Has anyone had a 'Merrill' converted to IR only?

[Lincoln Castricone]

I've had several Canon cameras converted to IR only with the 830nm filter applied directly to the sensor by MaxMax. These have been great IR to B&W cameras.

Dan Llewellyn at MaxMax says that he has not converted any as yet. Two issues he brought up are:

1. With the 830nm filter the Foveon sensor will be strictly a 15 mp sensor because the pixel sites will will receive only red light. 'Keep in mind, though, that a Bayer camera at 830nm is 99% a monochrome device.'

2. "The thickness of the glass changes the focal plane so you need to get the right thickness for focus to be correct. I would have to take one of the cameras apart and measure the ICF/AA."

Is the 15mp limit a problem in terms of IQ and/or print size? Any experience or thoughts about this conversion?

www.lincolncastricone.com/blog

 

[Kendall Helmstetter Gelner]

Lincoln Castricone wrote:

I've had several Canon cameras converted to IR only with the 830nm filter applied directly to the sensor by MaxMax. These have been great IR to B&W cameras.

<...>

Is the 15mp limit a problem in terms of IQ and/or print size? Any experience or thoughts about this conversion?

A more compelling conversion to me is one of the DP-M cameras. Since you can already easily remove and replace the IR filter in the SD-1M and use an IR filter over a lens, converting the camera full time to IR does not gain that much benefit beyond easier framing (and in fact wouldn't the focus sensor have to be adjusted too? Perhaps that's part of the service).

Regarding the printing though, here's where you see how nebulous a concept "MP" is for saying anything about anything. In a Bayer sensor a converted sensor (IR or other monochrome conversion) truly matches the photosite count instead of the upsampled output the camera generated before that just happened to match the photosite count. Thus it gains a huge boost in output quality even though it's the same "MP" camera it was before.

In a Foveon sensor you lose nothing in the process of removing data from the other color channels, it's still the same pixel-level sharpness the camera has always had to begin with. The Foveon chip retains the same level of detail regardless of scene color, and gathering IR in the red channel is no exception. Thus with a good lens, a 30x48 print off the data should still look really good, just as good and sharp as if the camera were shooting in color.

You can see an original sized SD-1 IR image (slightly cropped) here:

--
---> Kendall

Kendall

Explore Kendall’s 4,149 photos on Flickr!

www.flickr.com

http://www.pbase.com/kgelner

Galleries by The Sigma Users Group Galleries

www.pbase.com

 

[Lincoln Castricone]

Thanks Kendall.

On a IR filter to sensor conversion apparently the new IR focus adjustment is done by choosing the right filter/glass thickness for the camera's ICF/AA, whatever that is.

I agree that one of the DP-M cameras make the most sense for such an IR conversion.

Kendall Helmstetter Gelner wrote:

Lincoln Castricone wrote:

I've had several Canon cameras converted to IR only with the 830nm filter applied directly to the sensor by MaxMax. These have been great IR to B&W cameras.

<...>

Is the 15mp limit a problem in terms of IQ and/or print size? Any experience or thoughts about this conversion?

A more compelling conversion to me is one of the DP-M cameras. Since you can already easily remove and replace the IR filter in the SD-1M and use an IR filter over a lens, converting the camera full time to IR does not gain that much benefit beyond easier framing (and in fact wouldn't the focus sensor have to be adjusted too? Perhaps that's part of the service).

Regarding the printing though, here's where you see how nebulous a concept "MP" is for saying anything about anything. In a Bayer sensor a converted sensor (IR or other monochrome conversion) truly matches the photosite count instead of the upsampled output the camera generated before that just happened to match the photosite count. Thus it gains a huge boost in output quality even though it's the same "MP" camera it was before.

In a Foveon sensor you lose nothing in the process of removing data from the other color channels, it's still the same pixel-level sharpness the camera has always had to begin with. The Foveon chip retains the same level of detail regardless of scene color, and gathering IR in the red channel is no exception. Thus with a good lens, a 30x48 print off the data should still look really good, just as good and sharp as if the camera were shooting in color.

You can see an original sized SD-1 IR image (slightly cropped) here:

--
---> Kendall
http://www.flickr.com/photos/kigiphoto/
http://www.pbase.com/kgelner
http://www.pbase.com/sigmadslr/user_home

 

[KENTGA]

I'm a Canon shooter and bought an SD 14 a few years ago for IR. I popped out the IR blocker and use it with and without a Tiffen 87 filter. You can't see anything thru the viewfinder with the 87 filter on, but autofocus works great. You just have to guess about the zoom factor and where the subject is.



Kent