Jump to content
UltravioletPhotography

ZWB1 + QB21 2mm deep red light contamination


Avalon

Recommended Posts

Hello. I‘m happy that at last I was able to convert Sony A3000 to full spectrum camera without damaging delicate parts. I received various filters of which 49mm ZWB1 and QB21 2mm thick I wanted to use for ultraviolet photography. But after testing them with camera I noticed that filters leaked deep red light. I used standart f/3,5 lens and there seemed to be view lag indicating that not a lot that deep red light passed. In photos redness is more dramatic. When I stack another 2mm ZWB1 filter on top I still get redness visible. When QB21 used alone view looks normal. Here sample photos made during cloudy day and indoors under incandescent lamp:

https://drive.google.com/open?id=1Kn6rKq1sxEJXmaXkClnIuTQ3oaQdNt_5

https://drive.google.com/open?id=1fz0JJhvpzXdBmJkSKZRRwqlHi8h4Kl13

https://drive.google.com/open?id=1WZ47zC1IyOmgqzLiFWqvsKAC67NMRZK7

https://drive.google.com/open?id=1t8bUDfOowe21iA2GO-yGn-y9kAzst8Ui

https://drive.google.com/open?id=1JNZBl1IZtESc-IAh8qLa5IKYfm23xbzD

https://drive.google.com/open?id=1-AUWdqFOOvNDILId6oLYjPVIHT2g64oT

https://drive.google.com/open?id=1XW_sVpaZKkD_I5ZqYMfWWk0slGIN6Esv

 

So my question if it is possible to fix this problem with Schott BG40 or S8612 filter and at what thickness? In theory better red/infrared supressing hot mirror should allow only UV pass.

Link to comment

S8612 1.5mm or 1.75mm will stop the leaks. However part of your problem is that incandescent lights have far more IR than UV, worse than sunshine I think. Also note that most windows at least partly block UV, making things worse.

 

Finally, you have to white balance your images because UV pics have a strong magenta cast to them. (In fact, I wonder if you are mistaking the magenta cast as a red leak?)

Link to comment
I heard a lot of criticism of chinese filters but not much reasons for what. Maybe thicker QB21 filter wouldn't have this problem. Is red light not visible with S8612 or BG40?
Link to comment

S8612 1.5mm or 1.75mm will stop the leaks. However part of your problem is that incandescent lights have far more IR than UV, worse than sunshine I think. Also note that most windows at least partly block UV, making things worse.

 

This is not always true, especially for 1.5mm.

I would aim for a S8612, 2mm thick as with 2mm the IR attenuation is even stronger and the loss of UV-transmission is not very strong compared to 1.5 or 1.75mm.

At 700 nm 2mm attenuate 10 times better than 1.5mm, but at UV the difference is just a few percent.

BG40 is not at all as efficient. BG40 works better than S8612 as a VIS restoring filter to mimic the original BG filter that was removed from the camera during conversion.

Link to comment

I heard a lot of criticism of chinese filters but not much reasons for what. Maybe thicker QB21 filter wouldn't have this problem. Is red light not visible with S8612 or BG40?

 

Yes the Chinese blocking filters are bad. The S8612 2mm is expensive, but you only need to buy it once and it will block your IR leak.

Link to comment

Don't buy Chinese glass. You don't know what you are getting, and the stuff I tried (both of those glass types) is poor and inefficient. Dirt cheap indeed.

If you were using U-360 or UG11 at 2mm thick then S8612 1.5mm would work fine, 2mm is very secure, but it all depends on the balance of thickness of both glasses.

Frankly, I am not sure your photos show me an IR leak. They don't even look white balanced to me. You need to white balance.

Shoot in RAW and white balance in Photo Ninja or whatever, do that first, before you think about the filters.

 

Rule of thumb, 2mm S8612 ~ BG40 4mm+ approximately. But no one really wants to use 4mm+ thick BG40.

Compare BG40 3mm and S8612 1.5mm, those work about the same for UV stacking, same for 4mm/2mm versions.

post-87-0-30143300-1549741691.jpg

Link to comment
I have rarely seen a leak with my 1.5mm S8612. It has happened but not usually. Probably better to do 1.75mm. I think 2mm is overkill.
Link to comment

I have rarely seen a leak with my 1.5mm S8612. It has happened but not usually. Probably better to do 1.75mm. I think 2mm is overkill.

 

There is very little active UV-transmission gained by selecting 1.75mm over 2.0mm.

2.0mm might be an overkill in many situations, but not all. It gives a better margin with almost no practical penalty.

 

With a S8612 2.0mm, compared to a 1.75mm, for an identical light situation and UV-pass filter in the stack, the penalty would be a less than 3% longer exposure time, but you gain five times margin fot the IR suppression.

There is no reason to not do an overkill.

 

The need for IR-suppression depends of the UV to IR ratio in the light source and the characteristics of the UV-pass filter in the stack.

In some situations 1.75mm might not be enough. 2.0mm will be enough in almost all situations and with many more filter stack combinations.

Link to comment

When I stack another 2mm ZWB1 filter on top I still get redness visible.

 

There is always a strong red tint in the raw UV-images. This is due to differences in sensitivity for UV in the different colour channels in the image sensor.

To get a more normal UV-mage you have to do a rather extreme white balance. Some cameras can do that, but not all.

You also get best results by working with RAW format and process the images on the computer instead.

 

Images taken through a window, a clouded day, wintertime at your latitude do not have much UV and might be difficult to use for judging UV-capability and IR-leakage.

You would need a very high IR-suppression for that light-situation.

Link to comment

Let's get back to white balance.

Let's not go any further with this topic discussion until the photos presented in post #1 (or new examples) are white balanced.

Those are not white balanced, and we can not make judgment on them until they are.

Link to comment

There is always a strong red tint in the raw UV-images. This is due to differences in sensitivity for UV in the different colour channels in the image sensor.

To get a more normal UV-mage you have to do a rather extreme white balance. Some cameras can do that, but not all.

You also get best results by working with RAW format and process the images on the computer instead.

 

Images taken through a window, a clouded day, wintertime at your latitude do not have much UV and might be difficult to use for judging UV-capability and IR-leakage.

You would need a very high IR-suppression for that light-situation.

 

I came up also with idea just to remove red channel and leave blue which more sensitive to UV anyway.

S8612 2mm would be also better for mechanical durability so better to have less UV but clean spectrum.

Link to comment

NO! You need to forget about filters for now, and everyone needs to forget about posting in this topic until you white balance the pics.

There is no discussion here until there is a real white balance.

 

Someone please tell him how to white balance before you ramble about filters. OK?

Link to comment

Compared to a ZWB1 + QB21 stack, a UG11 +s8612 or a U-360 + S8612 stack would bot be better, you would get much less exposure time, and other aspect would improve, but what is wrong here is mostly the white balance,

and THAT is what is needed first. Work on the white balance before you buy or think about filters.

Do that first.

Link to comment

Avalon, Before you buy another filter to remedy this, lets first white balance one of your UV shots.

If you don't have a RAW file of one of those, then shoot a shot again.

It would be nice if the shot is outdoors using natural light, even cloudy, use a tripod, not through any window glass.

No need for a visual shot, just a UV shot using your stack.

Shoot a RAW file, upload it somewhere I (we) can download it from, and white balance it.

Give me access and permission to do that, and I (we) can post my white balanced version here.

I can't say much about filters until I see a white balanced version of what you are using.

 

I have ZWB1, but not QB21, I have QB39.

Frankly I can't say which Schott filters either of those 'really' compare to, because the Chinese graphs are so contradictory and varied.

I don't trust any of their info.

 

However, I will say this from my limited experience with the Chinese filters I have tested for UV stacking.

The main difference you will see between a ZWB1 + QB39 stack and a UG11 + S8612 stack is the exposure time, thus what I call 'efficiency'.

Even if your QB suppression glass were as 'leaky' as BG38 (for example), you would still not see red, you would more likely see warm brown in what should normally be UV black areas that have some IR reflection,

such as the center of a Rudbeckia flower (the perfect UV test subject, by the way).

 

So please provide a UV RAW file, and then we can go from there.

Thanks.

Link to comment

Does this red left hand photo look familiar? It is ZWB1 + QB39, Aperture Priority, 4s, f11, White Balance is set to Daylight.

The second version on the right is white balanced from RAW using 'Marquee' in Nikon Capture NX2.

Although the exposure is 4s, this stack is much darker than the UG11 stack shot at 2.5s. The difference is 2/3 stop, but the ZWB1 stack needs more exposure time to compare with the UG11

stack, so I added +2 stops of exposure compensation (from RAW) to the ZWB1 stack to match the exposure to the UG11 stack.

So the right hand version is Marquee white balanced (from RAW), +2 stops exposure compensation (from RAW), then auto levels in Photoshop (using TIFF, which basically adds contrast).

(you can do the white balance etc. with Photo Ninja)

 

Like I said before, the main difference between the ZWB stack and the UG11 stack (other than price) is the exposure time,

I think that ads up to 2 + 2/3 stops, which can be a significant difference in exposure time.

 

Keep in mind, your UV shots look like they have a blue sky, my sample sky was cloudy, so your sky will look different.

 

White balance 'gets the red out'.

post-87-0-11044100-1549862511.jpg post-87-0-98525100-1549862515.jpg

Link to comment

I fully agree with Cadmium that white balance is the first step for evaluation of the images.

 

When that is done some filter facts can be useful.

QB21 is the Chinese "equivalent" to BG38.

QB39 is the Chinese "equivalent" to BG39.

Here "equivalent" can mean not as good as and with more variations in characteristics between different melts.

 

BG38 2mm do normally not surpress IR enough to be usable in a UV-pass stack.

It is very unlikely (ca 0%) that QB21 would be better.

 

BG39 2mm do normally surpress IR enough to be usable in a UV-pass stack but have some UV attenuation.

The superior S8612 that I suspect Cadmium used in his comparison above has no Chinese "equivalent glass.

S8612 is like a better version of BG39 with 20-25% higher transmission in the UV range interesting for us.

post-150-0-78564700-1549867713.png

Link to comment

Really? We are still talking about S8612 1.75mm compared to 2mm?

 

Avalon wants to learn to white balance himself. Which of course he must learn to do.

I can't explain to you how to white balance in camera with a Sony A3000 camera, I don't have one, and I have no experience with that camera.

If you want to do it in your camera, then you will need to grab someone else around here who has experience with some camera similar to yours for instructions.

 

If you want to do it in Photo Ninja, I can explain that, but that program costs money... however, if you wanted to test it, you can download a trial version:

https://www.picturec...om/download.php

I would be happy to post instructions about how to white balance with Ninja.

 

White balance is totally essential, not only for UV but also for IR. Without white balance you don't see the true content of your capture.

White balance is not usually used as much with visual, because stock cameras do a pretty good job of white balancing visual photos automatically.

White balance has to be done from RAW or in camera, and only some cameras are good with full spectrum in camera white balance, I certainly can't do it with my Nikon cameras,

but I think your A3000 would do it well, ask someone here who uses that camera.

Link to comment
Sony can white balance in-camera. Just use some asphalt to white balance off of, if you don’t have Teflon. I don’t have an A3000 and the white balance menu is in a different place on my NEX-7 as on the A7S.
Link to comment

Really? We are still talking about S8612 1.75mm compared to 2mm?

 

NO, not with one word.

 

I was just comparing different Chinese inferior filters with better ones.

Sorry if something in my text disturbed you.

I think that what I wrote was correct. Do you not agree?

 

However you added the QB39 to the mix in your own comparison in your post just above. That was confusing

As Avalon is using a QB21 not a QB39 I wanted to make the difference clear between QB21 and QB39.

 

I just left both thicknesses of S8612 in the diagram to show how much better the S8612 in both thicknesses are even compared to a proper BG39.

Link to comment

Oh, I thought you were still discussing the issue of S8612 thickness with Andy.

 

As I said, "I have ZWB1, but not QB21, I have QB39.", thus my example, not the same as his stack, but the only example I have using a Chinese glass UV stack.

The main point of my example was to highlight the difference between poor white balance and optimal white balance.

Sorry I confused you.

Yes, good to point out, yes, just looked it up, QB21 is suppose to be the equivalent to BG38, so there will probably be some warmth to a Rudbeckia center. ;-)

Never use BG38 for UV stacking. The red is not from that however.

 

There was a guy on eBay, in California I think, that sold ZWB(1 or 2 ?) and QB21 for UV stacking. He may still be selling, not sure.

Link to comment

if QB21 is similar to BG38 then there will probably be some warmth to your Rudbeckia center. ;-) But the red is not from that.

 

Agreed!

 

Wasn't confused, but thought others might be and just wanted to help and clarify.

However, I have never ever photographed any Rudbeckia. :(

Link to comment

Sorry I confused confusing with confused. :huh:

 

I can't find any Rudbeckia I shot with a U+BG38 stack, but here are some lawn daisies shot with 4 different BG stacks, including a BG38 stack, which illustrates the idea.

The center of the daisies should be black in UV only, but you can see a lot of warmth in the BG38 pic, the foliage also of course, anything that reflects IR.

post-87-0-13133100-1549953422.jpg

 

https://en.wikipedia...Bellis_perennis

Link to comment

The same flower-type, at close up, using a Baader U. This filter gives a similar colour rendering as the UG1 + S8612 stack.

 

The image is from the second day I attempted UV-photography and has a trace of light leakage from the finder ocular.

The leakage can be seen as a slight green hazy discolouration between the flowers. Now I have learned to keep the ocular covered.

 

post-150-0-31900200-1549965516.png

Link to comment

Please sign in to comment

You will be able to leave a comment after signing in



Sign In Now
×
×
  • Create New...