which infrared filter for a beginner?

[Montanawildlives]

I am going to send my Xt-1 to Life Pixel for an IR conversion but can't decide on the filter I want installed. As a beginner, I guess that a bit of flexibility would be important (e.g., I probably would not want to be locked into the "deep BW" filter). I think the 720nm ("standard IR" to them) is the most popular, but I'm leaning toward the "enhanced IR" (665nm) or even super color IR (590nm). Is it true that the latter two are letting in some additional visible light, resulting in more saturated colors?

I don't think I'd be interested in the hyper color, super blue (too surreal) or full spectrum (don't want to carry around additional filters).

Any thoughts? Costs and benefits of the 720, 665, or 590?

Thanks.

 

[petrochemist]

I am going to send my Xt-1 to Life Pixel for an IR conversion but can't decide on the filter I want installed. As a beginner, I guess that a bit of flexibility would be important (e.g., I probably would not want to be locked into the "deep BW" filter). I think the 720nm ("standard IR" to them) is the most popular, but I'm leaning toward the "enhanced IR" (665nm) or even super color IR (590nm). Is it true that the latter two are letting in some additional visible light, resulting in more saturated colors?

I don't think I'd be interested in the hyper color, super blue (too surreal) or full spectrum (don't want to carry around additional filters).

Any thoughts? Costs and benefits of the 720, 665, or 590?

Thanks.

The numbers here reflect the approximate wavelength of the transition from opaque to transparent. So a 720 gives you roughly 720-1100nm, & a 590 gives 590-1100nm about 130nm more range. Above 830nm all three channels tend to give similar responses, so the color information is rather limited with a 720. Looking at the filters themselves 720 is pretty much black, 665 is a very dark red & 590 is red (actually identical to the old black & white contrast filters for darkening blue & greens).

If you get a 590 conversion you can later add a 720nm filter to get exactly the same results as a 720nm conversion. If you try adding a 590 filter to a 720 conversion you'll be unlikely to see any difference.

For me the only option was Full spectrum. I can add filters to the lens to get any of the three options you list as well as countless others. Some of the options giving blue, UV & IR work quite well (I like U340 & BG3 both somewhat similar to superblue) but I'd get tiered of the effect if it was the only option available. FWIW superblue is often not a surreal as a 590 giving similar landscape results to the 590 after the channel swap most shooters employ, with lenses that transmit unusually high amounts of UV the effects might be different - I've not got round to trying that properly yet.

IMO carrying a few filters is well worth the flexibility. Most of the time I have a 590 filter on my converted camera, with stepping rings to allow the same filter to be used on any of my routine lenses. My more specialist filters are usually only in 37mm size so use a restricted range of lenses.

If you really don't want to bother with the filters a 590 can be de-saturated to give results fairly similar to a 720 (not quite so contrasty) , It's much harder to get good color results from a 720.

 

[Montanawildlives]

I am going to send my Xt-1 to Life Pixel for an IR conversion but can't decide on the filter I want installed. As a beginner, I guess that a bit of flexibility would be important (e.g., I probably would not want to be locked into the "deep BW" filter). I think the 720nm ("standard IR" to them) is the most popular, but I'm leaning toward the "enhanced IR" (665nm) or even super color IR (590nm). Is it true that the latter two are letting in some additional visible light, resulting in more saturated colors?

I don't think I'd be interested in the hyper color, super blue (too surreal) or full spectrum (don't want to carry around additional filters).

Any thoughts? Costs and benefits of the 720, 665, or 590?

Thanks.

The numbers here reflect the approximate wavelength of the transition from opaque to transparent. So a 720 gives you roughly 720-1100nm, & a 590 gives 590-1100nm about 130nm more range. Above 830nm all three channels tend to give similar responses, so the color information is rather limited with a 720. Looking at the filters themselves 720 is pretty much black, 665 is a very dark red & 590 is red (actually identical to the old black & white contrast filters for darkening blue & greens).

If you get a 590 conversion you can later add a 720nm filter to get exactly the same results as a 720nm conversion. If you try adding a 590 filter to a 720 conversion you'll be unlikely to see any difference.

For me the only option was Full spectrum. I can add filters to the lens to get any of the three options you list as well as countless others. Some of the options giving blue, UV & IR work quite well (I like U340 & BG3 both somewhat similar to superblue) but I'd get tiered of the effect if it was the only option available. FWIW superblue is often not a surreal as a 590 giving similar landscape results to the 590 after the channel swap most shooters employ, with lenses that transmit unusually high amounts of UV the effects might be different - I've not got round to trying that properly yet.

IMO carrying a few filters is well worth the flexibility. Most of the time I have a 590 filter on my converted camera, with stepping rings to allow the same filter to be used on any of my routine lenses. My more specialist filters are usually only in 37mm size so use a restricted range of lenses.

If you really don't want to bother with the filters a 590 can be de-saturated to give results fairly similar to a 720 (not quite so contrasty) , It's much harder to get good color results from a 720.

Thanks so much for the comprehensive response! In addition to the increased hassle Factor one reason I didn't want to use filters was because they block light and can require tripods and very long shutter speeds of many seconds or longer. I know this is true for an unconverted camera using infrared filters. Is it also true for a full spectrum conversion using the various filters you describe Or not so much?

 

[petrochemist]

Thanks so much for the comprehensive response! In addition to the increased hassle Factor one reason I didn't want to use filters was because they block light and can require tripods and very long shutter speeds of many seconds or longer. I know this is true for an unconverted camera using infrared filters. Is it also true for a full spectrum conversion using the various filters you describe Or not so much?

Not at all. I very rarely put my converted camera on a tripod - If I do it's normally because the batteries died on my normal camera!

Having your camera converted drastically increases the response to wavelengths over 650nm. A 720nm filter on a Full spectrum camera will typically have similar exposures in sunlight to an unfiltered normal camera,

With a 590nm it will usually get more light than a standard camera. Even with some 10 stop filters you can still handhold the converted camera easily. Many ND filters transmit IR so the results may not be that different from a 720.

 

[JimH123]

I am going to send my Xt-1 to Life Pixel for an IR conversion but can't decide on the filter I want installed. As a beginner, I guess that a bit of flexibility would be important (e.g., I probably would not want to be locked into the "deep BW" filter). I think the 720nm ("standard IR" to them) is the most popular, but I'm leaning toward the "enhanced IR" (665nm) or even super color IR (590nm). Is it true that the latter two are letting in some additional visible light, resulting in more saturated colors?

I don't think I'd be interested in the hyper color, super blue (too surreal) or full spectrum (don't want to carry around additional filters).

Any thoughts? Costs and benefits of the 720, 665, or 590?

Thanks.

The numbers here reflect the approximate wavelength of the transition from opaque to transparent. So a 720 gives you roughly 720-1100nm, & a 590 gives 590-1100nm about 130nm more range. Above 830nm all three channels tend to give similar responses, so the color information is rather limited with a 720. Looking at the filters themselves 720 is pretty much black, 665 is a very dark red & 590 is red (actually identical to the old black & white contrast filters for darkening blue & greens).

If you get a 590 conversion you can later add a 720nm filter to get exactly the same results as a 720nm conversion. If you try adding a 590 filter to a 720 conversion you'll be unlikely to see any difference.

For me the only option was Full spectrum. I can add filters to the lens to get any of the three options you list as well as countless others. Some of the options giving blue, UV & IR work quite well (I like U340 & BG3 both somewhat similar to superblue) but I'd get tiered of the effect if it was the only option available. FWIW superblue is often not a surreal as a 590 giving similar landscape results to the 590 after the channel swap most shooters employ, with lenses that transmit unusually high amounts of UV the effects might be different - I've not got round to trying that properly yet.

IMO carrying a few filters is well worth the flexibility. Most of the time I have a 590 filter on my converted camera, with stepping rings to allow the same filter to be used on any of my routine lenses. My more specialist filters are usually only in 37mm size so use a restricted range of lenses.

If you really don't want to bother with the filters a 590 can be de-saturated to give results fairly similar to a 720 (not quite so contrasty) , It's much harder to get good color results from a 720.

I have 2 cameras converted to full spectrum. I like this because I can use any filter I want to.

Regarding filters, I use the cheap Chinese ones found on eBay and I have a number of sizes and wavelengths. I suspect that these cheaper filters have a fairly broad response, but they work just fine. I don't know if they actually extend to 1100nm. But I do have 590, 630, 720, and 830.

Did an experiment today. I live in California and we are experiencing lots of smoky haze from the fires. Going out today, I took shots using a non-modified camera and a modified camera using 720nm. There was an amazing difference in the detail the IR image was able to see though the haze that the regular image couldn't see.

 

[petrochemist]

Regarding filters, I use the cheap Chinese ones found on eBay and I have a number of sizes and wavelengths. I suspect that these cheaper filters have a fairly broad response, but they work just fine. I don't know if they actually extend to 1100nm. But I do have 590, 630, 720, and 830.

I'd be very surprised if they didn't extend to well beyond 1100nm. These filters are simple sharp cut longpass models, readily available from a wide variety of sources, filters that only transmit a narrower band of NIR are much more difficult to source & considerably more expensive.

The only longpass filters that have more gradual transitions & drop in transmission after reaching their peak are the very longest models like a 960nm. Even with these I've not seen any spectra that show less than ~50% peak transmission at 1100nm

IIRC Shanghai optics, Midwest optics & Edmunds optics offer bandpass type NIR filters, but even small sizes are >>$100 - Edmunds are the only ones who show prices on their website none are under $350 and I'd find a ½" diameter filter (their cheapest) difficult to use!

 

[JimH123]

Regarding filters, I use the cheap Chinese ones found on eBay and I have a number of sizes and wavelengths. I suspect that these cheaper filters have a fairly broad response, but they work just fine. I don't know if they actually extend to 1100nm. But I do have 590, 630, 720, and 830.

I'd be very surprised if they didn't extend to well beyond 1100nm. These filters are simple sharp cut longpass models, readily available from a wide variety of sources, filters that only transmit a narrower band of NIR are much more difficult to source & considerably more expensive.

The only longpass filters that have more gradual transitions & drop in transmission after reaching their peak are the very longest models like a 960nm. Even with these I've not seen any spectra that show less than ~50% peak transmission at 1100nm

IIRC Shanghai optics, Midwest optics & Edmunds optics offer bandpass type NIR filters, but even small sizes are >>$100 - Edmunds are the only ones who show prices on their website none are under $350 and I'd find a ½" diameter filter (their cheapest) difficult to use!

With those prices, it is why I use the cheap Chinese filters on eBay with prices around $7.

 

[petrochemist]

Regarding filters, I use the cheap Chinese ones found on eBay and I have a number of sizes and wavelengths. I suspect that these cheaper filters have a fairly broad response, but they work just fine. I don't know if they actually extend to 1100nm. But I do have 590, 630, 720, and 830.

I'd be very surprised if they didn't extend to well beyond 1100nm. These filters are simple sharp cut longpass models, readily available from a wide variety of sources, filters that only transmit a narrower band of NIR are much more difficult to source & considerably more expensive.

The only longpass filters that have more gradual transitions & drop in transmission after reaching their peak are the very longest models like a 960nm. Even with these I've not seen any spectra that show less than ~50% peak transmission at 1100nm

IIRC Shanghai optics, Midwest optics & Edmunds optics offer bandpass type NIR filters, but even small sizes are >>$100 - Edmunds are the only ones who show prices on their website none are under $350 and I'd find a ½" diameter filter (their cheapest) difficult to use!

With those prices, it is why I use the cheap Chinese filters on eBay with prices around $7.

Those prices relate to a different type of filter, but yes the Chinese ones are far more affordable than R72 & similar equivalents. Spectroscopically they appear to be just the same (unlike most people I have the equipment available to check!) - I don't have the means to measure optical flatness etc from them which might not be as good as the more expensive filters - but the results I get with the cheap ones are certainly still pretty good. If they are inferior it's not dramatically so.

 

[Montanawildlives]

Thanks so much for the comprehensive response! In addition to the increased hassle Factor one reason I didn't want to use filters was because they block light and can require tripods and very long shutter speeds of many seconds or longer. I know this is true for an unconverted camera using infrared filters. Is it also true for a full spectrum conversion using the various filters you describe Or not so much?

Not at all. I very rarely put my converted camera on a tripod - If I do it's normally because the batteries died on my normal camera!

Having your camera converted drastically increases the response to wavelengths over 650nm. A 720nm filter on a Full spectrum camera will typically have similar exposures in sunlight to an unfiltered normal camera,

And so, by implication, a "regular IR" camera conversion (720nm) would also have similar exposures to an unfiltered normal camera?

With a 590nm it will usually get more light than a standard camera. Even with some 10 stop filters you can still handhold the converted camera easily. Many ND filters transmit IR so the results may not be that different from a 720.

And again, by implication, the "enhanced IR" or "super IR" would be letting in more light than an unfiltered normal camera I guess.

I am trying to avoid having to buy and carry extra filters (the $275 to have the camera conversion done is about all of my fun budget for this unnecessary expense!). However, I am seeing the advantage of full spectrum....

Thanks.

 

[JimH123]

Regarding filters, I use the cheap Chinese ones found on eBay and I have a number of sizes and wavelengths. I suspect that these cheaper filters have a fairly broad response, but they work just fine. I don't know if they actually extend to 1100nm. But I do have 590, 630, 720, and 830.

I'd be very surprised if they didn't extend to well beyond 1100nm. These filters are simple sharp cut longpass models, readily available from a wide variety of sources, filters that only transmit a narrower band of NIR are much more difficult to source & considerably more expensive.

The only longpass filters that have more gradual transitions & drop in transmission after reaching their peak are the very longest models like a 960nm. Even with these I've not seen any spectra that show less than ~50% peak transmission at 1100nm

IIRC Shanghai optics, Midwest optics & Edmunds optics offer bandpass type NIR filters, but even small sizes are >>$100 - Edmunds are the only ones who show prices on their website none are under $350 and I'd find a ½" diameter filter (their cheapest) difficult to use!

With those prices, it is why I use the cheap Chinese filters on eBay with prices around $7.

Those prices relate to a different type of filter, but yes the Chinese ones are far more affordable than R72 & similar equivalents. Spectroscopically they appear to be just the same (unlike most people I have the equipment available to check!) - I don't have the means to measure optical flatness etc from them which might not be as good as the more expensive filters - but the results I get with the cheap ones are certainly still pretty good. If they are inferior it's not dramatically so.

Thank you. The cheapness of these Chinese filters has allowed me to purchase sets of IR filters in several sizes so that I can use them on several sizes of lenses, Also have step down rings so as to allow me to match additional lenses.

And they do seem to work pretty good. Or at least I am unable to determine that there are any issues.

Should bring up hot spots too. I do have a kit lens that if I stop down far enough, I do get a hot spot. The hot spot does go away as I open up the aperture. My other lenses do not seem to exhibit any such problem.

Also having two modified cameras. One, a Sony A55 which is a translucent mirror type camera. And a mirrorless Olympus EM5ii. The Sony requires me to manually focus to get a sharp image since with IR, the focus is a bit off. But with this camera, manual focus is a breeze. But with the mirrorless, there is no need to resort to manual focus as focus is on the sensor and it is always right on.

I should also mention that focus-by-wire is a real pain when doing manual focus. The Olympus lenses I have, including some pro level lenses, and even some Sony lenses are focus-by-wire. But for the Sony A55, I like to use older manual focus lenses, and manual focus is so easy and so sure with these lenses.

 

[JimH123]

Here is an example of using IR. Perhaps you are aware that we are having big fires in California. I live south of San Jose, so I am not close to any of the fires. But there is so much smoke, that even here is the Bay Area, the skies are very smoky hazy. So I decided to do an experiment and take a shot with a non-modified camera (Olympus EM1) and with a modified camera with a 720nm IR filter (Olympus EM5ii). I used an Olympus 40-150mm lens at 150mm on both cameras and shot at ISO 800 and f5.6 on both cameras. The two images were taken just minutes apart. But they don't overlap and were a few degrees apart from each other as I had forgotten exactly where I aimed for the 1st shot. With a 2.0x crop at 150mm, it is equivalent to a shot on a FF at 300mm, so it didn't much to miss the same spot. Distance was about 1 mile.

Here is an image with the non-modified camera.



And for comparison, one taken with the IR modified camera with the 720nm filter.



This gives quite a visual example of just how 720nm light can cut through the smoky haze which blocked the non-modified camera.

Also note that the IR shot was 1/1600 sec and the non-modded camera was 1/6400 which I attribute mostly to the overall brightness of the haze which was blocked on the IR camera.

 

[petrochemist]

Thanks so much for the comprehensive response! In addition to the increased hassle Factor one reason I didn't want to use filters was because they block light and can require tripods and very long shutter speeds of many seconds or longer. I know this is true for an unconverted camera using infrared filters. Is it also true for a full spectrum conversion using the various filters you describe Or not so much?

Not at all. I very rarely put my converted camera on a tripod - If I do it's normally because the batteries died on my normal camera!

Having your camera converted drastically increases the response to wavelengths over 650nm. A 720nm filter on a Full spectrum camera will typically have similar exposures in sunlight to an unfiltered normal camera,

And so, by implication, a "regular IR" camera conversion (720nm) would also have similar exposures to an unfiltered normal camera?

Yes they would be roughly the same in good sunlight. It's of course highly dependent on the light but within a couple of stops at least - very different to adding an IR filter to a normal camera!

With a 590nm it will usually get more light than a standard camera. Even with some 10 stop filters you can still handhold the converted camera easily. Many ND filters transmit IR so the results may not be that different from a 720.

And again, by implication, the "enhanced IR" or "super IR" would be letting in more light than an unfiltered normal camera I guess.

Indeed, my Full spectrum model typically sees around a stop more than a normal camera if I don't add filters to it - of course this screws up the colours but frequently not so severely that a viewer of the image would know.

I am trying to avoid having to buy and carry extra filters (the $275 to have the camera conversion done is about all of my fun budget for this unnecessary expense!). However, I am seeing the advantage of full spectrum....

Thanks.

If you have any other old filters around many of them also give interesting results on a FS conversion. Most Blue filters transmit IR giving both ends of the spectrum, likewise variable ND filters transmit IR & some standard NDs probably do too. Yellow, Orange & Red filters are the same as the various high/enhanced/super/ultra conversions...

Lighting, subject & white balance of course all affect how good the results for any particular filter will be giving a near endless range of combinations. With some filters even the lens used can make a drastic difference to the colours seen - some filters transmit UV with little visible, but most lenses block nearly all this UV, but thats getting into a new realm!