Custom ND project

[rob norton]

I have two of the Nisi Athenas in RF mount https://nisifilters.com.au/shop/nisi-athena-cinema-lenses/rf-mount/

You can see they have drop in rear filter options https://nisifilters.com.au/shop/nisi-athena-cinema-lenses/athena-drop-in-filters/

Their vari-ND is 1-5 stops and their solid NDs only go up to 7 stops. As @cyvideo will tell you, this isn't enough off the shelf filtration for the Australian sun. On my Canon C200s I very regularly use 8 stops with the sigma 18-35 @f/1.8 - f/2.2, but almost never use 10 stops. IMO Nisi's ND lineup isn't complete, particularly for those who want to take "advantage" of T1.9 in bright conditions with now almost unavoidable full frame cameras. It feels like a shame to need additional front filters when the drop in filters are so handy.

Each lens comes with a clear filter because something has to live there to block the slit where the filters go.

So, why not just commit a clear filter and larger solid ND to make a custom 8 stop filter?

Questions:

I believe I'd be able to cut an appropriate size disc from a larger ND, but since it needs to fit in the NiSi housing, I'm assuming thickness is also a big consideration. I don't think many people are asking about filter "depth". Do you think it's just a matter of asking NiSi how thick their drop in NDs are and trying to find a match? As an aside, are filters thicker the denser the ND? They probably can't deviate too much.

Say I cut the perfect size circle that is an appropriate thickness, is it then just a matter of disconnecting the filter holder, remvoing the clear filter, then glueing the new one in place? Or is it better to avoid glue and see if some micro screws etc. could hold it in place?

Is this a dumb idea?

Thanks everyone and have a great day!

 

[DustinSchmidt]

I have two of the Nisi Athenas in RF mount https://nisifilters.com.au/shop/nisi-athena-cinema-lenses/rf-mount/

You can see they have drop in rear filter options https://nisifilters.com.au/shop/nisi-athena-cinema-lenses/athena-drop-in-filters/

Their vari-ND is 1-5 stops and their solid NDs only go up to 7 stops. As @cyvideo will tell you, this isn't enough off the shelf filtration for the Australian sun. On my Canon C200s I very regularly use 8 stops with the sigma 18-35 @f/1.8 - f/2.2, but almost never use 10 stops. IMO Nisi's ND lineup isn't complete, particularly for those who want to take "advantage" of T1.9 in bright conditions with now almost unavoidable full frame cameras. It feels like a shame to need additional front filters when the drop in filters are so handy.

Each lens comes with a clear filter because something has to live there to block the slit where the filters go.

So, why not just commit a clear filter and larger solid ND to make a custom 8 stop filter?

Questions:

I believe I'd be able to cut an appropriate size disc from a larger ND, but since it needs to fit in the NiSi housing, I'm assuming thickness is also a big consideration. I don't think many people are asking about filter "depth". Do you think it's just a matter of asking NiSi how thick their drop in NDs are and trying to find a match? As an aside, are filters thicker the denser the ND? They probably can't deviate too much.

Say I cut the perfect size circle that is an appropriate thickness, is it then just a matter of disconnecting the filter holder, remvoing the clear filter, then glueing the new one in place? Or is it better to avoid glue and see if some micro screws etc. could hold it in place?

Is this a dumb idea?

Thanks everyone and have a great day!

Ha, I love the initiative, but to answer your question, I think this is a dumb idea.

Filtration behind the lens does need to be of a certain thickness or you'll run into issues with focus. Also, you're creating a problem where one doesn't exist really.

Just purchase and place an ND on the front of the lens if you need 8-10 stops. You can get an 8 or 10 stop ND for not very much cash and your problem is solved.

 

[rob norton]

Thanks Dustin!

I know with Canon when going 8-10 stops it says to check focus, but using the athenas/drop in filters with DJI lidar, it doesn't matter if it's clear/7 stops.

Is this just an extra consideration when moving beyond a certain density?

At the moment I have a smallrig mattebox (discontinued), where the matte box attaches to a 95mm diameter, which is provided by an adapter ring, threaded into the 77mm of the athena lens. A screw in filter between the lens and matte box would definitely work, it just wouldn't be as fast as the drop in system.

Just purchase and place an ND on the front of the lens if you need 8-10 stops. You can get an 8 or 10 stop ND for not very much cash and your problem is solved.

The "just...place" in your sentence is the part that sort of ignores reality. Pulling drop in filters in and out is quite good, but still pretty annoying. Adding a matte box removal to the equation is worse?

I forgot to add, the suggestion from NiSi is to add a 4 stop filter in front when using the 1-5 stop vari for example, giving a 1-5 then 5-9 when needed.

 

[cyvideo]

I have two of the Nisi Athenas in RF mount https://nisifilters.com.au/shop/nisi-athena-cinema-lenses/rf-mount/

You can see they have drop in rear filter options https://nisifilters.com.au/shop/nisi-athena-cinema-lenses/athena-drop-in-filters/

Their vari-ND is 1-5 stops and their solid NDs only go up to 7 stops. As @cyvideo will tell you, this isn't enough off the shelf filtration for the Australian sun. On my Canon C200s I very regularly use 8 stops with the sigma 18-35 @f/1.8 - f/2.2, but almost never use 10 stops. IMO Nisi's ND lineup isn't complete, particularly for those who want to take "advantage" of T1.9 in bright conditions with now almost unavoidable full frame cameras. It feels like a shame to need additional front filters when the drop in filters are so handy.

Each lens comes with a clear filter because something has to live there to block the slit where the filters go.

So, why not just commit a clear filter and larger solid ND to make a custom 8 stop filter?

Questions:

I believe I'd be able to cut an appropriate size disc from a larger ND, but since it needs to fit in the NiSi housing, I'm assuming thickness is also a big consideration. I don't think many people are asking about filter "depth". Do you think it's just a matter of asking NiSi how thick their drop in NDs are and trying to find a match? As an aside, are filters thicker the denser the ND? They probably can't deviate too much.

Say I cut the perfect size circle that is an appropriate thickness, is it then just a matter of disconnecting the filter holder, remvoing the clear filter, then glueing the new one in place? Or is it better to avoid glue and see if some micro screws etc. could hold it in place?

Is this a dumb idea?

Thanks everyone and have a great day!

Quite often I will use a 1 to 2 ND on the front of the camera, on the FX6 and FX3 especially. Your fast lenses will still retain their DOF characteristics, but you will just effectively have a slower camera. Which I haven't found to be a hindrance in extreme sunlight. Always remember, any filtration placed behind the lens will have a refractive index impact on the light path going through to the sensor.

There is a focus shift of about one third of the thickness of the filter. The exact value depends on the refractive index of the filter glass, but if you focus after fitting the filter, the on axis focus point can be found accurately. But this is only 100% true for manual lenses. With most modern cameras, especially those using phase detection AF with native lenses, the lens communicates with the camera body via a lens file stored in the lens to the firmware in the camera. Should you be able to shoehorn a glass filter in behind a native lens that calculates its focus based on the communication from the lens, it can throw the focusing calculations out, as a few have found out. Suffice to say, a 3 mm filter can change the back focus by about 1/3rd, that's 1 mm. That is a large amount.

That is why broadcast 2/3" ENG cameras with built-in ND filters, which are solid glass, and with later cameras with electronic NDs, like we find in the FX7, FX9 and FX6 family, these cameras still run a clear glass filter element when no NDs are engaged. To maintain correct back focus and maintain the full functionality of the lens correction circuits that most cameras and lenses rely on these days.

The other issues that can occur and that will not be corrected for by the camera's lens compensation circuits is a rear filter imposes a bit of spherical aberration on the native image of the lens. This is because increasingly oblique rays effectively traverse an increasing thickness of filter glass. This results in an increasing focus shift for each image ray as it becomes more oblique toward the edge of the format. The effect is most severe for wide angle lenses. However, stopping well down may create enough depth of focus to mask the problem. As always, test thoroughly.

Chris Young

 

[rob norton]

Quite often I will use a 1 to 2 ND on the front of the camera, on the FX6 and FX3 especially. Your fast lenses will still retain their DOF characteristics, but you will just effectively have a slower camera. Which I haven't found to be a hindrance in extreme sunlight. Always remember, any filtration placed behind the lens will have a refractive index impact on the light path going through to the sensor.

There is a focus shift of about one third of the thickness of the filter. The exact value depends on the refractive index of the filter glass, but if you focus after fitting the filter, the on axis focus point can be found accurately. But this is only 100% true for manual lenses. With most modern cameras, especially those using phase detection AF with native lenses, the lens communicates with the camera body via a lens file stored in the lens to the firmware in the camera. Should you be able to shoehorn a glass filter in behind a native lens that calculates its focus based on the communication from the lens, it can throw the focusing calculations out, as a few have found out. Suffice to say, a 3 mm filter can change the back focus by about 1/3rd, that's 1 mm. That is a large amount.

That is why broadcast 2/3" ENG cameras with built-in ND filters, which are solid glass, and with later cameras with electronic NDs, like we find in the FX7, FX9 and FX6 family, these cameras still run a clear glass filter element when no NDs are engaged. To maintain correct back focus and maintain the full functionality of the lens correction circuits that most cameras and lenses rely on these days.

The other issues that can occur and that will not be corrected for by the camera's lens compensation circuits is a rear filter imposes a bit of spherical aberration on the native image of the lens. This is because increasingly oblique rays effectively traverse an increasing thickness of filter glass. This results in an increasing focus shift for each image ray as it becomes more oblique toward the edge of the format. The effect is most severe for wide angle lenses. However, stopping well down may create enough depth of focus to mask the problem. As always, test thoroughly.

Chris Young

Thanks Chris, I knew you'd be the right person to comment about this topic.

I've never appreciated the communication the Sony cameras are doing in terms of AF coordinating with the filtration. It sounds like the clear filter is sort of an unsung hero as a piece that allows the whole system to work as you described.

This is all a bit over my head, but what's happening with something like a lidar system (DJI Focus pro)? Is it better to keep the clear filter in place and use NDs in front of the lens, but the drop in filters would be better used with a 1st AC pulling focus to avoid the refractive index impact you mention?

I shot a project moving between clear/0.6/1.2/2.1 ND filters and am pretty sure the focus was bang on. But for another project the focus was slightly off, except I don't know if the issue you're referring to was showing itself there or my calibration was done incorrectly (was on an island using a box of cereal as my chart to calibrate critical focus yikes!).

 

[cyvideo]

The right person? Don't know about that, but you are welcome to pick my brains for any useful tit bis I may have gathered along the way. A box of cereal. Luv that!

You still have a refraction effect with front mounted filters, but the path through the filter is effectively in a straight line and all rays are virtually parallel. So the impact of refraction is extremely minimal. Once the rays start getting redirected by the lens elements and start to converge, is where the accuracy of those converging light rays becomes very important. That is why, with film and rear mounting filters, we always used Kodak Wratten filters. They are 0.10 mm thick. This is a massive difference to the 30 x difference in thickness to a 3 mm rear glass filter. At 0.10 mm you can get away without too much impact from refraction as the refraction is in the order of 0.03 mm. That's 3/100th of a millimetre. A big difference to 1.0 mm with a 3.0 mm rear drop in filter. Also remember the wider the lens the worse the refraction will become as the converging of the light rays is pretty extreme compared to let's say a 600 mm lens which has a much gentler convergence path to the focus point.

As I said before. I've never had issues with 100% manual focusing, unless I have got it wrong on my part in focusing. But I have definitely seen issues with Sony E series lenses in AF mode when testing the idea of behind the lens optical. Which you can shoehorn into some of Sony's E Mount lenses.

Now, I haven't used these Kase filters. But if I needed to use any rear ND filtering on my A7 type cameras, or Canon cameras for that matter. This is what I am pretty sure I would use as they are supposed to have a very low refractive index, and they are made from SCHOTT 270 optical glass, reputed to be the best, and are only 0.3 mm thick. So you end up with a 0.1 mm refractive deviation. The fact that they sit right in front of the sensor also minimises refractive angles. They are very popular with Astro shooters because of their performance. Often used on long Milky Way exposures.

Chris Young

Some info here:

SCHOTT

https://www.schott.com/en-au/products/b-270-p1000313/technical-details

A practical demonstration of the Kase filters:

Wratten

https://www.edmundoptics.com.au/f/kodak-wratten-filters/11429/

 

[rob norton]

The right person? Don't know about that, but you are welcome to pick my brains for any useful tit bis I may have gathered along the way. A box of cereal. Luv that!

You still have a refraction effect with front mounted filters, but the path through the filter is effectively in a straight line and all rays are virtually parallel. So the impact of refraction is extremely minimal. Once the rays start getting redirected by the lens elements and start to converge, is where the accuracy of those converging light rays becomes very important. That is why, with film and rear mounting filters, we always used Kodak Wratten filters. They are 0.10 mm thick. This is a massive difference to the 30 x difference in thickness to a 3 mm rear glass filter. At 0.10 mm you can get away without too much impact from refraction as the refraction is in the order of 0.03 mm. That's 3/100th of a millimetre. A big difference to 1.0 mm with a 3.0 mm rear drop in filter. Also remember the wider the lens the worse the refraction will become as the converging of the light rays is pretty extreme compared to let's say a 600 mm lens which has a much gentler convergence path to the focus point.

As I said before. I've never had issues with 100% manual focusing, unless I have got it wrong on my part in focusing. But I have definitely seen issues with Sony E series lenses in AF mode when testing the idea of behind the lens optical. Which you can shoehorn into some of Sony's E Mount lenses.

Now, I haven't used these Kase filters. But if I needed to use any rear ND filtering on my A7 type cameras, or Canon cameras for that matter. This is what I am pretty sure I would use as they are supposed to have a very low refractive index, and they are made from SCHOTT 270 optical glass, reputed to be the best, and are only 0.3 mm thick. So you end up with a 0.1 mm refractive deviation. The fact that they sit right in front of the sensor also minimises refractive angles. They are very popular with Astro shooters because of their performance. Often used on long Milky Way exposures.

Chris Young

Oh sorry, I missed the manual lens part but that makes sense. Thanks for the links and those Kase filters look very cool.

With the 100% manual focus athena lenses (/with lidar in the mix too), it sounds like my DIY project might still be worth at least trying, if only for fun. Just need to find the right 2.4/8 stop donor filter.