Metering TTL for R72 Filter - Can it be done?

[jtg13]

I've been looking to get into IR photography for a bit and recently picked up several rolls of Rollei Infrared and an IR filter.
I'm working off a fairly modest budget so while a Hoya R72 would have been ideal, I elected to go for a cheap ebay special IR720. I know it's probably not the best filter, but I'm just looking to get my feet wet, not expecting absolute perfection at this point.

I've done some reading and it seems that the general consensus on the workflow is as follows:

1) Compose without filter -> Meter TTL without Filter - > Install FIlter -> Set shutter speed to compensate for filter factor
2) Compose without filter -> Install FIlter -> Meter with external meter -> Set shutter speed to compensate for filter factor

I just received the filter today and it is incredibly dark as expected so I have no doubts it probably works as an IR filter.
Just for the hell of it, I threw the filter on my Nikon FE with a 50mm F1.8 and pointed the camera out a window. To my surprise, I could very faintly see the image through the viewfinder! To top it off, the meter was actually able to detect a non-trivial amount of light; I was getting 1/30 on a brightly lit street with a wide open aperture.

My question is, how reliable is the meter in this configuration? Of everything I managed to dig up, most people seem to be of the opinion that metering TTL can't be done simply because the filter is too dark for the meter to even register. However, it seems in my particular case, the meter is able to detect the faint bit of light. Has anyone ever tried metering TTL with an R72?

 

[MattKing]

how reliable is the meter in this configuration?

It isn't.
Remember as well that the meter reads visible light, whereas the film + filter is only sensitive to near infra-red and a bit of infra-red. So what we do is meter the visible light, and use experience and knowledge about the typical correlation between the amount of visible light there is, and the amount of near infra-red and infra-red light there is.

 

[jim10219]

You can always give it a try. Is the reading without the filter of the same scene the same as the reading with the filter when compensating for the filter factor? If so, you're probably good.

Meters are supposed to read only visible light so that they're not confused by IR and UV light since most films don't react well to those wavelengths. If they did, they would likely underexpose outdoor scenes or overexpose ones under artificial light (since they don't have the same proportions of UV and IR to visible light as the sun). However some meters may read IR and UV okay, if they weren't properly filtered to address that issue.

Also since the focal point of lenses is different for IR and visible wavelengths, I wouldn't trust autofocus. But once again, it may work with some cameras/lenses.

My guess is what your camera is reading is what little visible light is passing through. Since you're able to see some image, that might be the same part of the spectrum that the camera is reading. If so, it'll likely overexpose the image pretty badly.

 

[Pieter12]

I would recommend the B+W ND Calculator app.

 

[abruzzi]

I know this isn’t helpful, but this is why I use a rangefinder with a non-TTL meter (Kodak Retina IIIC). I attach the filter, set the ISO to 12, focus with the rangefinder, adjust the focus to the infrared mark, frame and shoot. SLRs make IR more difficult.

 

[Maris]

If the TTL meter you are using is genuinely sensitive to infrared all you have to do is put the R72 filter over the lens, point your camera at your "standard scene", and then adjust the camera's ISO setting until it indicates the correct exposure. The correct exposure is discovered by shooting a bracketed series of exposures through the same R72 filter, same scene, same lighting. The calibration thus achieved is probably better than nothing or guesswork. Good luck.

 

[Sirius Glass]

I just started using a 720 filter instead of the Red 25 filter. I have tested my Red 23, Red 25, Red 29 and 720 filters with several calibrated light meters, and the through the lens and filter readings are not accurate. Instead take light readings without any filters and then adjust for the filter factors.

Red 23 use 2 f/stops
Red 25 use 3 f/stops
Red 29 use 4 f/stops
720 use 5 f/stops

 

[Marco Gilardetti]

I threw the filter on my Nikon FE with a 50mm F1.8 and pointed the camera out a window. To my surprise, I could very faintly see the image through the viewfinder! To top it off, the meter was actually able to detect a non-trivial amount of light;

It is likely that what the meter detects is mostly light coming in from the eyepiece of the finder, not from the lens. Stray light getting in from the eyepiece was a huge problem in early TTL cameras. The Nikon FE is already pretty well designed in regard to this, but still if the lens is almost completely darkened the meter will be mislead by the bright light coming from the eyepiece.

 

[neilt3]

I tend to use Rollei Infrared in a Minolta Dynax 9 when using it as 35mm .
Metering is spot on TTL with the cameras ISO set at 320 .
For focusing is spot on with AF .
I use a viewfinder like I use on range finder camera for framing and trust the camera for its spot on AF with the filter in place .
With manual focus camera I focus , correct focus for shift when I put the R72 filter on and meter with a Minolta spot meter .
The meter only goes down to ISO 12 , so I use that figure and double it as I treat the film as ISO 6 .

 

[Neil Grant]

I've been looking to get into IR photography for a bit and recently picked up several rolls of Rollei Infrared and an IR filter.
I'm working off a fairly modest budget so while a Hoya R72 would have been ideal, I elected to go for a cheap ebay special IR720. I know it's probably not the best filter, but I'm just looking to get my feet wet, not expecting absolute perfection at this point.

I've done some reading and it seems that the general consensus on the workflow is as follows:

1) Compose without filter -> Meter TTL without Filter - > Install FIlter -> Set shutter speed to compensate for filter factor
2) Compose without filter -> Install FIlter -> Meter with external meter -> Set shutter speed to compensate for filter factor

I just received the filter today and it is incredibly dark as expected so I have no doubts it probably works as an IR filter.
Just for the hell of it, I threw the filter on my Nikon FE with a 50mm F1.8 and pointed the camera out a window. To my surprise, I could very faintly see the image through the viewfinder! To top it off, the meter was actually able to detect a non-trivial amount of light; I was getting 1/30 on a brightly lit street with a wide open aperture.

My question is, how reliable is the meter in this configuration? Of everything I managed to dig up, most people seem to be of the opinion that metering TTL can't be done simply because the filter is too dark for the meter to even register. However, it seems in my particular case, the meter is able to detect the faint bit of light. Has anyone ever tried metering TTL with an R72?

...the problem is your meter's spectral sensivity isn't a good match for the film's 'response' to the minmal amount of light that passes though the R72 filter. Rollei IR 400 has almost NO sensivity beyond 700nm - just where the R72 filter starts to transmit. I don't bother with a light meter at all. I regard the IR 400/R 72 combination as 'no more than 4 ASA' and procede accordingly. In brilliant sunlight exposures are (of course) 1/4 @ f/16 - unless you want more shadow detail. Your cheap R 72 filter may be different to the costly Hoya version. Whatever, expect tripod operation in general.