Need help with filter problem UV/IR

[Stefano]

Could it be the XNite-330C filter, which has CONSIDERABLY BETTER infrared blocking?

https://maxmax.com/s...ry_pathway-9204

Since it is a coated filter, you can immediately see if that's the case by looking at the sensor.

 

Looking at the graph it should be about OD 2, not enough to prevent IR contamination, but considering that the sensor is monochrome (higher UV sensitivity)... it may work?

[Stefano]

Quote from them:

"Please note: This filter has about a 1% IR transmission at 720nm. If you are using this filter with a modified camera in sunlight, the camera will see more IR than UV because cameras are very IR sensitive and not very UV sensitive."

[Andy Perrin]

I think that's exactly it, I think the monochrome sensor is much more UV-sensitive than we are used to on this board (except for Jonathan and Bob) so with the coated filter it's enough to make the photo mostly-UV. The test scene doesn't allow us to really gauge the leak without color to help us, also. We need something like a dandelion or black-eyed susan that has both UV bright and UV dim regions on the same flower.

 

Stefano- I think he meant non-mono cameras.

[Stefano]

Stefano- I think he meant non-mono cameras.

Agree, that's why the monochrome conversion must be taken into account. Assuming 3 stops more UV sensitivity (that's probably unrealistically high), it would be as the filter (if it is that filter) has OD ~3 in the IR, and that is borderline visible contamination.

[Cadmium]

Yeah, if it is the 330C then that would explain it. That is probably what it is.

[StephanN]

Yeah, if it is the 330C then that would explain it. That is probably what it is.

 

 

Ahemm, just had a look at the email-correspondence with Dan from the time I bought the camera, and he does in fact mention that they're putting in the 330C. Am feeling pretty stupid right now

 

Thanks to everybody for their help

[StephanN]

Oh yes, and I will go and find some flowers to check, but probably won't have time for that until the weekend

[Andy Perrin]

Haha, don't feel too stupid, it's easy to miss that "C." Also this has been very informative. The indications are that if you have a MONOCHROME camera, then it may be possible to accept a 1-2% IR leak without drastically changing the UV-nature of the photo. We should verify with a dandelion or something, though.

[StephanN]

Haha, don't feel too stupid, it's easy to miss that "C." Also this has been very informative. The indications are that if you have a MONOCHROME camera, then it may be possible to accept a 1-2% IR leak without drastically changing the UV-nature of the photo. We should verify with a dandelion or something, though.

 

I've been thinking (yeah, ought to know better): If I want to determine the true amount or IR-contamination, wouldn't it suffice if I first took a random photo outside in the sun and then repeated it with, say, a 670nm filter? From zhe settings required to bring the two photos to the same level of exposure, one might estimate the IR-leak?

 

So far, I've never tried an IR-filter on this camera.

[Andy Perrin]

Yeah, that kind of test we do around here a lot. Usually we leave the settings alone and argue that if the photo with the long pass filter is black then there isn't any significant leak. The flower would be more of a practical test, since you want to know how does the signal compare to the "noise". Try both?

[dabateman]

What we argue a lot?

Once you see its black with the settings you would actually use. Then you can crank up the exposure to see how many stops more the leak will kick in.

That second part is actually relavent, as some subjects may not reflect back much UV and it could be an issue.

 

I saw that once with a really dark red flower outside. The UV reflected was extremely low and was pushed to the point of seeing IR reflectance. Lead to some cool brown blue black colors.

[StephanN]

Reason I'm asking: with homeoffice due to Youknowwhat I can do a quick test during my lunch-break on my balcony :-)

[StephanN]

Well, I did a test-series today, and here are the photos. All of them at ISO 1600 and f/3.8. The times can be seen in the screenshots, as well as the respective times and the histogramm. I compared the naked X300C with X300C plus an IR 630-filter and with X300C plus IR filter + BP1. [Edit:] Lens is the Soligor 21mm f/3.8[/Edit]

 

Conditions were sunny with some clouds, with the sun being almost at a right angle to my right, as you can see in the first photo with the shadow of the railing. Please forgive that I didn't really pay attention to correct focussing and consistent framing, saw no need for that

 

Tentative summary: seems that on a sunny day with plenty of UV, one might get away with omitting the BP1, if there are no strong IR-sources present apart from the sun, because the sun will also give plenty of UV. Or am I being too optimistic here? I don't really mind using the BP1, unless I want to use several lenses, then the unscrewing and screwing is a bit tedious. However, if I can "cheat" a little, I may be able to use lenses which don't lend themselves to filters easily, like fish-eye lenses.

 

The filter is from Optik Makario:

 

First, X330C alone, 1/50s:

 

Next, X330C + IR 630, 1/50s, 1/25s, 1/10s, 1/5s, 1/2s:

 

 

 

 

 

 

Finally, X330C + IR 630 + XBP1, 1/50s, 1/25s, 1/10s, 1/5s, 1/2s:

 

 

 

 

 

[Andy Perrin]

It looks like it's borderline. No leak should really truly look like a black image ("a picture of a black cat in the dark" as David puts it). But it's not a very big leak, and so I think you could get away with calling it a UV photo, especially since you know how bad the problem is now. One word of caution: if you use a different LENS and if that lens happens to block more UV than this one, then the problem could get a lot worse! That fisheye you mentioned, it will be hard to know if it's okay unless you test it again because it may not have good UV bandpass.

[StephanN]

The one fisheye I have is fairly new, so no UV-transmittance there But looking at those photos, I'm beginning to ponder the question whether I ought to get a filter which is blocking IR better than the BP1, especially once I get around to taking photos of plants.

[Stefano]

Yes, your BP1 seems to do almost nothing. You should get something better, otherwise it is like not having it.

[Andy Perrin]

Yeah, I somehow missed the "with BP1 also" photos but that filter is not helping.

[Cadmium]

You need more foliage in the pic if you want to test for Red/IR leaks. Or dandelion or rudbeckia... But just foliage will show a difference, blocked or not blocked comparison.

[dabateman]

You need more foliage in the pic if you want to test for Red/IR leaks. Or dandelion or rudbeckia... But just foliage will show a difference, blocked or not blocked comparison.

 

Yes true.

But his subject reflects a low amount of UV. So nearly worst case to see how soon the IR came in. With a real floral subject it might be better.

However, your bp1 wasn't good. Also you have only about 1 stop before leakage. Not great.

 

So if you strictly want UV you will need to stack a filter. S8612 might be best.

But this is a monochrome converted camera right? So may not matter. Just have a strong UV illumination source and we may not know. As in use a UV modified flash or use a 365nm LED flashlight. That should over power the IR and the monochrome image will help.

 

But if your UV images start to look too soft, than you will need a S8612 to block that IR leakage. A uv monochrome image looks quite sharp and contrasty.

[StephanN]

Hm, in the first photos there was plenty of foliage and it was pretty dark, but I'll try to find some more today after work. Not sure if I'll be able to find the right flowers, though.

 

But just so that this may be the conclusive test: is it better to take overview-photos with the Soligor/Nikkor or use the EL-Nikkor for macro or landscape, and the landscape consisting of grass and/or trees?

 

In either case, I can pair the camera with the ir630, bp1 and s8612 and combinations.

[Andy Perrin]

I think both of those lenses have good UV pass, but the answer does depend on the lens somewhat also. You should pick a lens and we will try to answer the question for that particular lens. The problem is that you need an object that has UV dark/IR light and also UV light/IR light (like the dandelion) because all we have to judge the leak qualitatively is the contrast, in a mono camera. Definitely vegetation also helps because it is UV dark/IR light so it will show the leak better for the filter tests, but if you want to be able to see if the leak is washing out your photos (which is what we care about, I think?), then we need a subject like the dandelion.

[StephanN]

Unfortunately no time today to go out hunting. But: I've found this blossom hiding on my balcony, will it be OK for testing? Just snapped quick VIS and X330C shot; I belive it's some sort of Malva.

 

Since it's on my balcony I can do all sorts of tests, apart from using sunlight, bad weather coming up

[Andy Perrin]

No idea. We have to have some kind of known expectations of what it should look like. If you know what kind it is, see if we have photos on the board already.

[StephanN]

It appears to be Malva sylvestris ssp.mauritiana, which also goes by the name of malva sylvestris zebrina . I've found malva sylvestris here, but not the zebrina variation.

[Cadmium]

Another way to test for a Red/IR leak is:

1) shoot an optimized UV exposure.

2) shoot another photo (same scene), using the exact same exposure, ISO, aperture, all the same settings as the first exposure, except stack a longpass filter in front to block everything except the range you are concerned about leaking.

For your filter, maybe stack with some red filter, 590nm, #25, etc..

This will show you how much Red/IR is mixed into your first UV shot.

If the frame looks black, that is good, that is what you want, if you see a faint ghost image, then that is the leak, and that is what you don't want, you want to suppress that, you want a black image.

Again, use the same setting and exposure for each shot, make sure the first shot is an optimal exposure.