Press "Enter" to skip to content

Ground glass / focusing screen

A ground glass focusing screen is something I usually rush through because when it’s time to make one  I’m usually at the end of a tedious camera build and want to hurry up and use the thing already. But I wanted to take some time and make a decent one- after all this is the one thing I spend the most time looking at on a camera. Usually I don’t bother to seam the edges so the little chips along the cuts refract light into the image area and generally looks pretty crappy, so I spent the time to carefully cut the screen, smooth out the all the edges, and bevel them before getting on with grinding the surface. I did this for a recent Arca Swiss CLA, and it turned out well so made another one for my shop made camera.


Places that carry rock-tumbling supplies carry grit, or it can be ordered online from lapidary or ATM supply sites. 12 years ago I bought a half pound each of 5 different grits from and haven’t even begun to put a dent in it. I use 320 grit silicone carbide to seam edges and lap the glass flat (most glass isn’t dead flat, there’s a faint ripple pattern that shows up when ground with finer grits first).  In the foreground are synthetic diamond stones for quickly deburring edges so they are safer to handle. The work surface is plain 1/4″ plate glass, placed on a towel to keep it from sliding around during grinding. I like to set up next to a sink so I can rinse the screen often to check progress.

How fine to grind? It’s really a personal preference, and may depend on the focal lengths of the lenses and the focusing aids you use.  I decided to directly compare the grits when I made the Arca Swiss screen, so I cut a 5×7 piece of glass in two, ground one half to 5 micron, and the other half to 12 and then 9 micron. I compared them along the way by installing both halves in the camera with a 180mm lens and then focusing on a range of subjects both indoors and outside,  flipping the back 180º occasionally to make sure there was no dominant eye bias skewing the appraisal. I use 3x reading glasses mostly for composing and focusing, then switch if needed to a 4x loupe for more critical focusing. Under these conditions each grit had its benefits but overall I liked the 9 µm the best, which gave a good balance of illumination and grain. The illumination was slightly dimmer than the 5 µm, but more even across the screen, with no obvious hot spot in the center. The 5 µm screen was pretty bright in the center, but quite dim off axis. I should have compared a range of lenses, but 180mm is by far my most used lens on the format. Just for laughs I ground one half to 3 micron, but didn’t like it at all- at this point, the glass is essentially starting to get clear again.



Before and after edges. Much easier to just have the local glass shop cut and seam a few screen blanks for you, but if like me you’re bored and don’t feel like driving to town then 320 grit silicone carbide works pretty well, but it does take a while. I put a 1/8th teaspoon of grit on a piece of flat 1/4″ plate glass, add a few drops of water, and grind the screen edges in a random sawing motion across the plate glass work surface,  parallel to the cut edge. Dragging the glass perpendicular to the edge can cause  chip-out. I first deburr the glass with synthetic diamond stones and of course use gloves while doing all this, the edges are surgically sharp after cutting. I also slightly bevel the ground surface side of the screen during this step, which makes getting an even grind around the perimeters of the screen a little easier.


I mask off the side I want to keep clear by placing it on some paper with a few drops of water to keep it from sliding around. This keeps the slurry from wicking underneath and scratching the other surface. An 1/8th tsp. of git is placed on the ground side of the screen with a few drops of water (eye droppers are nice for adding tiny amounts of water to the slurry as it evaporates). For a grinding tool I use a flat bottomed element from an old zoom lens, it’s been dropped a few times but still works well enough. I usually start by lapping the bottom of the lens element on the plate glass with the 320 grit  just to make sure it has a good flat surface.  For this glass I started with the 25 μm, and it does take a  while longer to get the surface even with grit this fine. 320 silicone carbide is a little quicker but does tend to leave tiny craters that can be time consuming to remove with finer grits. I ground in 3 steps from 25 to 12 to 9 µm. Obviously it’s important not to contaminate the grit with coarser grit, so I clean the teaspoon before moving from bag to bag. After the screen is fully ground, I put a little more grit on the plate glass work surface and finish the edges and bevels a little more with 25 µm oxide.


For a grid, a quick CAD sketch is printed on an inkjet printer, placed under the glass on a lightbox, then scribed into the ground side of the glass using a T-square and diamond burr. Many different things will work- I’ve used a stainless steel rivet pin and even a cheap ballpoint pen. This time I tried a fine-point diamond burr from a rotary tool bit set. It does help to have a piece of scrap ground screen to practice on- some tools will really chip the glass a lot along the scribe line, others are so fine it’s difficult to see the grid at all. Somewhere between the two is ideal. A little chip-out is needed to make the grid light up so it’s easy to see. I’ve tried using a really fine point scribe line then doubling over that with a fine-point red Sharpie, but the ink doesn’t last very long and tends to bleed into the ground surface of the screen. I’ve also tried just printing the grid on clear overhead projector media and attaching that to the ground glass, but there is about half a stop light loss through the extra material.