Kolari UV bandpass in front of lens or behind?

[moondigger]

I plan to purchase the Kolari UV bandpass filter for my first attempt at UV photography. I'm wondering if anybody has advice regarding whether it's better to buy a lens filter or utilize the drop-in filter for the EF-EOS R lens adapter?

 

I'm leaning toward the drop-in filter, because it would cover use of any EF-mount or adapted-to-EF mount lens, rather than having to buy filters in multiple sizes, or one large lens filter and a bunch of step-up/step-down rings. But I wonder if there's a technical reason for favoring a lens filter over the drop-in filter, or if there are other reasons to avoid the drop-in filter. I'd hate to get one and then find out that (hypothetically) light leaks are an issue or something like that.

 

Thanks -- 

[ulf]

From an optical point of view one thing happens when you place a filter or optical window behind a lens like with a drop-in filter:

The the focal point is shifted a bit, making you focus beyond infinity.

 

That might help a lens working at UV reaching infinity if that was a problem due to that it was not doing so as it was designed for VIS only.

 

The strength of the effect is dependent on the FL of the lens.

For a tele lens, barely noticeable, as the extension for focussing closer is rather long.

A Wide angle lens will be affected more by this phenomenon.

 

I think the Kolari UV-pass filter is rather thick, making this effect noticeable.

 

Quality lenses designed for film might lose sharpness in the corners, especially if your camera's conversion was done by replacing the internal filters with an optical window of equal optical thickness. Then you have the combined effect of that window and the UV-Pass filter.

 

This is a phenomenon is even reported when using some Leica WA-lenses on normal digital cameras, as those lenses, designed for film, on a digital camera suddenly see the internal filter stack.

I do not remember the exact lens but the phenomenon is real and can be explained by optical design theories. Could be found by searching on the web.

 

On the plus side, if you keep the drop-in filter in place when switching lenses it will block passage for dust reaching the sensor.

Mirrorless cameras are much more sensitive to that.

That can be extremely disturbing with macro, stacking and high magnifications, especially if the conversion type was by removing the internal filters and no replacement window.

 

As a side note the stated blocking of non UV wavelengths is not that good for UV-Pass Kolari filters.

Sometimes it is good enough, sometimes it is not.

They claim a minimum of OD3.5. (Typically around 700nm where the sensor is close to maximum sensitive)

If you have lower UV levels like me at 55° N , especially in the spring, that might not be enough.

 

I even saw a hint of IR-leakage with my Baader U (measured to OD 4.2) on a EL-Nikkor 80mm old metal type, with good UV-performance.

The UV-signature of a flower, that should be pitch black, was not quite black.

 

The sensitivity for such kind of leakage increase when the lens is more UV-marginal as the ratio between UV and IR increase. 

 

Hope you have some use for my answer and please do not hesitate to ask if something was unclear or too technical. I am happy to answer or clarify.

Sorry for the long answer.

[moondigger]

ulf -- No need to apologize! I appreciate your detailed answer. 

 

Having spent a few weeks here lurking before I registered, I was able to gather a lot of information I didn't previously know. However I don't have the hands-on experience with UV imaging to understand all the details I've read up to this point.

 

I think I have most of the basics covered to begin my UV-imaging education. My EOS R has been converted to full spectrum, and is on its way back to me. I don't have a dedicated UV lens, but I already own a Canon EF 40/2.8 pancake lens, which I understand will suffice for a beginner. I just need to obtain my first UV bandpass filter. I'm leaning toward the Kolari because it's an easy/convenient option for me, and the performance (based on samples I've seen) seems adequate. I live near 43° N, so UV levels should be a bit better than you describe at 55° N. Would a 40 mm lens like the Canon pancake be strongly affected by the shifted infinity focus you describe, if I use the drop-in filter?

 

[ulf]

39 minutes ago, moondigger said:

I think I have most of the basics covered to begin my UV-imaging education. My EOS R has been converted to full spectrum, and is on its way back to me. I don't have a dedicated UV lens, but I already own a Canon EF 40/2.8 pancake lens, which I understand will suffice for a beginner. I just need to obtain my first UV bandpass filter. I'm leaning toward the Kolari because it's an easy/convenient option for me, and the performance (based on samples I've seen) seems adequate. I live near 43° N, so UV levels should be a bit better than you describe at 55° N. Would a 40 mm lens like the Canon pancake be strongly affected by the shifted infinity focus you describe, if I use the drop-in filter?

 

I do not think the Canon EF 40/2.8 pancake lens would see any problem with a rear mounted filter at all. The AF of the camera will often be able to function too. The rear filter gives more margin as it shift the maximum focus beyond infinity.

 

The IR leakage issue is only visible when you have strong IR reflectance at the same time as the UV-reflectance is close to zero, typically for some flowers with really dark UV signature.

 

If you like UV-photography and want to expand into macro I recommend getting the EL-Nikkor 80mm old metal type and a suitable helicoid.

The image quality from those enlarger lenses are really good

I posted how I made that work 5-5 years ago. If you search the forum you can see how I did.

 

General thoughts about filters:

If you want to test more filter combinations like a BUG stack..., then Kolaris rear mounted filter system will not be suitable and front mounted filters will be more flexible.

I standardised to 52mm filters for UV related photos and 77mm filters for IR related photos. Lately

I have complemented each group crosswise.

When playing around testing filters better quality from china makes such explorations not that expensive.

 

[DonPilou]

Hi,

Some practical feedback here: I use the UV bandpass filter from Kolari Vision in drop in mount with a converted Canon R5C, compared with a front filter there is no visible difference of quality with lenses from 20mm to 200mm on the 45Mpx sensor. The drop in adapter is designed to take into account the thickness of filters placed between the lens and the sensor, so there will be no noticeable focus shift. 

If you want to test another UV bandpass filters, you can also purchase one of their empty drop in mount and place a 52mm filter in it.

[ulf]

15 minutes ago, DonPilou said:

The drop in adapter is designed to take into account the thickness of filters placed between the lens and the sensor, so there will be no noticeable focus shift. 

Do you mean that all drop-in filters made by Kolari are parafocal, meaning that they have identical optical path length?

If not, then how noticeable depends very much on in what situation they are used.

This is mostly important for close up and you can rarely just swap filters without refocussing.

Do you really have 20mm lenses that are good for UV-Photography?

[DonPilou]

2 minutes ago, ulf said:

Do you mean that all drop-in filters made by Kolari are parafocal, meaning that they have identical optical path length?

If not, then how noticeable depends very much on in what situation they are used.

This is mostly important for close up and you can rarely just swap filters without refocussing.

Do you really have 20mm lenses that are good for UV-Photography?

 

No, of course an infrared filter made of a single layer of glass is less thick than an UV bandpass filter made of 2 layers, but the shift is less important than if the converter was not calibrated to take into account an average drop in filter. The minium focusing distance will not be impacted significantly in real use. And I do not see any problem checking the focus distance when I change filter, whatever it's mount.

I use a Nikkor AF-D 20mm F/2.8 for wide UV photography, it's transmission is not at the same level of EL lenses, but it has great homogeneity , can be used handheld and has enough UV transmission to achieve coloured post processing.

 

[ulf]

23 minutes ago, DonPilou said:

If you want to test another UV bandpass filters, you can also purchase one of their empty drop in mount and place a 52mm filter in it.

That is correct, as long as the filter is not too thick. a remounted Baader U would fit though, but is a bit more expensive. The Baader U is also much thinner that the Kolari UV-Pass filter and defenitely not parafocal

 

A non dichroic UV-pass filter normally consists of a stack of two filters stacked together a BG glass like S8612 and a UG1 or UG 11 or similar such a stack is likely too thick to fit.

 

If using a separate BG-glass you can combine with several other filter types for interesting imaging effects

[ulf]

2 minutes ago, DonPilou said:

The minium focusing distance will not be impacted significantly in real use. And I do not see any problem checking the focus distance when I change filter, whatever it's mount.

I was not referring to the minimum focussing distance. A change there, if any would be quite difficult to detect.
 

When describing the optical effects above I mostly do not see them a problems. I am just trying to describe what happens and where things change the most.

Your Nikkor AF-D 20mm F/2.8 is a modern lens designed for digital cameras with the thickness of the internal filter stack.

 

What I wrote about above was the effect in images from lenses designed for film, without the added filters in a modern digital camera.

Using such lenses can in some situations have problem with corner sharpness. You cannot say that just because your Nikkor AF-D 20mm F/2.8 work well with rear filtering that all 20mm lenses will do that! That is simply not true. 

[ulf]

14 minutes ago, DonPilou said:

 

No, of course an infrared filter made of a single layer of glass is less thick than an UV bandpass filter made of 2 layers, but the shift is less important than if the converter was not calibrated to take into account an average drop in filter.

As long as the adapter is short enough to allow focus  at infinity without any filter at all there is no problem adding more glass in the light path for the ability to reach infinity.

An approximation of the shortening of the back flange distance by adding a glass window or filter is ca 1/3 of the filter material's thickness.

 

If you add glass behind the lens the distance information from the lens, recorded in the exif information will be a bit off. It will be more off for a WA lens than for a Tele lens.

[DonPilou]

My intervention was motivated by giving @moondigger my practical experience of UV photography with a drop in filter (which is, in my point of view, a real game changer).

I honestly do not have time to spend 1h per message to give all the data and take into account all the theorical side effects of all lenses or alternative filters. I know what has been said in the many quotes of my answers, I simply do not see why I would have had to expose all these data for a simple ask about practical experience. 

 

So to summarize for the asker: from my experience with lenses offering acceptable to very good results from 20mm to 200mm, drop in mount has many advantages from a photographic aspect. Maybe not from a scientific aspect, but it seems that you are looking to take real pictures, not to stay in a lab.

[moondigger]

I want to thank everybody for the detailed replies and advice. I have decided to go with the Kolari drop-in filter for now. Knowing my personal history with tools or technology that's new to me, this drop-in filter probably won't be my only option for very long. But it's a place to start. 

 

Thank you!

[ulf]

I think that is a good decision as a starting point, Russ!

 

I hope that you understood that I am not opposing a rear filter mount, just trying to explain all potential pitfalls and technical effects just as you asked for in your first post in this thread.

I really like the idea of rear mounting of filters.
For my mirrorless camera I started a now dormant project to create such an adapter that can handle really thick filter stacks and filters. the goal was to be able to handle a filter stack with two filters or a really thick 8mm ZWB2.

 

Eventually you might want to try using front mounted filters too. 

 

As a bonus from other comments in this thread we eventually learned that a reasonably modern Nikkor AF-D 20mm F/2.8 performed rather well and even worked for some upper UV-A.

That can be good enough in some situations.

Deep UV reach is not always a must. It can be better to focus on enough UV reach and good image quality. 

I am sorry that I seams to have offended DonPilou that seams to not understand why I try to clarify the technical details.

 

When learning about this rather technical type of photography you can avoid much confusion if you understand what happens technically from scientific point of view.

Only then you can fully take advantage of the possibilities and take great pictures in the real world. 

For me that is the absolute goal, even if I am not as talented as some professional photographers like DonPilou.

 

FYI, He have some business relation to Kolari that sells his packages of LUTs and also have a French friend that has designed the Kolari IR Chrome.

Not saying that he must be biased, but at least have more than just a relation with them as a satisfied customer.
He appears to be happy with the adapter.

I hope you will be content wit yours too when you get it.

[moondigger]

ulf, I understand the reasons for your detailed answers, and I appreciate them. I'm sure I'll find the information useful as I learn more about UV photography.

 

[ulf]

Please do not hesitate to ask in the future.

It is also OK to message me separate if you prefer that in some situation.

 

[cazza132]

I use the EF to RF filter slot adaptor with the Kolari UV bandpass filter and it works great! I have a full spectrum converted Canon R5.

The added advantage is that any yellow fluorescence that can occur in some lenses is also filtered out by default. This yellow fluorescence can be a problem with my Baader U (Venus) when front filtered on lens.

All of Kolari's slot in filters are parfocal except for the UV bandpass and IR Chrome filters as these filters are slightly thicker. This has never been a problem for me though. The non-parfocal issue only really shows up with wide angle lenses and wide apertures. 

[DonPilou]

10 hours ago, ulf said:

FYI, He have some business relation to Kolari that sells his packages of LUTs and also have a French friend that has designed the Kolari IR Chrome.

Not saying that he must be biased, but at least have more than just a relation with them as a satisfied customer.
He appears to be happy with the adapter.

I hope you will be content wit yours too when you get it.

 

Well, I mean, this kind of stupid allusion is exactly why I was out of this forum for so much time... For your information, I used home made drop in filters way before Kolari Vision decided to launch their own. Sorry that this is the only company offering this kind of mount for IR and UV bandpass filters for general customers. 

 

If I really wanted to promote Kolari Vision here, there are tons of other posts dealing with their products where I could have defended them. As you can see, this is not the case.

 

That's done for me.