Selenium Meter Repair

on the

Olympus Auto Eye II



Selenium is a non-metallic element which sits at number 34 on the Periodic Table.  In its vitreous form the material acts as a semiconductor, which is both photoconductive and photovoltaic.  This last property was initially harnessed in 1883 with the development of the first practical photovoltaic cell.   Although these cells put out very little power, they became the basis of the first electronic light meters, and gave photographers a simple and accurate tool for determining correct exposure values.   Because the selenium produces a current proportional to the light that's shining on it, the meters don't need batteries.  The meter circuits are very simple and dependable, typically consisting of only the selenium cell, an ammeter, and some resistors.  These advantages made selenium cell meters popular for many decades.

There is a persistent myth passed around on the web that selenium cells "wear out" with age.  Needless to say, any electromechanical device is liable to have problems after 50 years.  There can be various reasons why a selenium cell meter does not work, and a defective cell is only one of these.  The cells do not invariably deplete with age, like a battery.   But if the cell is left out in the light in a humid environment, the internal current will corrode the cell and increase the internal resistance.  This reduces the electric output of the cell.  If this has occurred, the only solution is to replace the cell.  Fortunately, this is relatively easy to do using readily available materials.

I came across the beautiful Olympus Auto Eye II shown in the pictures.  Everything on it was working great, with the exception of the meter.  The camera is capable of fully manual operation, but it just didn't seem right to have one of the very first Auto Exposure cameras ever made without the Auto Exposure part working.   The instructions below give a step by step process for checking the meter assembly of the camera and replacing the selenium cell.  Much of the information here may be useful in troubleshooting other types of selenium meters.  Only you can determine if you have the skills and ability to perform a repair of this type.  This author is not responsible for any damage you may do.


Tools you will need:

- Jeweler's screw drivers

- Needle nosed tweezers

- Hemostat

- Ice cube tray

- Spanner or similar tool (a pair of dividers works  well)

- Popsicle stick

- Soldering iron

- Q-tips

- Lens cleaning cloth

- Volt - Ohm - Meter


Material you will need:

- Silicon solar cell, 10mm x 25mm, .5V, 60mA, Edmund Scientific P/N 30373-37

- Aluminum sheet, .025" x .57" x 1.09"

- Electrically conductive epoxy

- Small gauge wire

- Solder

- Resistors

- Powdered graphite lubricant

- Nail polish

Note: All threads are right handed.




1. Remove the top cover from the camera.

Remove the film advance lever.  The screw that holds this in place has two small holes and can be removed with needle nosed tweezers.  Once the screw is removed, lift off the lever and the drive dog under it.  Note the assembly of dial, face-plate, brass shim and elastic ring (from top to bottom) for the film type reminder.  There is a ring underneath this which can be removed with a spanner.   Put all the pieces in the first cup of the ice cube tray.

Remove the film rewind crank by opening the back cover of the camera.  Put the popsicle stick into the fork of the rewind shaft, then unscrew the film rewind crank.  There are two small screws under the film rewind crank.   Remove the screws and put them and the crank in the second cup of the ice cube tray.

Lift off the top cover and set it aside.  The accessory shoe is a "cold shoe" - there is no electrical connection, so there is no wire to have to watch out for.


2. The viewfinder/rangefinder, meter assembly, and film advance mechanism are exposed.   The viewfinder/rangefinder can be cleaned after removing the cover over them, using the lens cleaning cloth over a Q-tip.  Take a look at the meter mechanism.

This is a type of "trapped needle" mechanism, except in this case the meter doesn't have an actual needle.  The meter armature carries the cam that sets the aperture opening.  There is an aperture stop that engages this cam.  The stop is carried on an threaded aperture adjustment rod.  When the shutter release is pressed down the aperture adjustment rod travels down with it.  This action opens the aperture.  The rod carries the aperture stop down with it until the stop engages the aperture cam; this sets the aperture to the correct opening.  There is a clip that gently presses against the aperture stop to hold it steady.  Even with this, the mechanism has some play in it, which can affect the repeatability of the meter reading by about 1/3 stop. 


3. Remove the meter assembly. 

First, remove the clip that rests against the aperture stop.  Then unscrew the nut that holds the aperture stop in place (the hemostats work well for this.).  Remove the aperture stop.  Remove the screw on the left side of the meter, and remove the threaded collar around the shutter release.  This can be removed just like a screw, by placing a screw driver in the slot of either side of the shutter release.  Put all the parts in the next cup of the ice cube tray.  Lift out the meter assembly.

4.  Test the meter assembly. 

There are several reasons that the meter may not be functioning properly.   The meter assembly is seated in a cast housing that serves as one of the contacts for the circuit.  A wire connecting the front of the of the cell to the resistors is the other contact.  Test the output of the cell, check for a broken or corroded contact in the circuit, and test the meter movement (the output of the ohm meter works very well for this).  For this camera, the selenium cell was the problem.  The cell slides into a housing with two spacers and the lens assembly.  Two spring clips hold everything tight and insure a good contact is made with the housing.  Finally, cut the wire where the cell attaches to the resistor


5.  Replace the cell. 

Not having any specs on the cell or meter, this requires some experimenting.  The first problem is that selenium cells are not commonly available.   Selenium has been replaced with silicon in most applications.  A modern silicon cell about the same size as the original selenium cell should put out as least as much, most likely more, current.  A search of the web turned up the one listed above, from Edmund Scientifics.  These cells are very fragile and need to be mounted on a conductive substrate.  Cut a piece of .025" thick aluminum the same size as the original selenium cell (.57" x 1.09") and mount the cell to it with conductive epoxy - use the epoxy sparingly and be very careful not to short the front and back surfaces of the silicon cell (epoxy can be cleaned up with denatured alcohol before it sets).  Attach a wire to the front of the cell with either the epoxy or solder (the epoxy is less likely to damage the cell).  The surfaces of the silicon cell have the same polarity as the original selenium cell, so the meter will turn the correct direction.

Make sure the wire is long enough to test the assembly.  An alligator clip on the end of the wire makes this easier.

Reassemble this into the meter housing.  Note that the half spacer will probably not be needed.


6.  Test the assembly. 

Connect the wire from the cell to the resistor where the original cell was connected.  Hold the cell up to a bright light source, one that reads about "sunny 16".  The meter armature should rotate just short of the entire length of the cam.  It doesn't have to be exact, as there is the threaded adjustment as well as the film speed setting.  The cell appeared to be reading just about right, so the wire is clipped and soldered into place.


7. Install the meter in the camera and check the exposure value against a known meter.  With the aperture stop adjusted to the end of its travel, this meter still read about 2/3 stop too far open.  That is pretty close, so before changing any of the circuit resistance I decided to set the film speed 2/3 stop fast and shoot a roll of film.  When the film came back, the exposure was pretty close to right on.  In fact, many people would consider it to be correct.  But I like a bit of a darker exposure for increased contrast and saturation.  It would be fine to just leave things at this point, and adjust the film speed 1 stop fast.  But "in for a penny, in for a pound".


8. Adjust the resistance of the circuit. 

This circuit needed slightly less resistance.  This could be done by replacing a resistor in the circuit, but this is not recommended.  Trying to resolder a new resistor where one has been cut out can be very difficult.  It is much safer and easier to add resistance in parallel, to decrease the overall resistance of the circuit.  It is simple to calculate the required parallel resistance value using the equation 1/Requivalent = 1/R1 + 1/R2 + 1/R3 ....  But since I had an assortment of resistors, it was easier to just test them in the circuit.  It turned out the two 47K ohm resistors in the circuit brought the meter into the correct range.


9.  Adjust the mechanical position of the aperture stop.   The stop must be held securely on the aperture adjustment, but if it is clamped too tightly between the adjusting nuts, it will bind and the meter will not read reliably.  If it is too loose, it will also tend to bind and give inaccurate readings.  There should be about 1/2 turn of play in the adjusting nuts for the aperture stop.  Since the nuts will be loose, apply a drop of nail polish to lock them in place - first to the bottom nut, then let the polish dry.  Reassemble the aperture stop and the top nut, recheck the aperture.  If everything is OK put a drop of nail polish on the top nut.  Reassemble everything and recheck the exposure.  If it all looks good, load up a roll of film and test it.


Replacing the selenium cell with a silicon cell presents a couple of additional issues that need to be kept in mind, besides the current output of the cell with respect to the meter movement.  One of these is the spectral response of the cell - that is, whether the two types of cells respond the same way to various wavelengths of light.  The other is whether the current vs. photon flux functions are the same, to make sure that the meter reads correctly at all light levels.   After shooting numerous frames at all light levels across meters range, the meter has shown itself to be very accurate. 

The camera does seem to be very susceptible to backlight.  This is most likely because the cell has a much larger sensing angle than the lens.  This situation might be slightly worse after this repair, since the silicon cell is slightly smaller than the original selenium cell.  But there is really not much difference in the size of the cells, so it is most likely that this camera always had problem with backlit conditions.