Home

Please support mflenses.com if you need any graphic related work order it from us, click on above banner to order!

SearchSearch MemberlistMemberlist RegisterRegister ProfileProfile Log in to check your private messagesLog in to check your private messages Log inLog in

DIY dust seal in adapter for mirrorless camera
View previous topic :: View next topic  


PostPosted: Mon Feb 14, 2022 4:05 pm    Post subject: DIY dust seal in adapter for mirrorless camera Reply with quote

Folks,

Some 7 years ago,, I mounted a high quality coated clear filter inside the mount adapters for my mirrorless cameras, in order to seal the sensor from any dust whilst changing lenses (which I do a lot! Wink ). I share my experience with you in case it is of interest.

For those not interested in the details, let me first get to what matters most: my experience after 7 years use:

Dust elimination: Over the last seven years of fairly extensive use, with several hundreds of lens swaps at least, I have never had to clean my camera sensors, not once. I have also never had to spend any time cleaning dust bunnies in PP Smile
Aberrations: The expected increased aberrations have turned out to be totally unnoticeable in the vast majority of my photos; only where I use very fast apertures of f/1.2 and f/1.4 can I see some very minor extra aberrations. Do please bear in mind that I only use vintage lenses (mostly Rokkor), a lot of which already have some residual aberrations when shot at wide open apertures f/1.2 and f/1.4, so this will likely hide any issues in the first place.
Sensor reflections: The filter has pretty much been an unexpected and welcome irrelevance as far as sensor reflections are concerned. This is likely a result of the very effective B&W Nano AR coating on the glass I used.

So, for me (I have to stress that this isn’t an appropriate solution for everyone) this has been a highly successful modification.




Now the details for those interested:


The Problem:

As we all know, digital cameras with interchangeable lenses are prone to getting dust on the sensor, which builds op over time. It was a worry when I bought my first digital system camera (SONY A7S), and sure enough, despite all my best efforts (changing lenses away from the wind, lens mount facing down etc.), after my first outing I already had plenty dust bunnies covering my sensor, visible at smaller apertures. Sad

Now I don't have any patience for removing dust bunnies in PP, so a solution was imperative, otherwise I would have to stick with film.

I needed some way of covering the sensor at all times, so I started analysing the problem:

- Dust WILL enter a lens mount when changing lenses. This is not factually true, but having experienced that the usual suggested preventative measures don’t work well, I stipulated that as a fact in order to isolate the problem.
- I only shoot legacy lenses of one specific make (Minolta Rokkor) on a mirrorless camera, so I can leave the lens adapter on the camera at all times, which greatly reduces the problem. Whilst therefore my solution will not work for everyone all the time, it is perfect for me.
- Dust ON the sensor will be visible at smaller apertures, somewhat dependent on the size of the dust particles
- Dust some distance in front of the sensor will barely be visible, if at all, at any aperture


The Solution:

The obvious solution therefore would be to seal the lens adapter with an optically perfectly transparent seal of zero refractive power:

1) at some distance in front of the sensor, enough to render the inevitable dust well out of focus, but
2) with enough clearance behind the lens to allow for any protruding aperture levers etc.

This will therefore work for a mirrorless camera, but likely not for a DSLR.


This solution presents two potential serious issues:

1) Aberrations: An optical flat inserted in a perfectly converging beam of light, will introduce aberrations (spherical, chromatic, & astigmatism), a well known problem. This is also the reason that lenses designed for use with dedicated rear filters always need to have one of those fitted, as use without any filter will deteriorate the performance. This effect is more pronounced the greater the angle of convergence, i.e. the greater the lens aperture used (hence not much of a problem with filters used in front of the lens, as the incident light is near parallel).

2) Sensor reflections: Digital sensors are far more reflective than film ever was, so an optical flat inserted behind the lens is likely to increase issues of flare & ghosting. How much of an issue this would be was unclear to me, only one way to find out. In any case, in order to minimise this problem it is imperative that the optical flat has the most effective anti-reflection coating possible.


Candidates:

Candidates I considered were:

clingfilm (yes, that thought experiment did cross my mind! Smile ). Being very thin it eliminates the problem of increased aberrations, but it is optically far from flat, and would have a hopelessly poor uniformity of refractive index across its surface, resulting in a very poor IQ. Plus it can’t be coated.

Plastic filters (gelatine, polyester, resin). Thin, but no AR coating and prone to scratching. Also, they can potentially create issues with polarised light.

A glass anti-ghosting UV filter. This would likely reduce the problems of reflections of the sensor to some degree, but it wouldn’t eliminate it very much I suspected. Anti-ghosting filters aren’t optical flats, but rather concave-convex lenses of effective zero power, think of it as a uniformly curved sheet of glass of (almost) constant thickness. Mounted with the concave surface facing the sensor it would limit the on-axis additional aberrations a bit (off-axis possibly a bit worse), but likely not by much. However, the production of an optical flat is much easier than that of a concave-convex glass surface of zero refractive power, so I would expect IQ to be noticeably impacted. Plus it would probably impact on distortion (marginally). Finally, anti-ghosting filters are a bit of a fad from the past and ones with modern coatings I am not aware of.

A clear glass UV filter. This is readily available from top quality manufacturers, in a range of diameters, with highly effective modern AR coatings. This seemed the most promising option, but it was difficult to ascertain what would be the thinnest filters on the market at the time (often quoted is the thickness of the filter mount, but rarely is the glass thickness itself documented). Extensive research indicated the only ones that clearly stood out were the B&W XS-Pro Digital MRC Nano filters, with a glass thickness of only 1.3mm as opposed to the more usual 2mm thicknesses. These filters are manufactured by Schneider-Kreuznach, and should be pretty much on par with the likes of Heliopan filters (which is Zeiss, really), unfortunately also reflected in the price.


Observations on the filter thickness & sensor reflections:

So, how much would the image be affected by the presence of a 1.3mm thick optical flat inserted behind the lens? A few considerations came to mind here:

Focus register would be affected, depending on the exact refractive index of the glass. For a 1.3mm thickness an approximate 0.4mm register adjustment would be anticipated, which is fortunately easily achieved and fine-tuned by shimming the adapter lens mount.

Roger Cicala of Lensrentals had already published an article (links to parts I, II & III below) investigating the impact of the sensor stack thickness of various DSLR’s, when used with legacy lenses that were designed for use without any sensor stack at all.

https://wordpress.lensrentals.com/blog/2014/06/the-glass-in-the-path-sensor-stacks-and-adapted-lenses/
https://wordpress.lensrentals.com/blog/2014/06/sensor-stack-thickness-when-does-it-matter/
https://wordpress.lensrentals.com/blog/2014/07/sensor-stack-thickness-part-iii-the-summary/

The sensor stack (a.k.a. filter stack) referred to are the IR filter & anti-aliasing filter sandwich in front of the CCD / CMOS in most DSLRs and MILCs. Different camera makes employ quite different sensor stack thicknesses, and their concurrent lens lines take this thickness into account when correcting for aberrations. Particularly thick is the sensor stack of the Micro 4/3 cameras at a whopping 4mm! (I can understand why, having such a thick sensor stack would will reduce the effect of dust quite markedly even at smaller apertures, precisely the effect I was trying to achieve myself). Roger’s database showed my SONY A7 series cameras sensor stack thickness to be at about 2.5mm, which puts the additional 1.3mm anticipated into some perspective. Furthermore, in part III of his article Roger shows that for on-axis performance, at f/2 or smaller the impact will be difficult to notice even for a 4mm sensor stack thickness. For larger apertures however things deteriorate rapidly. Add to this the fact that Micro 4/3 users with their thick sensor stack seem quite happy to use legacy lenses, so I suspected I could probably live with the additional aberrations introduced by 1.3mm of glass.

Re. reflections, the AR nano coatings on the B&W filters are first class. Reflections of oblique rays towards the edge of the frame would never make it back to the sensor, but rather would be mostly absorbed by the walls of the lens adapter, thus I was expecting the residual effect of reflections to be a loss of contrast predominantly in the centre part of the frame, but the only way to find how much this would be noticeable in practice was to try. As I already alluded to, this turned out to be an almost non-issue.


Fitting the filters:

I have three high quality SONY NEX to Minolta SR adapters, one Novoflex, one Fotodiox Pro and one Rayqual. With a bit of plastic surgery the B&W filters fitted into the Fotodiox and Rayqual adapters in their original mounts, but to fit the filter into the Novoflex adapter I had to take the glass out of the B&W filter (not easy)! The Novoflex adapter had a partially cut internal thread to prevent internal light reflections, which turned out very useful as I could use some old lens mounting rings to secure the glass in place. I used a Mitutoyo Vernier depth-gauge to ensure perfect parallel alignment of the filter to 20 microns accuracy. The adapter lens mounts were shimmed and fine tuned to 20 microns levelness as well to regain infinity focus for all lenses.

Interestingly the B&W filters seem to come with coatings that either have a green or purple residual reflection, depending on either availability or exact type of filter (UV, clear, etc.). Ideally one would look at the colour of the residual reflection of the sensor, and match with a complementary colour coating on the filter. I.e. a sensor with a green refection should ideally be matched with a filter with a purple reflection, to minimise the overall secondary reflection.

The extracted filter glass fitted in the NOVOFLEX Minolta SR - SONY E adapter:



A complete filter was fitted in the FotoDiox Pro Minolta SR - SONY E adapter (see also the detailed photo at the start of this subject):



PostPosted: Mon Feb 14, 2022 5:24 pm    Post subject: Reply with quote

Thank you very much for the detailed and comprehensive topic.

Contrary to popular belief, living near the coast where air travels over 3000 miles of open ocean, then through a couple miles of wilderness forest, dust remains a major issue, even in winter time after feet of rainfall. Moments after removing a lens cap there is dust on the front element. Th sensor of my Canon 5D required repeated cleanings; blow dust from mirror box, clean sensor, lower mirror, mirror up, blow dust mirror flipped onto cleaned sensor and mirror box, repeat -- the first time I swear the procedure was repeated at least 20 times before getting mirror box and sensor to remain clean (I used 10x loupe to examine). The Sony A7RII doesn't seem to been as problematic as the 5D. Your DIY may be a solution for me.

I always try to aim the camera down when changing lenses; seems to help, not certain.


PostPosted: Mon Feb 14, 2022 9:00 pm    Post subject: Reply with quote

visualopsins wrote:
Th sensor of my Canon 5D required repeated cleanings; blow dust from mirror box, clean sensor, lower mirror, mirror up, blow dust mirror flipped onto cleaned sensor and mirror box, repeat -- the first time I swear the procedure was repeated at least 20 times before getting mirror box and sensor to remain clean (I used 10x loupe to examine).


That is exactly what I had no desire to get myself into, hence my "fix" to the problem.

It has the added advantage that although I still take care when changing lenses, I don't have to be worried about pointing the mount opening down or turning away from the wind etc. Plus I can use bellows and lenses with rear flocking without getting paranoid about dust or flocking fibres getting to the sensor. It does really work... Smile !


PostPosted: Sat Apr 23, 2022 3:07 pm    Post subject: Reply with quote

As a permanent option, would be to insert a filter behind the Sony mount body , but maybe not possible.
My camera needs sensor cleaning , and I haven't done it yet, I've read about scotch tape etc. but ,haven't found a good dyi step by step video . Probably it should be interesting to have a section for cleaning sensors also ,mine is the a7ii,


PostPosted: Sat Apr 23, 2022 4:23 pm    Post subject: Reply with quote

kiddo wrote:
As a permanent option, would be to insert a filter behind the Sony mount body , but maybe not possible.
My camera needs sensor cleaning , and I haven't done it yet, I've read about scotch tape etc. but ,haven't found a good dyi step by step video . Probably it should be interesting to have a section for cleaning sensors also ,mine is the a7ii,

STC and others make clip in filters that do just that.
I got one to reverse the IR conversion of my A7ii, but sadly I cracked it trying to refit it in a hurry.


PostPosted: Sat Apr 23, 2022 8:07 pm    Post subject: Reply with quote

DConvert wrote:
STC and others make clip in filters that do just that.
I got one to reverse the IR conversion of my A7ii, but sadly I cracked it trying to refit it in a hurry.


I did see those STC filters and I did consider them, but decided on the DIY route as described for three main reasons:

- They are primarily designed as optical filters and they don't form the good airtight seal I was looking for, especially when used with "air pumps" ( Wink ) like certain zooms and bellows
- On the SONY A7 FF cameras they (marginally) reduce the mount throat a bit, which is small enough as it is and as such would also act as an unwelcome marginal field stop for bellows work. Again, marginal but the corners of the sensor can already get a little obscured on the FF E-mount when used with very fast lenses or very forward exit pupils. One drawback of the E-mount for FF cameras unfortunately, it is really a bit on the narrow side Sad .
- I was looking to mount the filter further forward to reduce the ghosting effects in the centre of the frame when shooting high contrast at small apertures. The B+W MRC nano coatings are very good at limiting reflections, but still something I would further reduce the visible effects of when mounting the filter further away from the sensor (a more spread-out contrast impact rather than a more localised one). The STC filters sit around 4-5 times closer to the sensor compared to my DIY solution. But as I said, in practice this has turned out to be a non-issue for the vast majority of images.

But, for those not keen on some DIY and not shooting on bellows or with f/1.2 lenses, those STC filters could be a reasonable alternative. As long as you don't break them Wink

EDIT: how thin is the glass in those STC filters? I assume thick enough to still merit shimming the adapter to prevent any MF legacy lenses focusing a bit beyond infinity? (not an insurmountable problem if it does though, but I'm a bit OCD like that... Rolling Eyes )


PostPosted: Sun Apr 24, 2022 4:06 am    Post subject: Re: DIY dust seal in adapter for mirrorless camera Reply with quote

RokkorDoctor wrote:
This is also the reason that lenses designed for use with dedicated rear filters always need to have one of those fitted, as use without any filter will deteriorate the performance.


The manual for the Tamron 500-b is fairly clear on this matter- something I was not aware of with the sample I had bought sans filters. I had to look the manual up on line.
I suspect it was part of my infinity troubles with that particular lens.

I have no desire for a mirrorless camera at present.

-D.S.


PostPosted: Sun Apr 24, 2022 9:33 am    Post subject: Re: DIY dust seal in adapter for mirrorless camera Reply with quote

Doc Sharptail wrote:
I suspect it was part of my infinity troubles with that particular lens.


Yes, it wouldn't be able to reach infinity and that would be very noticeable on a catadioptric lens, especially as the core inside the circle of confusion is missing. Rear filter glass is generally around 2mm thick. That would give about 0.7mm displacement of the focal plane. On an f/8 mirror lens that would give a "doughnut of confusion" just under 0.1mm in diameter, nearly three times the size of the usual circle of confusion criteria for the 35mm film format.

Those rear filters really should be considered an essential lens element in the optical calculation, one of zero power but required for correction of various aberrations. Omitting a 2mm thick rear filter is not quite the same as, but also not too different from, replacing the rear element with one that is 2mm thinner. No one would doubt that noticeably would mess up performance Wink


PostPosted: Fri May 06, 2022 1:21 pm    Post subject: Reply with quote

Excellent post on an interesting DIY concept, Rokkor Doctor! This is the second post of yours that I've read, and I'm already a fan of yours!

By the way, yours truly is a Rokkor lover!


PostPosted: Sat May 07, 2022 10:30 am    Post subject: Reply with quote

Sakyaputta wrote:
Excellent post on an interesting DIY concept, Rokkor Doctor! This is the second post of yours that I've read, and I'm already a fan of yours!

By the way, yours truly is a Rokkor lover!


Welcome to mflenses!
Glad to see another Rokkor lover. You'll find there are quite a few of us on this forum Friends

You'll also find that there are many very well informed members on this forum that all bring different expertise to the mix. I'm learning a lot here myself!

Opinions may vary sometimes but discussions here are always friendly and there is a lot of respect between forum members; you will find this a pleasant place Wink


PostPosted: Sun May 08, 2022 5:11 am    Post subject: Reply with quote

RokkorDoctor wrote:

Welcome to mflenses!
Glad to see another Rokkor lover. You'll find there are quite a few of us on this forum Friends


Thank you very much, Mark! 🙏
I joined this forum, not long ago, to learn from everyone, and a lot has been gained already. Marvelous place this is! Like 1