The GetDPI Photography Forum

Great to see you here. Join our insightful photographic forum today and start tapping into a huge wealth of photographic knowledge. Completing our simple registration process will allow you to gain access to exclusive content, add your own topics and posts, share your work and connect with other members through your own private inbox! And don’t forget to say hi!

GF1-IR, take 2

D

DavidB

Guest
You would find that B+W 486 won't cut it for the NMOS sensor. Its IR sensitivity is sky high.
With a Baader Planetarium 2" fringe killer filter (pricey) and custom WB, it comes close but not there yet.

With the B+W filter you could try adding a cyan filter. BG39 and the like. I will check those combos shortly.
How did you go with that testing?

At the moment I'm not sure about your assertion that the B+W 486 lets through too much IR.
I finally got my camera back last week and have only had limited opportunities to test it, but my testing so far shows decent performance for visible-light work with the 486 filter on. It doesn't match the behaviour of the original internal IR filters exactly, so the default WB settings don't work, but custom WB does a pretty good job.
Going further, custom DNG profiles combined with custom WB (using a Colorchecker Passport) do a great job.
Without the 486 filter there is definite IR "contamination" reflected from various subjects, but with it I haven't noticed any. Even though it's winter at the moment I do have some IR-strong light sources in my studio.

Filters such as the BG39 do their work by absorption, while the 486 does its work by reflection. The IR-blocking filters in most digital cameras involve both (in dual-filter setups like the G-system sensors where the front filter vibrates dust away, the "front" filter is the one with an IR-reflective surface. Absorption filters usually also introduce a cyan cast to the visible wavelengths, and looking down the mount at the sensor it's obvious that this colour cast has gone.
Rather than the sensor being SO IR-sensitive that the 486 "isn't enough", do you think it might be just that this visible cast is throwing off the camera's default WB mechanism?

The primary purpose for my G1 is as an IR camera, but with the B+W 486 filter it's also producing visible images I'm happy with so far.
 
V

Vivek

Guest
How did you go with that testing?



(in dual-filter setups like the G-system sensors where the front filter vibrates dust away, the "front" filter is the one with an IR-reflective surface. Absorption filters usually also introduce a cyan cast to the visible wavelengths, and looking down the mount at the sensor it's obvious that this colour cast has gone.
Rather than the sensor being SO IR-sensitive that the 486 "isn't enough", do you think it might be just that this visible cast is throwing off the camera's default WB mechanism?

The primary purpose for my G1 is as an IR camera, but with the B+W 486 filter it's also producing visible images I'm happy with so far.

It did not get anywhere as I am yet to find time to organize everything.

The dust shaker in G1 is completely transparent to IR. All the filtration is done from the glass stack that is in the middle. The sensor has a fixed glass (~1mm thick), the AA/UV/IR cut stack (2.75mm thick) and then the dust shaker (~1mm thick). The middle stack is both absorptive and reflective towards IR. I have shown pics of that (visible/UV and IR).

It is not just color balancing that is problematic. Even if that is taken care of and an image is apparently devoid of IR contamination, there is considerable amount of chromatic aberrations, color fringing and such that pose problems.

For me, it easier to use another (unmodified) G1 (prices are so low, it is cheaper to do that than buying band pass filters).
 
J

JoepLX3

Guest
I am not in the financial position to get 2 GF1's (one normal, one IR), but curious to see how far you would get with just an IR filter on top of GF1 (or LX3). Or is that not worth a try at all?
 
D

DavidB

Guest
Joe, this is similar to what I've done with my G1. It's been converted with quartz (i.e. it lets full IR+visible in, which you can control with external filters). It's not as convenient to use for visible-light photography as an unconverted camera (mainly due to trickiness with WB) but gives much better IR results than an unconverted camera.
But if you just want to experiment, get an IR filter (e.g. an R72-ish one) and try it on the front of your un-modified camera. You'll probably need to use a tripod to cope with the long exposures, but it's a cheap experiment.


Incidentally, I saved myself some serious $$ today. At the PMA show (it's on in Melbourne at the moment) I tried an M.Zuiko 9-18mm on my modified G1. I'd been thinking I might get hold of one of these. One of its benefits for me is that the filter thread is the same as that on the 14-45mm lens.
But no, I won't be getting one: with just a 720nm IR filter on the end it will not focus any further away than a couple of metres! The flare in IR is quite nasty too.
I suppose I could modify my camera further by moving the sensor forward slightly, but to bring this lens in range I might end up stuffing up operation with the 14-45mm. I'll leave well enough alone (at least for now).

The staff on the Olympus stand obviously didn't quite understand why I was attaching filters and wanting to try the lens on my G1, but they were happy to indulge me. Actually I don't think they caught up at all, still talking about being happy to sell a lens to a "fellow μ4/3 user" when I'd finished and was walking away. The words "won't focus to infinity" didn't really seem to register...
 
V

Vivek

Guest
If the focus isn't going past 2 meters, I would blame the (botched) camera conversion. See posts #137 onwards.
 

Jack

Sr. Administrator
Staff member
Agree. There can be some slight loss of true infinity focus, but it is usually so slight that normal DoF handles it.
 

cjlacz

Member
Pulling this thread back from the dead. Wonderful shots in here. Since I've preordered the OM-D I wanted to convert my GF1 to IR and thankfully found this thread. I'm email Precision Camera tonight about having it done. I thought about picking up a newer camera, but since I have 4 (now 5) batteries for the GF1 and the view finder seems like I should just stick with it. I found a mint condition Red GF1 at my local used shop for only $100 US so I picked that up and it's going to get used for the conversion. My beater will keep me busy until March 30th. Really looking forward to getting it back now.
 

RichA

New member
LX3 is a bad choice, its lens hot spots in IR. Which is too bad, it would make a wonderful compact IR solution otherwise!

Ken
IR work tends to exaggerate noise. I would not convert a P&S with a small sensor unless you are going to keep the images small (screen posts, not prints). At a minimum, a old Nikon D70 or a Canon Rebel, 4/3rds, etc., is a good idea.
 

RichA

New member
With the colour IR shots, you can get a normal colour balance or close to normal on everything except folliage which stands out in sharp purple contrast.

 
Top