Jump to content

Infrared


clairelucas

Recommended Posts

Hey guys. I'm new to the site, so hello! I just thought I'd ask the question, as infra red has only just got onto my radar as it were! I've been reading about ir filters etc, but wanted to ask - what happens if you stack red, green and blue filters? Does that work in the same way as an ir filter? I'm assuming not, but can someone explain in layman's terms why?

Thanks!

Link to comment
Share on other sites

Hey guys. I'm new to the site, so hello! I just thought I'd ask the question, as infra red has only just got onto my radar as it were! I've been reading about ir filters etc, but wanted to ask - what happens if you stack red, green and blue filters? Does that work in the same way as an ir filter? I'm assuming not, but can someone explain in layman's terms why?

Thanks!

Not something I've tried, but I doubt it, at least unless you are VERY careful in selecting your coloured filters.

 

I shoot quite a bit of IR & have access to a decent research spectrometer, so have recoded the transmission spectra for quite a few of my filters.

IR filters rarely transmit even 0.1% of light at wavelengths more than 50nm below their transition wavelength.

Normal coloured filters can - one of my red filters (nominally a 590nm long pass) transmits 4.3% at 400nm (the short end of blue).

Most ordinary filters transmit IR, but some of my green & blue ones don't transmit much.

 

I have used a variable ND for recording IR with a modified camera when I didn't have a proper IR filter the right size with me.

 

It's important to remember that while we only see light as three colours it's in fact made up of a continuum of different wavelengths. Having a filter that transmits more light in the 600-700nm region than any other visible band will make that filter look red, but it doesn't mean it reduces the blue & green wavelengths completely or even that significantly.

 

There are plenty of cheap options for trying IR photography, IR filters from China are usually under £20 each (720nm is the basic filter higher numbers are more restrictive & will generally want a converted camera. Shooting through an old floppy disc works (but gives longer exposures which are already long if you don't have a modified camera.) Processed but unexposed B&W film works better, but still not as well as a dark ND filter...

  • Like 2
Link to comment
Share on other sites

Not something I've tried, but I doubt it, at least unless you are VERY careful in selecting your coloured filters.

 

I shoot quite a bit of IR & have access to a decent research spectrometer, so have recoded the transmission spectra for quite a few of my filters.

IR filters rarely transmit even 0.1% of light at wavelengths more than 50nm below their transition wavelength.

Normal coloured filters can - one of my red filters (nominally a 590nm long pass) transmits 4.3% at 400nm (the short end of blue).

Most ordinary filters transmit IR, but some of my green & blue ones don't transmit much.

 

I have used a variable ND for recording IR with a modified camera when I didn't have a proper IR filter the right size with me.

 

It's important to remember that while we only see light as three colours it's in fact made up of a continuum of different wavelengths. Having a filter that transmits more light in the 600-700nm region than any other visible band will make that filter look red, but it doesn't mean it reduces the blue & green wavelengths completely or even that significantly.

 

There are plenty of cheap options for trying IR photography, IR filters from China are usually under £20 each (720nm is the basic filter higher numbers are more restrictive & will generally want a converted camera. Shooting through an old floppy disc works (but gives longer exposures which are already long if you don't have a modified camera.) Processed but unexposed B&W film works better, but still not as well as a dark ND filter...

Thanks - there's a lot to think about there! I was looking at the Hoya r72, but do you have any experience with the cheaper Chinese filters? I'm guessing you get what you pay for. Yep, my camera is unmodified - I don't object to getting a dedicated converted one, but imagine that it will be prohibitively expensive? That's interesting about the floppy disc so does it essentially work as an ir filter?

Link to comment
Share on other sites

I use Hoya R72 Filters. I mainly use a Fuji X100s (unconverted) for my IR Photography, though sometimes use Canon DSLRs.

 

I think that we usually always get what we pay for with Filters, but I haven't tried cheaper options so I cannot say for sure with respect to IR Filters. You'll probably need a Tripod and Head, so budget for those.

 

Here are some samples, Lake Te Anau was made hand held, and braced, but it took several attempts, I think it was about 2 seconds exposure - LINK

 

What camera(s) and lens(es) will you be using?

 

WW

Link to comment
Share on other sites

I shot some infrared with a cheap Chinese R72 filter and was very pleased with the results. What camera will you be using? Early DSLR's are more sensitive to infrared than than later ones, which have stronger internal filtering. Just get hold of an infrared filter to fit your lens. You can't see anything through this filter, so you have to compose and focus before fitting it, then shift the focus to the infrared mark on the lens. A tripod is essential as the shutter speed is going to be measured in seconds, so the technique is mostly restricted to landscape shots. I use manual exposure and judge it from the histogram, as auto exposure doesn't work at all well with infrared. Here's a shot I did with an old 5.4MP Nikon D1X - with much post processing:

 

391068142_Infraredcopy.thumb.jpg.5e8bdf40cfc28a17cc6ad67ac7c95d4a.jpg

  • Like 2
Link to comment
Share on other sites

. . . Shooting through an old floppy disc works (but gives longer exposures which are already long if you don't have a modified camera.) Processed but unexposed B&W film works better, but still not as well as a dark ND filter...

 

I am unclear of the meaning of the above -

 

Processed but unexposed B&W film works better than Shooting through an old floppy disc on an unmodified camera?

Is a dark ND Filter (in the sentenced underlined) only used with a modified camera?

Thanks.

 

WW

Link to comment
Share on other sites

I use Hoya R72 Filters. I mainly use a Fuji X100s (unconverted) for my IR Photography, though sometimes use Canon DSLRs.

 

I think that we usually always get what we pay for with Filters, but I haven't tried cheaper options so I cannot say for sure with respect to IR Filters. You'll probably need a Tripod and Head, so budget for those.

 

Here are some samples, Lake Te Anau was made hand held, and braced, but it took several attempts, I think it was about 2 seconds exposure - LINK

 

What camera(s) and lens(es) will you be using?

 

WW

Hey, sorry for tardy respon

I use Hoya R72 Filters. I mainly use a Fuji X100s (unconverted) for my IR Photography, though sometimes use Canon DSLRs.

 

I think that we usually always get what we pay for with Filters, but I haven't tried cheaper options so I cannot say for sure with respect to IR Filters. You'll probably need a Tripod and Head, so budget for those.

 

Here are some samples, Lake Te Anau was made hand held, and braced, but it took several attempts, I think it was about 2 seconds exposure - LINK

 

What camera(s) and lens(es) will you be using?

 

WW

I'm using a canon 750d at present. I have a range of lenses, but I was thinking a 58mm filter as a happy medium to step up/down from.

Link to comment
Share on other sites

I am unclear of the meaning of the above -

 

Processed but unexposed B&W film works better than Shooting through an old floppy disc on an unmodified camera?

 

Is a dark ND Filter (in the sentenced underlined) only used with a modified camera?

 

Thanks.

 

WW

I've only used them on (full spectrum) modified cameras,

 

As I have ready access to my spectral information while at work I can give a bit of indication -

Diskette material transmitted ~0.1% of light at 600nm (red), 0.9% at 700nm (border of red & IR), & 1.6-3% from 750nm to 1100nm (modified cameras IR region)

The processed negative gave ~1% throughout the visual, 24% at 800nm, 77% at 900nm & over 80% above 1000nm

One of my ND filters gave less than 0.1% in the visual. 20% at 800nm & over 70% from 900nm to 1100nm. these can be somewhat hit & miss as some ND filters block IR too.

 

Stock cameras typically have very little IR sensitivity and will generally see the short IR wavelengths much more than the longer ones. My old K100d which is fairly IR sensitive would probably show IR character with the ND filter and just perhaps (to a lesser degree) with the negative. I doubt the diskette would have been practical.

  • Like 1
Link to comment
Share on other sites

I don't object to getting a dedicated converted one, but imagine that it will be prohibitively expensive?

It depends, getting a camera modified is generally expensive (very roughly $250 but varied widely with the company) Some can make a bit of a mess of it too.

 

The first true modified camera I got, I brought preconverted with some cosmetic damage for less than a typical conversion.

I've modified a cheap point & shoot myself, but haven't been impressed with the focus. another I tried proved MUCH more awkward to open up.

Several older DSLR can work well without modification, The Pentax K100d & Nikon D70 are among the better of these.

Some of the Sigma DSLRs have a easily removable dust trapping filter, that also provides the IR blocking function. removing this (30s work on my SD14) makes the camera converted! However it records colour in a different way to most cameras & sees IR almost exclusively in the red channel.

Some old Sony cameras have a 'Nightshot' function which makes them IR sensitive my DSC V1 is a point & shoot that only cost £15 and gives IR sensitivity with limitations (forced high ISO, wide open only, green monochrome results)

  • Like 1
Link to comment
Share on other sites

Pictorial films (panchromatic emulsions) are not sensitive to radiations beyond the visible red, starts at 700nm. Infrared sensitivity is in the range of 700 to 900nm. In other words, hot objects emit IR within this range. We are talking about objects with temperatures 250 to 500°C (482 to 932°F).

 

Because film and digital cameras are sensitive to light, when imaging you need to exclude. Best is a Wratten 87. However, depending on the desired effect you can use a dark red.

 

Likely you have an IR TV remote. You can use this to make tests. If you focus on the business end of at IR remote, your camera will likely allow you view this emission. Now try filters you already have like, deep red or deep orange. Likely some extermination will find a good combination. You can even try cellophane candy or Xmas wrappings.

  • Like 1
Link to comment
Share on other sites

[...] Several older DSLR can work well without modification, The Pentax K100d & Nikon D70 are among the better of these.

The Nikon D70 is also not difficult to modify. You can find how to open it and remove the IR blocking filter on Internet (Nikon D70 IR cut filter removal). Given how cheap they are nowadays, worth a try.

Only 6 MP (and DX format) though.

Edited by q.g._de_bakker
  • Like 1
Link to comment
Share on other sites

Hey, sorry for tardy respon

 

I'm using a canon 750d at present. I have a range of lenses, but I was thinking a 58mm filter as a happy medium to step up/down from.

 

Using a Step-Down Ring to attach a Filter of a smaller diameter than the Lens's thread diameter may present an Optical Vignette.

 

i.e. a darker outer edge, to the image

 

WW

Link to comment
Share on other sites

Using a Step-Down Ring to attach a Filter of a smaller diameter than the Lens's thread diameter may present an Optical Vignette.

 

i.e. a darker outer edge, to the image

 

WW

Yeah, my only logical reason was that I have a couple of 58s, and some converter rings from my old Olympus SLR that scale to that size. I guess if it works I can look to buying more!

Are there any tutorials you guys would recommend for post processing/actually taking the photos? There seems to be a lot online, but I doubt know which sites are correct!

Thanks

Link to comment
Share on other sites

Pictorial films (panchromatic emulsions) are not sensitive to radiations beyond the visible red, starts at 700nm. Infrared sensitivity is in the range of 700 to 900nm. In other words, hot objects emit IR within this range. We are talking about objects with temperatures 250 to 500°C (482 to 932°F).

 

Because film and digital cameras are sensitive to light, when imaging you need to exclude. Best is a Wratten 87. However, depending on the desired effect you can use a dark red.

 

Likely you have an IR TV remote. You can use this to make tests. If you focus on the business end of at IR remote, your camera will likely allow you view this emission. Now try filters you already have like, deep red or deep orange. Likely some extermination will find a good combination. You can even try cellophane candy or Xmas wrappings.

To clarify most IR photography is done using reflected IR using the sun, flash or incandesant lights as the IR source. High subject temperatures are not needed for this. It's actually 'near infra red' that's being used & is very different to thermal cameras. Near infra red will see through many dyes, give high contrast clouds (white against a black sky), show most foliage as bright & see slightly under the skin in portraits typically giving a (much) improved complexion.

 

Wratten #87 (without a letter) is a 760nm long pass and is one of the popular IR gels, Most people prefer a Hoya R72 or equivalent (720nm)

Other Wratten IR gels include #87A (880nm), #87B (820nm), #87C (790nm), #88 (710nm), #88A (735nm) & 89B (690nm). I'd suggest the #88 or perhaps 89B would be best of these if using a stock camera as this maximises the amount of IR recorded while effectively blocking all visible light.

With a modified camera #25 (Red) is one of my favourites, allowing either false colour shots or monochrome IR.

 

The tip about the TV remote is a very good one - I used it in the early days for checking IR sensitivity of cameras, but not for judging the affects of filters :)

Link to comment
Share on other sites

Yeah, my only logical reason was that I have a couple of 58s, and some converter rings from my old Olympus SLR that scale to that size. I guess if it works I can look to buying more!

Are there any tutorials you guys would recommend for post processing/actually taking the photos? There seems to be a lot online, but I doubt know which sites are correct!

Thanks

There is no real wrong or right with IR, so pretty much all of them will be correct even when they contradict 8)

Many factors will depend on the camera being used. Many references make a big thing of the focus shift when shooting IR, if you using a mirrorless camera (or live view) this is no longer an issue as your focusing based on what the sensor sees. with DSLRs & film cameras you typically have to focus using visual light then move the focus so the distance focused is by the IR mark (old lenses usually have one - with newer lenses you may have to add your own mark by experimentation).

 

I've always found experimentation is the best way to learn how your camera/lenses/filters work with IR. There are just to many factors involved for on-line information to be totally reliable. I've found a lens that gives hotspots on one of my cameras but not on another, aperture, focusing distance & direction of the sun also can all play a huge roll in hot spots. White balance can also have a big effect but photographers have successfully used: paper, skin, skies, concrete, PTFE... as well as the generally recommended 'fresh grass'.

 

I find various IR groups on Flickr, & an IR forum to be among the most useful sources of inspiration sometimes suggest approaches I wouldn't have thought to try - like using a Foveon sensor with X1 filter to get aerochrom type results;

aa_SDI1437.thumb.JPG.143edbc25381678133fab4aa138dad3e.JPG a_SDI1435.thumb.JPG.7bab3aaabd2334ee0397d2d1b749f334.JPG

 

The look of these (SOOC) is very different than my other IR cameras can produce.

For comparison I think this was using a 590nm filter (on a full spectrum modified Panasonic GF2) with subsequent due adjustment (easier than the typical red-blue channel swap):

1280106316_ManingtreestationwatertowerfalsecolourIR.thumb.jpg.048bf184e0d705b3e5dec9a515232e39.jpg

  • Like 1
Link to comment
Share on other sites

  • 2 weeks later...

It is likely that a combination of color filters passes IR, though how much and of what wavelength is still in question.

 

I bought one of the cheap Chinese 720nm IR filters to play around with.

I tried it on one non-SRL, and then on a DSLR, maybe the D70s.

 

Note that the eye is slightly sensitive at 720nm, so with a sunny enough subject you might know which way to point.

 

Since it was only to have a little fun going around with an IR camera, my budget was low.

The filter looks just fine, and is plenty good enough for the tests I did.

 

There is a lot you can buy from China for a low price, especially when you don't need the highest quality.

 

I also bought an M39 body and lens cap, which work just fine.

 

Is there anyone in the US making M39 body caps? And how much do they cost?

-- glen

Link to comment
Share on other sites

I've just had a quick play with transmission data I have available for filters - only using transmission values every 100nm.

 

Combining Watten #25 (red), #47 (blue) & #58 (green) I don't see any transmission in the visible, but IR isn't too good either 0.01% @700nm, 0.2% @800nm, 0.5% @ 900nm, & 1% for 1000nm & 1100nm.

Changing blue filters the #47 for a Cokin A020, allows transmission at 600nm (total 0.004% through 3 filters) with no more than a doubling of IR

 

The budget 'Photo-R' lighting gels I have might be more suitable if the camera is highly IR sensitive (if not the transmission at 600nm will swamp everything else).

transmitting 0.002% @600nm, 0.7% @700nm, 5.5% @800nm, 20% @900nm, 24% @1000nm & 28% @1100nm.

 

To put these into context my budget Chinese IR filters typically transmit over 90% more than 30nm above their cut off & rarely anything more than 30nm below the cut off.

Probably also much less issues with reflections & distortions with only a single filter too even if it is only a $20 one.

 

Just playing with tabulated data can introduce big errors. If all three filters transmit at the same visual wavelength that's not in the table, there can be significant leakage, but lunchtimes come to an end so I can't play any longer.

Link to comment
Share on other sites

Ektachrome False-Color Infrared

Kincaid-aerial-6908.jpg.2bca8c65079f4baa5cf67cc249563296.jpg

1969

an archaeological mound site in southern Illinois. The film was used here to see if there were crop markings. Bare earth images after plowing were more useful. Bright red is vegetation that is thriving, blue is unhealthy vegetation or water, etc.

Link to comment
Share on other sites

  • 3 weeks later...
Stacking red, green and blue filters will only create a neutral density filter, just reducing all the light. No point in doing that.

 

It is well known that they dyes in color film are transparent in the IR, probably at that of GaAs IR LEDs around 850nm.

 

There is a set of filters, red, green, and blue, that more or less let through that color and block the other two.

These are used for color separation negatives, densitometers, and other places that need photographically

separate red, green, and blue images.

 

Might be #25, #58 and #47B

 

It looks like the #47 is transparent at 800nm and longer.

 

I suspect that most are transparent above about 800nm, but didn't find the spectrum for all of them.

-- glen

Link to comment
Share on other sites

Stacking red, green and blue filters will only create a neutral density filter, just reducing all the light. No point in doing that.

There's a very good chance the result will not be neutral. It will probably transmit significant IR, reducing visual light while leaving IR will work but there are much easier ways to get that effect, A variable ND filter set to it;s darkest works if you want to avoid true IR filters.

 

This was taken with a Vari ND filter on a converted camera:

50746448723_bd704e8c07_b.jpgPK 28mm vari ND by Mike Kanssen, on Flickr

The filters transmission works out close to a 830nm long pass filter (not quite one I've seen available but close enough to a 860nm)

 

Some blue / green filters are poor IR transmitters. I have several X1 filters (green) one of them transmits lots of IR & another is below 50% throughout the cameras IR range & the third is well below 10% throughout the 700-1100nm region. I suspect the difference is down to the anti reflection coatings, but have no ay to check.

Link to comment
Share on other sites

A reminder that the OP is using a Canon EOS 750D.

 

WW

 

The color film dyes question comes from digital ICE on color film scanners, using IR to detect dust.

 

The question had to do with the transmission spectrum of some dyes.

 

In any case, it seems that until about 800nm either color film or common color filters aren't so IR transparent.

 

Otherwise, the filters from China are reasonably priced, and work just fine.

-- glen

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...