Metering shadows and IR

[Arvee]

Out of sheer curiousity I did the following casual experiment with my collection of exposure meters. I went into the kitchen last night and turned the electric range on the lowest setting and let it come to temperature. Then, in an almost totally darkened room, I metered the IR coming from the stove.

I was amazed by the amount of "light" the meters saw!

I didn't record the numbers and may do that in another experiment but here in the desert in UT there is a ton of radiated IR in the shadows from the latent heat in boulders where temps can get upwards of 140F. Both CdS and Si registered strong numbers with Si being higher but, of course, my old trusty Sekonic L398M didn't flicker. which I expected from a selenium meter.

I need to run some more tests but this sure explains some of the inconsistencies I see in summer desert photo sessions. Maybe the rule here is selenium in the summer sunshine and use the others during the other seasons although I am not sure how much reduction of IR occurs during the colder seasons, or, just stick to the older technology for landscape shooting.

IR experts, please comment....

Thanks!

Fred

 

[keithwms]

Well, bear in mind that the IR light you capture in an IR photograph is reflected IR. E.g. foliage and soil filter out the wavelengths that make it to your film. So you're not recording the full sunlight spectrum- only particular pieces of it, even if your film has extended sensitivity. And I think you are seeing zero radiated, long-wave light in an IR photograph.

If you point your meter directly at a glowing object, that is a rather different thing. That glowing object is emitting all over the place. And the meter might also be detecting over a much wider range, not over the fairly limited near-IR range that ordinary IR film picks up.

Offhand I would think that the "stored heat" that is gradually radiated from rocks etc. would be long wavelength, way past what ordinary IR film can see. I think we are talking about micron wavelengths, say 8000nm or more. Film cuts out at a micron or so at best. Photodetectors that can see further are specialty items; some digital cameras can see out to 1200m or so but not much further than that. But a meter might see quite far. So care must be taken when trying to meter IR!

Anyway, why not take a picture of your stove and see what you get!

 

[Arvee]

Keith,

I am more concerned about the introduced error in the meters from the broadened spectral response that seems to include near IR sources even though the spectral response of the film wouldn't see them. I am pretty certain the IR component is influencing the overall exposure reading.

Which leads me to wonder about a less than obvious possible advantage of the spot meter in its remote measurement of light sources. I have been walking into the shadows with my wide field meters to take measurements and stand relatively close to these radiating monoliths.

Maybe Helen B. will catch this thread and log in.

F.

 

[keithwms]

Obviously the meter will give an erroneous reading unless it's range of detection matches the range of sensitivity of your film + filter.

You could meter through the same filter that you shoot, that'd be more accurate. But the film probably still truncates part of the pass band of the filter.

Or are you saying that even when not shooting IR film with an IR filter, you worry that your meter is picking up IR wavelengths and that is skewing your readings?

 

[Arvee]

Or are you saying that even when not shooting IR film with an IR filter, you worry that your meter is picking up IR wavelengths and that is skewing your readings?

Yes.

 

[keithwms]

Ah okay. Well, the monoliths radiating at ~10 micron wavelengths don't matter much at all to your meter. If it really worries you, then put a hot mirror over your meter and then you'll see how much it changes your meter readings. Offhand, without googling around, I think your meter might see out to ~1200 nm or so. The rocks are emitting at much longer wavelengths. The hotplate is emitting all over the place.

 

[Arvee]

Thanks for the enlightenment.

Most probably not a factor in determining correct exposure and my inconsistencies are then more likely cockpit errors in judgment.

I found some data that supports Gossen meters out to about 1000nm before tailing off.

Fred

 

[dslater]

Ah okay. Well, the monoliths radiating at ~10 micron wavelengths don't matter much at all to your meter. If it really worries you, then put a hot mirror over your meter and then you'll see how much it changes your meter readings. Offhand, without googling around, I think your meter might see out to ~1200 nm or so. The rocks are emitting at much longer wavelengths. The hotplate is emitting all over the place.

Hmm - the ~10 micron value is the peak of the black-body radiation curve for a black-body radiating at 333deg K (==140 deg F). The meter sees out to ~1200 nm == 1.2 microns. So, the peak radiation is indeed not within the meter's sensitivity. However, at this low temperature, the black-body curve is rather flat and it is quite possible that significant radiation is happening down at the 1 micron range.

The OP said he turned his range on to the lowest setting - this usually is around 200deg F == 366deg K - at that temp, the peak radiation is at 7.9 microns as opposed to 8.7 microns for 140degF - if he sees an effect from the range at 200deg, then I would expect 140degF to also cause an effect.

 

[keithwms]

A rock and a hotplate have very different emissivity characteristics, so you can't transfer temperatures quite like that. Yes, you're right that rock has a kind of weak "tail" of emission right into the near IR, but that output is very, very negligible compared to directly reflected visible and near IR light. Rock is not at all like blackbody, and only the outermost surface is hot. Rock/soil has all kinds of diffuse emission peaks that vary with mineral content etc. A hotplate or filament is much more like a true blackbody, it is a well-defined thermal mass at one peak temperature.

This matters because a blackbody at a given temperature will concentrate most of its emissions in one wavelength range. That's not the case with (most) rocks, as far as I know.

N.b. the earth itself emits longwave IR at everything from ~3-12 microns and there are peaks and such. The longwave emissions are a big part of the so-called greenhouse effect that we always hear about.

It would be fun to point that meter at some lava and see what you get!

Sorry for the rambling physics lecture

 

[dslater]

Rock is not at all like blackbody, and only the outermost surface is hot.

I dispute this assertion - I believe a rock is like a black-body to a very good approximation - it's radiation pattern is a continuous spectrum with some absorption lines due to it's mineral content. The interior temp of the rock is irrelevant since no radiation reaches you from the interior. In the same way, although the core of the sun is at 10 million degrees, it's radiation curve is that of a black-body radiating at a temperature of 5500deg - the temperature of the sun's surface.

The earth also radiates as a black body as does the sun and most anything that radiates due to simple heating as opposed to something like a nebulae or an excited gas that is radiating due to an emission at a couple of specific frequencies.

 

[Roger Hicks]

From memory, anything beyond about 1300nm is absorbed pretty quickly by water vapour in the atmosphere. But I could be misremembering.

 

[dslater]

This matters because a blackbody at a given temperature will concentrate most of its emissions in one wavelength range. That's not the case with (most) rocks, as far as I know.

A black-body radiates with distribution of wavelengths that has a peak at a given wavelength. This does not mean it radiates most of it's radiation at one radiation - most of the radiation is concentrated is a range of wavelengths centered around the peak. Also, the higher the temp, the more concentrated this radiation is.

 

[Arvee]

Thanks, guys, for the responses. I see from the info that there is a possibility there may be an introduced error in exposure determination from readings taken in the desert where temps may be quite high on the objects photographed. Whether or not the film sees this "invisible light" is immaterial, I was more concerned about the errors in the exposure meter picking up extraneous energy and influencing the final exposure measurement.

So, if I understand the conclusion correctly, there is a real possibility the meter would respond to desert boulders at 140f much like it responded to the range element at 200F.

BTW, I just measured the range element cold and it measured EV 5. The measurement rose to EV 10 when the element was just beginning to radiate.

About five f stops just from the warm element!

I will try to do some testing next time I am in the desert but it may be difficult since most of the inanimate objects will be at ambient temperature.

Thanks!

Fred

 

[dslater]

Thanks, guys, for the responses. I see from the info that there is a possibility there may be an introduced error in exposure determination from readings taken in the desert where temps may be quite high on the objects photographed. Whether or not the film sees this "invisible light" is immaterial, I was more concerned about the errors in the exposure meter picking up extraneous energy and influencing the final exposure measurement.

So, if I understand the conclusion correctly, there is a real possibility the meter would respond to desert boulders at 140f much like it responded to the range element at 200F.

BTW, I just measured the range element cold and it measured EV 5. The measurement rose to EV 10 when the element was just beginning to radiate.

About five f stops just from the warm element!

I will try to do some testing next time I am in the desert but it may be difficult since most of the inanimate objects will be at ambient temperature.

Thanks!

Fred

Hey, maybe you could get yourself a cheap digital P&S camera and remove the IR filter from it. I think all most digital cameras have an IR filter in front of the sensor. Or you could try a piece of heat-absorbing glass intended for use in an enlarger. You could then meter hot rocks through the glass/filter and see if there is a difference.

 

[keithwms]

Yes that is why I suggested metering through a hot mirror. They are inexpensive, no need to take apart a p&s.

You will see that it makes zero difference when metering boulders. Hotplates and boulders are apples and oranges, in terms of emissivity. But try it and let us know.

 

[Arvee]

Excellent thought!

I used my HAG from my Beseler 23 and did a couple of measurements and that reduced the meter measurement by EV 2. The measurement is still inflated by 3 EV numbers over the cold range element.

This is a piece of special heat absorbing glass that is roughly 6x6x1/8" thick. I see now why the HAG isn't that effective in reducing negative pop.

Sorry, Keith, I am not familiar with a 'hot mirror.'

Fred

 

[keithwms]

A hot mirror is just a piece of glass that reflects IR. One is found over almost all DSLR sensors.

Look at the description of the tiffen hot mirrors at b&h:

Some digital still cameras can have increased sensitivity to infrared light that may cause color rendition problems. The Tiffen Standard Hot Mirror is designed to remedy this problem by reflecting most infrared light.

You should be able to find a small one for very low cost.

Some meters have them, some don't.

But again, a range and a boulder are apples and oranges. You need to think about emissivity and actual radiant output. A surface-heated rock isn't radiating significantly in the near IR. Hot lava would Look at the blackbody plot posted by dslater, the high tail is to the longer wavelength side.

 

[dslater]

I have attached a document showing the black body spectrum for both a 200°F Black Body and a 140°F Black body. As you can see, in the range on 1 micron to 8 microns, the 200°F body radiates about twice as much power as the 140°F object, So if the 200°F object gives you a meter reading, then the 140°F object should read about a stop less.

Dan