens at f11. The results are far too many images to show here, but I went through them in detail.

Here are my informal results (YMMV):

The best lenses: with virtually no hot spots and good sharpness

Canon EF-M 22mm f2 STM

Canon EF-M 32mm f1.4 STM

Canon EF-M 18-150mm f3.5-6.3 IS STM

Canon EF-M 55-200mm f4.5-6.3 IS STM

Canon EF-S 55-250mm f4-5.6 IS STM

Sigma 56mm f1.4 DC DN C

Rokinon 135mm f2.0 (EF mount)

Good lenses with slight issues

Canon EF-M 11-22mm f4-5.6 IS STM - very mild (correctable) hot spot and slight lack of corner sharpness in IR at 15mm, fine from 16mm - 21mm but hot spot appears at > f5.6 at the longer end of the zoom

Canon EF-M 15-45mm f3.5-6.3 IS STM - very mild hot spot at 15mm at > f4 but fine from 28mm - 45mm

Usable lenses with caveats

Rokinon 8mm f2.8 fisheye - no place to mount filters due to built-in lens hood; tested by holding a 67mm filter in front of lenses shows a mild (correctable) hot spot at all apertures

Rokinon 12mm f2 rectilinear ultra-wide - moderate 'color cast' contrast-reducing hot spot at all apertures, can be corrected with work

Sigma 16mm f1.4 DC DN C - no hot spot at maximum aperture but noticeable hot spot starts around f2.8 and gets progressively worse as you stop down lens

*****

I tested a Kenko 1.5x SHQ teleconverter, Kiron 2x MC7 teleconverter, and Viltrox 0.71x EF - EOS M2 speed booster with the EF-S 55-250mm IS STM (modded) lens. results were quite similar for all of them --- at all focal lengths of the zoom no hot spots when the lens was wide open aperture, but stopping down 1-2 stops started showing noticeable hot spots.

IR performance for EF-M lenses is hard to find on the internet --- please feel free to comment or ask questions about particular lenses.

If you have used these and especially other lenses on the M system, please comment and share you experiences.

Here are some images of my results:


Full-spectrum Canon M200, Hoya R72 filter held in front of Rokinon 8mm f2.8 fisheye lens at f5.6


Full-spectrum Canon M200, Hoya R72 filter, Sigma 16mm f1.4 lens at f1.4


Full-spectrum Canon M200, Hoya R72 filter, Sigma 16mm f1.4 lens at f2.8 - note the 'hot spot' compared to the f1.4 image


Full-spectrum Canon M200, Hoya R72 filter, Canon EF-M 18-150 IS STM at 18mm f3.5

 

[ken_in_nh]

Shot many a roll of BW IR, and a few rolls of EIR back in my film days. Last year (maybe 2 years ago?) I bought and had converted a M6 to IR. My interest is BW IR, not false color. I'll post some when I'm near my photo computer (which tends to get ignored in the lovely days of summer...

 

[HaroldC3]

My goto lens has been the 15-45mm. I tried the 18-150mm but the performance just isn't there for me. Although the 15-45mm isn't the greatest either, it is compact. I am using an 830nm converted M6. I have also used the EF-M 55-200mm and a Rockstar 10mm f8 lens.

I have used a 590nm M6 and 720nm M5 before finally getting an 830nm M6. I don't care for the color stuff so I don't need full spectrum or 590nm.

 

[Larry Rexley]

Images from sunset tonight at Sunset Beach in Tarpon Springs, FL.

Testing out what different subjects look like in color IR, including people...

All images shot with full spectrum M200, using a Hoya 25A filter which has a similar IR transmission curve to the 'super color infrared' filters that IR conversion companies sell (the Hoya 25A passes less orange light than a 590 nm color IR filter).

Initially processed in DxO PhotoLab 5 (sharpened and de-noised), saved to DNG format, imported into Lightroom classic for white balancing and red/blue color channel swap, then back into DxO PL5 for presentation processing.


EF-M 18-150mm, 21mm, f5, 1/100s, ISO 100


EF-M 18-150mm, 28mm, f5, 1/80s, ISO 200


EF-M 18-150mm, 150mm, f6.3, 1/320s, ISO 640


EF-M 18-150mm, 18mm, f5, 1/100s, ISO 100

DxO turns out to be great for IR processing. Due to the fact that shooting IR puts a lot of light into just one or two color channels, the noise is much higher even at ISO 100, especially when stretching the contrast of the image to more extremes and changing the white balance quite a bit.

Using DxO allows you to get rid of the noise very well at the first de-bayering step. Tried doing some of these directly in lightroom only and could not get results as good.

I do see a bit of a mild hot spot in some of these images. Guess the 18-150 is not perfect for color IR where a color cast might make a difference, but it's not so bad... the hot spot show up as a vague yellow veiling in the colors-swapped images. I had to correct it in the last image somewhat... what's left of the hot spot does not seem objectionable and I thought added a little to the image.

 

[antmanik]

Great thread, thanks for sharing.
I'm interested in how the conversion works, what changes are actually made to the camera?
Is 'color' of non-visible light filtered to pixel sensors with a new color filter array?
What is the range of wavelengths captured in "full-spectrum"?

I look forward to seeing more of these photos, thanks again.

 

[Larry Rexley]

Great thread, thanks for sharing.
I'm interested in how the conversion works, what changes are actually made to the camera?
Is 'color' of non-visible light filtered to pixel sensors with a new color filter array?
What is the range of wavelengths captured in "full-spectrum"?

I look forward to seeing more of these photos, thanks again.

Thanks. The conversion to full spectrum for the M200 that was done was just to remove both 'internal' factory filters from the sensor (the UV-IR 'cut' filter and the secondary near-IR 'cut' filter. To do this the technician takes the camera apart all the way to the sensor, removes the filters, then puts the camera back together, in a clean-room environment. This leaves the 'raw' sensor (which still has a glass cover on it so it's not really raw'). The sensor is also recalibrated so the focus distance is still correct for the EF-M lenses. Mine was perfect as all the lenses still work after my conversion.

The camera should now be sensitive to a 'full range' of wavelengths, from ultraviolet in the 250-300nm range, all the way to infrared, around the 1000 nm range. To take photographs now a lens filter must be used to cut all but the desired portion. I have filters for the ranges shown in the first post.

When you use a 'color infrared' filter --- it will cut off wavelengths shorter than about 590 nm, and mostly the red channel will record the data, but the other color channels will still record a little data. Using a deep infrared filter like an R72 which only passes around 720 nm and longer, will use almost entirely just the red channel.

With a 'hot mirror' filter or Visible + H-alpha (IR) filter all the channels get used, more like in normal photography. With any IR filter you generally use a custom white balance, with the 'visible-range' filters you get an image which would be similar to the unmodified sensor, with the 'non-visible-range' filters you get an image which looks more 'appealing' instead of being mostly red or pinkish.

The full-spectrum conversion removes the sensor's anti-aliasing filter, which softens the image slightly to avoid moire. I do notice that the modified sensor is sharper at the pixel level and shows moire, DxO PhotoLab 5 has a moire reduction feature that helps with this for images that show moire.

For cameras which have an ultrasonic 'sensor cleaning' feature a full-spectrum conversion removes the piezoelectric 'cleaner' but the M200 does not have this feature to begin with so there is no loss there.

Here are a couple more images I shot this morning


Originally 'green' pine tree on the Gulf of Mexico; Tarpon Springs FL. Full-spectrum Canon M200, Hoya 55mm Red 25A filter (around 600 nm cutoff for 'color IR'), EF-M 11-22mm, 11mm, f4, 1/250s, ISO 100, red/blue channels swapped in Adobe Lightroom after initial processing with DxO PL5


Anclote Key light from Howard Park beach 2 miles away; Tarpon Springs FL, Full-spectrum Canon M200 , 67mm Hoya R72 filter (deeper IR 720 nm cutoff), EF-S 55-250 IS STM with Kenko 1.5x SHQ teleconverter, 375mm, f8, 1/400s, ISO 1000, processed only with DxO Photolab 5 with aggressive Deep Prime de-noise, white-balanced so that trees were 'gray' and contrast greatly enhanced. (The lighthouse is just a 'dot' on the horizon in the first image)


Brackish Cove at Howard Park, Tarpon Springs. Full-spectrum Canon M200, Hoya 25A filter, Canon EF-M 15-45mm IS STM, 15mm, f4, 1/320s ISO 100, red/blue channels swapped in Lightroom, also DxO PL5


Brackish Cove at Howard Park, Tarpon Springs. Full-spectrum Canon M200, Hoya 25A filter, Canon EF-M 15-45mm IS STM, 15mm, f4, 1/320s ISO 100, red/blue channels NOT swapped, processed with DxO PL5


Post-sunset dusk sky over Klosterman Bayou, Palm Harbor FL, at low tide. Full-spectrum Canon M200, Hoya 25A filter, Canon EF-M 15-45 IS STM, 28mm, f4.5, 1/80s, ISO 100, red/blue channels swapped in post. Hard to tell this is an IR image!

 

[Larry Rexley]

My goto lens has been the 15-45mm. I tried the 18-150mm but the performance just isn't there for me. Although the 15-45mm isn't the greatest either, it is compact. I am using an 830nm converted M6. I have also used the EF-M 55-200mm and a Rockstar 10mm f8 lens.

I have used a 590nm M6 and 720nm M5 before finally getting an 830nm M6. I don't care for the color stuff so I don't need full spectrum or 590nm.

Thanks for your input. That's a striking image, very nice.

I see what you mean about the kit 15-45... I've shot more with the kit 15-45 and the 18-150 zoom, and the 18-150 does seem to have a slight hot spot, the 15-45 seems better suited. The latest image I posted with the 15-45 is quite nice, and I will use it more now.

https://www.dpreview.com/forums/post/66314468

 

[TimCastleman]

I'm just curious what the conversion cost was for your m200.

 

[Larry Rexley]

I'm just curious what the conversion cost was for your m200.

I had the conversion done by Issac Szabo after seeing positive posts on his conversions on other dpreview forums. The cost was $175 and he did it very quickly, and I'm very happy with the result.

https://www.isaacszabo.com/infrared.html

 

[TimCastleman]

Thank you, what filter did you go with? I'm thinking the 590.

 

[Larry Rexley]

Thank you, what filter did you go with? I'm thinking the 590.

The filters I use on my full spectrum converted M200 are mentioned in the various images I've posted in this thread, but to summarize:

Hoya Red 25A filter - this is very similar to the 590 nm 'super' or 'ultra' color filter the conversion vendors like Kolari Vision and LifePixel sell, but are widely available new and used. The Hoya red 25A filter passes IR very well but its visible light cutoff is slightly closer to the infrared, so it appears to be 'red' visually as opposed to the 590 nm filters which look more orange

Spencer Camera 590 nm filter - similar to the Kolari Vision and LifePixel 590 nm filters. Foliage appears a little more 'yellow' saturated (when images are red/blue channel swapped) -- I am finding I seem to prefer the 25A filter and use it more

Hoya R72 filter - shows a little color but mostly more contrasty, more B&W rendering

Spencer Camera Visible + H-alpha filter - I got this filter for astrophotography but found it works pretty well also as a normal visible light filter, like a hot mirror filter, although you have to use a custom white balance and colors are more 'rich' and Velvia-like

Just today I ordered a Tiffen Hot MIrror filter as they were deep discounted for a brief time on Amazon Prime sale ($50 - $100 off for some reason). I expect that filter to be quite similar to the Spencer camera Visible + H-alpha filter, after reading a lot of comments that it seems to let a little more IR light through than the Kolari Vision or LifePixel hot mirror filters.

 

[Larry Rexley]

I've been processing super-color infrared images using the standard technique of swapping red and blue channels in post.

I was uncomfortable using the process from a more 'purist' point of view... It seemed like a gimmick invented to make the photos 'pretty' --- for me this kind of turns the corner from 'photography' into 'art.' It didn't seem scientifically logical.

As I worked with the images --- I was increasingly puzzled by the behavior of images with different filters... Knowing a little about sensor technology and having studied physics and light, I expected that as the IR-spectrum sensor recorded light deeper into the infrared, the image data would be coming almost entirely from the red-sensitive pixels on the sensor, and that green and blue channels would record almost nothing.

That isn't what happens at all. Using a 'super color filter' such as a 590 nm filter or the Hoya 25A red filter, the images out of the camera, after white balancing, do have red skies, but blue foliage. Where's the blue coming from if this is supposedly infrared? Also the blue and green histograms of the image straight out of the camera show a LOT of data, which also seemed counterintuitive.

Foliage is very dark, almost black, at the shorter IR wavelengths such as 650 nm, and bright at the deeper (longer wavelength) IR 800 nm. Shouldn't it appear red or orange in an infrared image?

Also, a clear 'blue' sky gets darker and darker as you go deeper into the infrared (less and less 'red' light). If color infrared is truly just 'sliding' the whole spectrum into the infrared --- re-white-balancing, as it were - then shouldn't the sky should 'stay' bluish with the color shift?

I believe the answer to this is that there's a lot more going on that meets the eye (pun intended).

The answer seems to lie with the color curves for digital camera sensors, of which there are several examples on the pages below:

https://www.researchgate.net/figure...f-unfiltered-Canon-350D-sensor_fig1_229046541

https://maxmax.com/faq/camera-tech/spectral-response/canon-40d-study

Those curves are both for older Canon camera sensors, however I think the sensor technology hasn't changed much, and modern sensors likely have similar curves, for all camera brands.

The curves show that around 650 nm in the near-infrared, the green and blue sensitivity curves hit a 'minimum' where they are recording very little light. This makes sense as it's the 'red' and infrared part of the spectrum.

In the visible spectrum, the order of the peaks is blue, green, red, as our eyes expect and the sensor records. Color image processing shows the short wavelengths as more 'blue' and the longer wavelengths as more 'red' with a full spectrum in between.

However ---and this is key --- the green pixel sensitivity curve rises again in the deeper infrared, and has another peak around 750 nm. The blue pixel sensitivity curve also rises again and has a peak around 800 nm!

This means the green and blue pixels start functioning again in the deep infrared - perhaps taking on new identities, and doing things we might not expect!

From the spectral curves in the links above we can see In the infrared spectrum between 550 nm and 1000 nm the order of the peaks is red, green, blue! This is the OPPOSITE of the order of the peaks in visible light!

So, in the deep infrared the 'red' pixels are recording more of the shorter wavelengths than the green pixels, and the blue pixels are recording more of the longer wavelengths that the green pixels (once we normalize the color levels by 'white balancing')! In other words, the red pixels have become the infrared 'blue' and the blue pixels have become the 'infrared' red.

It would appear that channel swapping red and blue for infrared makes logical sense from a physics and 'perceptual' point of view. That's probably why red-blue color swapping results in a more 'natural-looking' image to our eyes.

The channel swap results in an image that's correctly color-shifted temperature-wise into the infrared part of the spectrum. The channel swap is needed because the relative sensitivity of the blue, green, and red pixels reverses in the IR spectrum between 550 and 1000 nm.

 

[Larry Rexley]

I posted in another thread a method for swapping red and blue channels directly in DxO Photolab 5.

If you are a DxO PL5 user, this can save a lot of time and trouble processing color IR images.

https://www.dpreview.com/forums/thread/4645101#forum-post-66324221

 

[Larry Rexley]

)Having heard that vintage lenses often work well for IR, I decided to take a couple of vintage lenses I recently got with a large set off Craigs list. The whole set was someone's 'junk' lenses they couldn't sell on eBay - some had loose barrels, most were pretty dirty, a few have some fungus on glass inside the lens. Projects for another day!

I cleaned up the two best ones as much as I could to try them out with my full-spectrum M200, with various filters.

I chose a Star-D 28mm f2.8 lens in Minolta MD mount, which looks like a lot of 28mm lenses such as Albinar, Promaster, etc, I've had pretty good luck with this type of lens in the past, stopped down a bit. This copy had something smudged on the front element which I mostly got rid of, a few tiny scratches, and a bit of haze inside.

I also chose a Montgomery Ward 60-300mm f4-5.6 lens in Nikon F mount. A little haze in the lens but otherwise not too bad.



I shot the Tampa city skyline from an observing platform at McKay Bay, in bright sunlight on a clear morning. Here are the results:

Star-D 28mm f2.8 vintage lens results


Unconverted Canon M6ii, f8


Full-spectrum Canon M200, no filters (full-spectrum), f8


Full-spectrum Canon M200, Spencer camera Visible + H-alpha filter, f2.8 (forgot to stop down!) - this filter is very similar to the TIffen Hot Mirror filter more widely available (amazon.com)


Full-spectrum Canon M200, Visible + H-alpha filter plus Hoya Red 25A filter, f8 - Two filters make this a very narrow-bandpass monochromatic image from ~590 nm to ~680 nm in the deep red. The V+Ha filter is similar to the TIffen Hot Mirror filter more widely available (amazon.com)


Full-spectrum M200, Spencer camera 590mm filter, f8, red/blue color channels swapped in post


Full-spectrum Canon M200, Hoya Red 25A filter, f8, red/blue color channels swapped in post


Full-spectrum Canon M200, Hoya R72 IR filter, f8, red/blue color channels swapped in post


Full-spectrum Canon M200, $6 Neewer 950 nm filter off eBay, f8. Very deep infrared

For the visible + H-alpha shot I forgot to stop down to f8, but you get to see the lens' poor performance wide open.

Interestingly the major flaw with the lens appears to be chromatic aberration, and in the narrow-bandwidth shots, especially the super-narrow 590-680 nm shot, the lens is quite sharp. There isn't much of a range of wavelengths for the CA to 'spread out' into the classic CA blur.

In post, images were processed as similarly as possible using DxO PL5 given that they are at different wavelengths. Similar sharpness settings were used, all downsampled to 2160 pixels high. The M200 has its sensor filters removed, including the anti-aliasing filter, which helps the lens sharpness, but moire shows on the square windows of the building, even with the anti-moire-setting in DxO PL5 maxed out.

I also shot directly into the sun just after sunrise, expecting the image to be the unusable 'mess' of glare & internal reflections all my Canon EF-M zooms all produce in IR with the sun in the frame. I was quite surprised to see this vintage lens do a far better job with only a few ghosts!! This is a great lens for IR photography, with little to no hot spots.


Full-spectrum Canon M200, Hoya R72 filter, shotting straight into the morning Sun at f8

Montgomery Ward 60-300mm f4-5.6 vintage lens results


Unconverted Canon M6ii, 60mm, f8


Full-spectrum Canon M200, 60mm, f8, Spencer camera 590 nm filter, red/blue color channels swapped in post


Unconverted Canon M6ii, 300mm, f8 (as marked on the lens which was probably f11)


Full-spectrum Canon M200, 300mm, f8 (as marked on lens), Spencer camera 590 nm filter, red/blue color channels swapped in post

In visible light this lens does pretty well stopped down to f8. I do not show wide-open images here, they weren't nearly as sharp - nothing to write home about.

The 60-300mm obviously does much better with IR at the short end of its range. At both the short and long ends of the zoom range, the foliage is blurred --- which isn't surprising as foliage reflects strongly in infrared light - so there was a lot of IR and some orange light from it, which focus at very different places behind the lens, blurring any foliage detail.

 

[Larry Rexley]

If you're considering modifying a camera to shoot infrared, or just want to try out IR photography without spending money to convert you camera to IR, 720 nm is a good middle-of-the-road IR conversion to get, or filter to try. It gives you a bit of contrast and you can get good B&W renderings.

A 720 nm filter is almost the only IR filter that can be used with unconverted digital cameras if you want to try IR photography without modifying your camera, but in bright sunshine you'll have long exposures on the order of 15 seconds, and must use a tripod. I bought the Hoya R72 filter and used it on my camera for a while before deciding to dedicate a camera to be converted to IR.

The longer cutoff filters (850 nm, 900 nm, 920 nm, 950 nm give you even more of a B&W high-contrast result than 720 nm, but all of them require some kind of IR-converted camera.

If you go the 720 nm camera conversion route, you can use the longer IR cutoff filters mentioned above as well, but the shorter cutoff filters like the 590 nm and 650 nm filters would have almost no effect since the 720 nm IR filter installed over the sensor for the 720 conversion already blocks the shorter wavelengths.

720 nm gives you just a little bit of 'IR color', if you bring up the saturation and vibrance of the image in post-processing. If you color swap the red and blue channels (a common infrared technique) you can get a bit of blue in the sky and the foliage will show a pale yellow.

it's worth considering a full-spectrum camera conversion, which gives you the largest world of creative visible light, IR, and even UV photography opportunities. With full spectrum you'd be shooting with a lens filter on the camera for most of the creative options. Some folks always use a filter for lens protection anyway.

if you shoot night photography or astro, you can use a full-spectrum camera without any filters, which lets in about twice as much (or 1 EV more) light as an unconverted camera. it almost steps you up to the light-gathering equivalent of a full-frame, unconverted camera.

A post in this thread (see link below) shows a full range of creative options for a full-spectrum camera (and the options for other conversions as allowed by the conversion filter's cutoff):

https://www.dpreview.com/forums/post/66344116

Kolari and LifePixel, two popular conversion services have great sections on their web sites explaining the various conversion options. I had my conversion done by Isaac Szabo, at $175 for the M200 he was $100 less than the standard conversion services, and he did a great job.

If you go the full-spectrum conversion route and only have the EF-M 15-45mm kit lens, Amazon has an unbelievable deal on the Tiffen Hot MIrror filter for the 49mm size only (if you're in the USA) - this is the size needed for the EF-M 15-45 kit lens... it's only $10.65.

I've thought that this might be a price mistake as most other versions of the filter around that size are over $50! The 77mm filter is also fire-sale priced (showing for me at $60.81 today - it fluctuates wildly in that size) -- I bought it a few weeks ago when it was less than $50! The next day it jumped to $175 which is the 'normal' price for that size.

https://www.amazon.com/Tiffen-49mm-Hot-Mirror-Filter/dp/B00004ZCL4?th=1

The Tiffen hot mirror filter turns your camera back into a visible light camera for normal photos. Its spectral curve is not exactly the same as the factory hot mirror filters removed from the camera during a full-spectrum conversion, but you can do a custom white balance either in camera or in post-processing (if shooting raw) and get a normal-image with vibrant green foliage instead of reddish-tinted foliage, and slightly more vibrant reds.

A good filter for color IR is the Hoya 25A Red filter, which cuts off around 600 nm and passes IR light very well. It's a high-quality, common filter that's less expensive than the dedicated Kolari, LIfePixel, or Spencer camera multi-coated 590 nm filters which are slightly more orange but give similar results. You can often find the 25A filter at good prices secondhand.

A Hoya or Tiffen deeper Red #29 filter also is great for IR, with a higher cutoff around 620 nm, but they're less common and more expensive than the 25A filter.

The best-known filter for IR is the Hoya R72 filter, with a 720 nm cutoff. It's a very high-quality filter you can often find secondhard for a good price, since they're fairly common.

I just bought a brand-new Neewer 950 nm ultra-contrast monochromatic IR filter in 67mm size for only $6 off ebay, which included shipping! I thought, at that price I didn't have much to lose, but it turns out to be a great filter with good sharpness and high contrast! 950 nm cutoff requires longer exposures than the other IR filters since it cuts off so much of the IR spectrum, but in bright sunshine or on a tripod it's still usable with a converted camera with shutter speeds of 1/30 second or so at higher ISOs like 400-800, and gives good results.

For me the full-spectrum camera is the 'safest' route to go. If it turns out that IR is not your cup of tea, you just use the hot mirror filter and you can use your camera as a normal visible-light camera again.

I often take the M200 along in my camera bag as the second or third body, with the EF-M 15-45mm kit lens and the 49mm Tiffen Hot mirror filter on it. I can swap and use that lens with that filter on it on my unconverted M6ii cameras, and it gives normal images on those cameras. Then if I want to shoot IR, I can remove the hot mirror filter and put on the Hoya Red 25A or R72 for color or B&W IR and shoot with the full-spectrum M200.

Generally filters in 49mm size are cheaper than other 'standard sizes' such as 55mm or 67mm so if you're sticking with just the kit lens (which is a very good infrared lens for the EF-M system) you can do it inexpensively. If you think you may get other lenses for the system, you'd be safer going with a larger filter size such as 55 mm or 67mm.

The most common filter size I've bought for IR is 67mm since I can use that size with all of my lenses (some with step-up rings). The Rokinon 12mm f2.8 and Sigma 16mm f1.4 lens use 67mm filters. I've also bought a 77mm to 67mm step-down ring so I can use all the 67mm filters on the Ef-mount Rokinon 135mm f2 lens for astrophotography. 67mm filters do not vignette on that lens.

I also have a few 55 mm filters like the Hoya Red 25A filter which I found for cheap on eBay in that size, and 55mm will work on all the EF-M lenses (some with step up rings). I have permanent 43 - 55mm step-up rings on my EF-M 22mm f2 lens and 32mm f1.4 lenses, and don't understand why Canon didn't make 55mm the filter size for ALL EF-M lenses as it works perfectly for the EF-M standard 'lens barrel' size. I have a 52-55mm step up ring for the EF-M 55-200 lens, but unfortunately this ring blocks the lens hood from being mounted on that lens.

I have several 55-67mm step-up rings so I can use the 67mm filters on the EF-M lenses as well, and a 58-67mm step up ring so I can use 67mm filters on the EF-S 55-250 IS STM lens. I bought lens caps off Amazon.com in packs of 4 in both 55mm size and 67mm size for less than $10 per pack. I sometimes leave the step-up rings on lenses for shooting sessions, swapping lenses and filters without fiddling as much with step-up rings.

The lens hood for the 32mm f1.4 fits nicely over the 55mm step up ring. I have a 49mm to 55mm step up ring I often use with the 15-45 kit lens, and a 52mm to 55mm step-up ring for the EF-M 55-200 lens. The Canon EF-M 18-150 and Sigma 56mm f1.4 lenses are already 55mm filter size.

 

[Larry Rexley]

Finally I had a beautiful clear night, even in the Florida summer, to try out my converted Canon M200 for one of things I got it for: astrophotography! I had to go out at 2 AM... it had rained and everything was soaking wet, but it was worth it!

I used the full-spectrum-converted M200 on a SkyGuider Pro star tracker and tried different lenses and filter combinations.

After a few trials I settled on shooting with the Spencer Camera Visible + hydrogen-alpha which allows the visible spectrum and near-IR but blocks around 700 nm and longer. I stacked that with the K+F Concept Clear-Natural light pollution filter (from Amazon) which appears to be quite effective in handling the light pollution from my shooting location in Tarpon Springs on the Gulf of Mexico. All shots below with the M200 were taken using both those filters stacked.

The summer Milky way, with the manual Rokinon 12mm f2 lens in native EF-M mount:


Ultra-wide summer Milky way from Cygnus to Sagittarius. Full-spectrum Canon M200, K&F LP filter, Spencer camera Visible + Ha filter, Rokinon 12mm f2 lens, f2, 30 seconds, ISO 100 (!)


Rich Milky Way star field in Cygnus, and the North American Nebula upper right corner. Full-spectrum Canon M200, K&F LP filter, Spencer camera Visible + Ha filter, Rokinon 12mm f2 lens, f2, 30 seconds, ISO 100, cropped center of frame

The Milky Way images show far more 'richness' and a lot more 'red' in Visible + Hydrogen-alpha light than they do in plain visible light. The dark 'coal sack' nebulas in Cygnus and the dark lanes through the galactic core (to the lower left in the first image) don't appear so dark, they must be giving off a good deal of IR light.

Images with the EF manual Rokinon 135mm f2 & Canon EF to EOS M adapter:


Trifid and Lagoon nebulas, and a globular cluster. Full-spectrum Canon M200, K&F LP filter, Spencer Visible + Ha filter, Rokinon 135mm f2 lens, f2, 30 seconds, ISO 400, cropped at pixel res


2 globular star clusters in Sagittarius. Full-spectrum Canon M200, K&F LP filter, Spencer Visible + Ha filter, Rokinon 135mm f2 lens, f2, 30 seconds, ISO 400, cropped at pixel res


The Dumbbell Nebula in Cygnus. Full-spectrum Canon M200, K&F LP filter, Spencer Visible + Ha filter, Rokinon 135mm f2 lens, f2, 20 seconds, ISO 800, cropped at pixel res


The Ring Nebula in Lyra. Full-spectrum Canon M200, K&F LP filter, Spencer Visible + Ha filter, Rokinon 135mm f2 lens, f2, 30 seconds, ISO 400, cropped at pixel res

Another image of the Ring Nebula with an unconverted Canon M6ii, using the EF-S 55-250mm IS STM lens, plus my home-made 2x MC7 teleconverter with vintage Kiron optics, with no filters on the lens. 20 separate 1-minute-long exposures on the star tracker at 500mm/f11/ISO 1600, stacked in Deep Sky Stacker with flat and bias frames to make a single 20-minute exposure. Sharpened in Registax with wavelets, final processing in DxO Photolab 5:


The Ring Nebula in Lyra. Unconverted Canon M6ii, Canon EF-S 55-250 IS STM, vintage Kiron 2x MC7 teleconverter, 500mm, f11, ISO 1600, 20 exposures of 1 minute each stacked with Deep Sky Stacker

Notice the Ring nebula appears an aqua color in the unconverted camera, this is from the green emission lines given off by the gas in the central nebula. In the other visible + hydrogen-alpha light image, the outer parts of the nebula appear orange-yellow from the H-alpha emissions that the unconverted M6ii can't see almost at all.

 

[Alan WF]

I am impressed by your photos! The comparison of the two images of M57 nicely shows the need to include H-alpha in nebular photography.

The dark 'coal sack' nebulas in Cygnus and the dark lanes through the galactic core (to the lower left in the first image) don't appear so dark, they must be giving off a good deal of IR light.

More precisely, the interstellar dust that blocks the background light and forms these apparently dark features becomes more transparent at longer wavelengths.

Regards,

Alan

 

[barbara j]

I recently had an M200 converted to full-spectrum infrared (IR), and have been experimenting with it.

All images were shot RAW, white balance was set in post using the gray part of the cloud on the horizon.

I'd love to hear other folks' experiences with IR on M!

I'm shooting IR on a different camera but on your M, try setting the white balance on location, using green instead of a cloud. Anything green works, I have a green plastic card, an old library card, that works well when there isn't any grass. This gives a much cleaner look to your image.

 

[Larry Rexley]

I recently had an M200 converted to full-spectrum infrared (IR), and have been experimenting with it.

All images were shot RAW, white balance was set in post using the gray part of the cloud on the horizon.

I'd love to hear other folks' experiences with IR on M!

I'm shooting IR on a different camera but on your M, try setting the white balance on location, using green instead of a cloud. Anything green works, I have a green plastic card, an old library card, that works well when there isn't any grass. This gives a much cleaner look to your image.

Thanks for the suggestion, I'll try that. I have experimented with setting white balance before shooting at a scene... seems some folks recommend setting it on something neutral like a cloud or concrete, some who want their foliage to go 'white' suggest doing it on foliage or something green. I agree that setting it in camera during the shoot gets you in the ballpark and makes post-processing easier.