The Bronica S2A is my favourite medium format camera. It is also the camera that has given me the most trouble over the last six months. This is the third video in a series of focusing problems and the one where I finally worked out what was actually going on.
The short version is: focusing screens have to sit at exactly the right distance from the back of the lens for the image on the ground glass to match the image on the film. The “exactly” in that sentence is doing a lot of work. We are talking about tolerances of fractions of a millimetre. When I tried to upgrade the screen in my S2A to a brighter aftermarket version (a Brightscreen by Rick Oleson, which is a very well-regarded product) I learned the hard way that 1mm of plastic in the wrong place can put your focus 30 to 60 metres out at infinity.
The long version is below. Worth reading if you have any vintage camera with a ground glass focusing screen (which is most SLRs and all TLRs) because the lesson is universal even if my specific symptoms were Bronica-flavoured.
Why I was tinkering with the focusing screen at all
A bit of background. I have had my Bronica S2A for about a year and a half by this point. It is my most-shot camera and my favourite camera, and also the one I trust the least. Almost from day one I have had intermittent issues with focusing not being where the ground glass said it should be. Sometimes frames came back sharper than I expected. Sometimes softer. The pattern was random enough that I could not initially work out whether the camera was the problem or whether I was the problem.
About six months before this video I had replaced the foam pad that holds the ground glass in place against the waist-level finder. That improved things. Frames started coming back better. But not perfect. Some shots were still off, and I was getting the classic “cannot reach infinity” symptom on landscape frames. The ground glass said the lens was focused at infinity but the negative showed the trees in the background as soft.
The decision I made was to upgrade the focusing screen itself to a Brightscreen by Rick Oleson, which is a well-regarded aftermarket plastic screen with a fresnel lens and a microprism focusing aid built in. Rick (working with Lance, who handles shipping) runs the operation out of the US at bright-screen.com. The S2A screen costs about $80 plus shipping, which is really reasonable for a quality-of-life upgrade on a camera I use every week. I ordered one. While it was in transit I took the existing setup out for what I assumed would be its last outing.
Last shoot on the old screen
My friend Andy Ham is the host of Two Chaps Motoring, a super fun YouTube channel covering affordable classic and modern cars. He has a 1972 MG Midget called Margot, which is the kind of vehicle that makes you reconsider whether sensible cars are worth bothering with at all. The MG Midget is one of the British roadster icons (produced from 1961 to 1979, in this case the Mark III variant with the 1275cc engine and the wire wheels) and Margot is a particularly clean example.
Andy agreed to bring Margot down to the Severn for a morning shoot. I took the S2A with the existing screen still in it, three rolls of FP4 Plus and the standard 75mm Nikkor PC kit lens.
The shoot itself was lovely. Andy is patient with the slow workflow of a manual medium format SLR and Margot is such a photogenic car. We did some shots of the car against the Severn estuary, some shots of Andy leaning on the bonnet and one frame where Andy sat on a stool with Margot framed behind him. The light was decent. The compositions were considered. The film advance worked correctly. The shutter fired cleanly. Everything was working except the focus, which came back roughly right but not quite right on about half the frames. Classic S2A behaviour. Sharp where I had not been trying for sharp, soft where I had wanted sharp.
I will get those frames printed and put them up somewhere. The best of the set is a half-length of Andy with Margot’s hood pushed down behind him. It is so nice as a portrait even with the focus slightly off.
Installing the Rick Oleson Brightscreen
The Brightscreen arrived a few days after the Margot shoot. The screen is about the size and shape of the original S2A ground glass but is different in construction.
The original Bronica setup is two separate pieces stacked on top of each other: a fresnel lens at the bottom (the embossed concentric rings spread the light evenly across the ground glass so you do not get a bright spot in the middle and dim corners) and the ground glass itself on top (which is the matte surface that the image forms on). Together they sit in a frame held down by a foam pad and clipped in by the waist-level finder above.
The Brightscreen replaces both pieces with a single sheet of plastic about 1.2mm thick. The fresnel is etched into one side. The ground-glass matte surface is on the other side, with the microprism focusing aid in the middle. So you go from two pieces of glass to one piece of plastic, which is the source of the trouble that came next.
While I had the camera open I also replaced the foam underneath the mirror, which I had not done before. This is a routine maintenance job on these cameras. The foam absorbs vibration and prevents light leaks, but it deteriorates over decades and turns into the gummy black residue that fingerprints itself onto everything. A small bottle of isopropyl alcohol and a cocktail stick gets the old foam off. New self-adhesive 1mm closed-cell foam goes back on. Fifteen minutes if you are careful.
Rick’s instructions say to install the Brightscreen with the fresnel side up, towards the waist-level finder. The original Bronica configuration is fresnel side down, towards the mirror. So Rick is asking you to put the screen in upside down relative to what the camera was built around. There is a reason for this (the optical path is slightly different with a combined screen versus a two-piece stack) but it confused me at first.
I installed it fresnel up as instructed. I cleaned everything. I reassembled the camera. I loaded a roll of FP4 Plus.
The first test shoot. Disaster.
A few days later I had a paying commercial shoot booked. I was not planning to use the S2A as the primary camera but I took it along to fire off a few extra frames as a bonus for the client. I rushed those frames. I metered, focused, fired, advanced, repeated. Twelve frames in maybe five minutes.
The results were bad. None of the frames was sharp. The focus was consistently off (back-focused this time, which is the opposite problem to what I had been having before with the original screen). I initially blamed myself for rushing the shots but the pattern was too consistent for that to be the explanation.
I needed a real test in conditions where I was not rushed and could compare against a known reference.
The real test shoot at Chepstow
I drove to Chepstow with a fresh roll of FP4 Plus and the camera. Chepstow is where I had done the last shoot with the original screen, so the location was familiar and the subject matter (the river, the bridge, the castle ruins) was the kind of distant landscape that would tell me immediately whether the camera was reaching infinity properly.
The test was simple. For each scene I would shoot two frames. One with the focus set to where the ground glass said was correct. One with the focus set to the infinity mark on the lens itself (the engraved ∞ symbol on the focus barrel). If the camera was calibrated correctly, these two settings should produce identical frames. If they were different, I would know the screen position was wrong.
The frames came back showing exactly that disagreement. The infinity-mark frames were sharp at infinity. The ground-glass-says-infinity frames were soft. The camera was overshooting infinity according to the ground glass. To get an object at 100 metres in focus, I needed to focus past the infinity mark on the lens, which is obviously impossible because the infinity mark is the end of the focusing range.
The diagnosis was clear. The new screen was sitting in the wrong position relative to the film plane. The image was forming below the matte surface of the screen, so by the time I had brought it into focus on the screen I had overshot the actual focal plane on the film.
The tolerance discovery
Here is the bit that genuinely surprised me, and the thing I wanted to share in the video more than anything else.
The Rick Oleson screen is about 1.2mm thick. The original Bronica setup has the fresnel on one side and the ground glass on the other, with the matte surface (the surface where the image actually forms) about 1mm from the bottom of the stack. Rick’s design has the matte surface at the top of the stack (when installed fresnel-up) which means the matte surface is about 1mm higher than the original setup.
That 1mm difference in screen position translates to a focus error at the film plane that, at infinity focus, equates to roughly 30 to 60 metres of subject distance. In other words, if your ground glass is 1mm too high, the camera will think it is focused at infinity when the actual point of sharp focus is around 100 metres away. If your ground glass is 1mm too low, the camera will think it is focused at infinity when it has overshot infinity and the actual point of sharp focus is past where the lens can physically focus to.
I flipped the screen over (against Rick’s instructions) to see if installing it fresnel down would put the matte surface back at the original Bronica position. That created a different problem. The screen was now too far down, putting the matte surface lower than the original ground glass position. Now I had the original problem back: the camera could not reach infinity at all.
I had effectively confirmed that the Bronica was built to put the matte surface at a very specific position, and the Brightscreen (in either orientation) put the matte surface either slightly above or slightly below that position. Neither orientation matched the camera’s expectations.
The fix
I put the original Bronica screen back in. I did not run a roll of film through it; I just took the back off the camera, opened it in a darkroom, focused on infinity through the ground glass with the bellows extended and looked at the actual image forming on the empty film plane. Even with the original screen back in, the camera was still not reaching infinity properly.
The only remaining variable was the foam under the waist-level finder. The foam holds the ground glass and fresnel stack down against the camera body, and if the foam is compressed, the stack sits slightly lower than it should. I had replaced this foam six months earlier but it had clearly already compressed and settled. I replaced it again with fresh foam, reassembled the camera and tested infinity focus through the ground glass.
This time the camera reached infinity correctly. The original Bronica screen, properly seated on fresh foam, gave me accurate focus from the ground glass.
The final verification shoot with Tom
My friend Tom (the one who modelled for the Viking shoot piece on the 4x5 a few months later, but at this point still just my friend Tom) came along to help me verify the fix. I gave the camera every chance to be as sharp as possible. Tripod. f/4 at 1/250. Twelve frames of Tom in various poses around a field and against a wall.
The frames came back sharp. The focus on the ground glass was now telling the truth. The camera was fixed.
The best frame was Tom holding a small ice cube against his nose because I had asked him to look “pretentious”, which is a direction Tom takes very seriously. It is a bit of an absurd portrait that I am genuinely fond of.
What I learned
A few things stood out from this whole exercise, none of which I had really understood before I started.
Camera tolerances are minuscule. The S2A was built in the early 1970s by a Japanese company that had to hit fractional-millimetre tolerances on the ground glass position for the camera to work at all. Half a century later, the foam underneath the screen is compressing by amounts you cannot see with the naked eye, and that compression is enough to make the camera unusable. Every vintage camera you own has the same issue lurking in it somewhere. When you knock a camera and the focus goes off, this is usually why.
Aftermarket screens are not a magic upgrade. The Brightscreen is such a beautiful piece of engineering and works brilliantly in cameras that were designed around its thickness. The S2A was not. Other reviewers on the Bright-Screen forums note the same thing about the Bronica S series. The screen works but you may need to shim or adjust the camera to compensate for the different stack height. I should have read more carefully before I ordered.
The original Bronica screen is fine. Once it is sitting at the correct height on fresh foam, the original Bronica ground glass is still usable. It is not as bright as the Brightscreen would be, but it is bright enough, and crucially it is in the correct position relative to the film plane. The S2A was designed around this exact screen and works best with it.
Control is the point. I sometimes get comments on the channel asking why I get so worked up about focus. The answer is that I am not chasing perfect sharpness in every frame. Some frames look better with slightly soft focus. What I want is the choice. If the camera is showing me focus on the ground glass and that focus does not match what hits the film, I have lost the choice. The decision about whether a frame is sharp or soft is being made by random foam compression and screen-stack height, not by me. That is what I object to.
A properly calibrated camera that puts focus where I want it gives me back the choice. That is what I have now.
A note on Rick Oleson’s product
I do not want this article to read as a criticism of the Brightscreen. It is a really well-made product that has improved countless TLRs and SLRs for photographers all over the world. The reviews on the bright-screen.com site are enthusiastic for a reason. The fact that it did not work in my S2A is partly down to the stack-height issue and partly down to me not understanding the implications of the different stack design before I ordered.
If you have a Bronica S2 or S2A and you are tempted by the Brightscreen, my honest recommendation is to first try replacing the foam under the original screen and seeing if that fixes whatever issue you are having. If it does, save your $80 and stick with the original. If it does not, the Brightscreen is worth trying with the awareness that you may need to shim it or compensate elsewhere.
For Rolleis, Hasselblads, Mamiya RBs, Pentax 6x7s and other systems where the camera was designed around interchangeable screens or where the original screen has yellowed beyond use, the Brightscreen is a really great option. Rick’s customer service is excellent and the screens are made in the US to high standards.
What’s next
The S2A is back in service and behaving again. The next video on the channel is going to be a GB Kershaw 110 review which is the polar opposite of an S2A in every way (a £5 fixed-focus folder from 1954 made out of bakelite) but which has reminded me that not every camera needs to be calibrated to fractional-millimetre tolerances to make really nice frames.
Big thanks to Andy for bringing Margot down to the Severn for the last-shoot-on-the-old-screen session. The MG Midget is a photogenic car and Andy is the kind of friend who agrees to drive an hour for a friend’s camera test. His channel is at Two Chaps Motoring and is worth a subscribe if classic and affordable cars are your thing.
And to Tom for the verification shoot. Tom gets paid in beer.