RF 16mm f/2.8

[Ming]

RF 16/2.8 transmits some UV.  Three sets of test images are shown below to demonstrate its capabilities, using El-Nikkor 80mm f/5.6 old metal version as the reference.  Set 1 is common daisy, set 2 a cactus, and set 3 Black-eyed Susan.  The cactus is Opuntia humifusa, commonly known as eastern prickly pear and native to the eastern US.  It is endangered in Massachusetts (https://www.mass.gov/doc/eastern-prickly-pear/download).  Set 3 includes EF 28mm f/2.8 and EF 40mm f/2.8 for comparison.

 

The filter diameter of RF 16/2.8 is 43mm.  If ZWB2+TSN575 of 52mm diameter is attached, there is serious vignetting.  I have remounted a Baader-U in a spare 52mm ring -- the original tenant, a ZWB2, has striations.  Unfortunately, this one ring on top of a step-up ring still causes vignetting that prevents proper white balance in camera.  So I attach a Kolari-U of 77mm diameter on RF 16/2.8.  The same filter is used on the other lenses, although all of them can take Baader-U.  Pictures are taken with a modified R5, white-balanced off gray teflon tape, RAW to JPG by Canon DPP's faithful mode, with no cropping and no post-processing.

 

I must warn you that RF 16/2.8 emits IR.  The IR contamination manifests itself with a greenish tint if the white balance is set for UV.  When you look at the pictures below by RF 16/2.8, see if you can discern the IR contamination.  An image of the contamination by itself is shown at the bottom.

 

Set 1, common daisy
 

El-Nikkor 80/5.6, f11, 1/4 sec, ISO 8000

 

RF 16/2.8, f11, 1/4 sec, ISO 8000

 

Set 2, Opuntia humifusa

 

El-Nikkor 80/5.6, f16, 25 sec, ISO 800

 

RF 16/2.8, this is a stack of five pictures; f5.6, 1/2 sec, ISO 2500

 

RF 16/2.8, f5.6, 15 sec, ISO 100

 

Set 3, Black-eyed Susan

 

El-Nikkor 80/5.6, f11, 1 sec, ISO 2500

 

RF 16/2.8, f11, 1/2 sec, ISO 4000

 

EF 40/2.8, f11, 2 sec, ISO 640

 

EF 28/2.8, f11, 2 sec, ISO 800

 

The amount of IR contamination depends on shutter speed, ISO speed, and focusing distance.  Aperture setting has no effect.  Obviously, the longer the exposure and the higher the ISO, the more contamination.  As for focusing distance, IR contamination is at the least at the infinity and at the most at the minimum focusing distance -- as the floating elements move forward for closer focusing, the gap in the throat of the lens mount opens wider and admits more IR.  The following is an image of the contamination.

 

RF 16/2.8 at minimum focusing distance, 30 sec, ISO 100

 

I plan to try Andy's subtraction method (https://www.ultravioletphotography.com/content/index.php?/topic/4203-tree-stump-uvivf-with-contaminating-light-removal-by-image-subtraction/) and see if the contamination can be reduced.

 

[Stefano]

Interesting. It seems like all RF/EF lenses pass some UV, probably the 40 mm one slightly more than the others. They are small and simple, which helps.

 

Here's a comparison between the 40 mm and 24 mm: https://www.ultravioletphotography.com/content/index.php?/topic/5200-lens-test-canon-24mm40mm-my-el-nikkor-80s-head-to-head

 

[Andy Perrin]

Here is the contamination with an autocontrast:

 

That contamination does NOT look like it was emitted by the lens - it looks like you have a light leak somewhere in your setup. Instead of trying to subtract it off, it would be better to trace the source of the leak by systematically covering the camera and lens with aluminum foil and peeling it back until the source is discovered. 

[Ming]

7 hours ago, Andy Perrin said:

Here is the contamination with an autocontrast:

 

That contamination does NOT look like it was emitted by the lens - it looks like you have a light leak somewhere in your setup. Instead of trying to subtract it off, it would be better to trace the source of the leak by systematically covering the camera and lens with aluminum foil and peeling it back until the source is discovered. 

But I took this picture in a closet.  The only light was this contamination on the LCD screen of the camera.

 

[ulf]

It could be some internal LED based position detector that bounces around and leaks into the image path.

[Andy Perrin]

Yeah internal LED in lens would be a good next guess. Can you take a picture of the RF lens powered on in the closet with a different camera, and see if you can spot the source of the light (if visible externally)? Could be something you could tape up…

[Andrea B.]

Ming, those are very good botanical photos with either lens.

 

I don't think I've seen contamination quite like that before. Light leak contamination is typically streaky not swirly. (...laughing at my technical description there....)

What camera were you using? Any chance that there is an internal IR shutter monitor LED?

[dabateman]

16 hours ago, Andrea B. said:

Ming, those are very good botanical photos with either lens.

 

I don't think I've seen contamination quite like that before. Light leak contamination is typically streaky not swirly. (...laughing at my technical description there....)

What camera were you using? Any chance that there is an internal IR shutter monitor LED?

Andrea it may not be the camera at all.

Many of the Canon Rf lenses have IR leds inside them to read off the focusing of the lens. So it can change with use and an odd new source of IR contamination for use who like full spectrum converted cameras.

 

[Ming]

Andrea, I used a modified Canon R5.  The camera is okay.  The problem is caused by the IR LEDs in many RF lenses.  As Ulf and David described, the IR LED is used to monitor the positions of lens elements.  The forum has a thread concerning RF 70-200mm leaking IR (https://www.ultravioletphotography.com/content/index.php?/topic/3869-canon-rf-70-200-experience/).  For a zoom lens, the problem is more complicated because the contamination changes with focal length as well as focusing distance.

 

Andy asked me to take a picture of the lens from the front while the lens is powered on.  I have a Canon 5D II with a 830nm filter fixed to the sensor.  So I attach the lens to a reverse mount adapter and attach that to 5D.  Basically, R5 and 5D are imaging each other through the lens.  The plan is simple.  Here is the twist.

 

The following two images are taken in a closet.  5D is in live view, so the mirror is up.  I also seal the optical viewfinder for good measure.

R5, f2.8, 30 sec, ISO 6400

 

5D, 30 sec, ISO 6400

 

The image by 5D should be rotated about 120 deg clockwise to align with the orientation in the "twist" picture.  It seems the LED is located in the bottom half of the lens and closer to the camera than the filter, but I am not sure it can be covered without causing malfunction.  For example, the lens front protrudes for close focusing.  I can push it back in, but immediately the stepping motor fights back.  So the camera is watching.

 

 

[Andy Perrin]

Darn it! Oh well, I guess this lens is not so useful for our kind of photography, unless you can rig up some rear filtration. 

[Kai]

Very helpful and useful investigation, Ming. Many thanks :)

[Ming]

Thank you, Kai.  Best regards.