I've been working more on the Ektra when time permits. I already have it together and have shot a couple of test rolls through it (soon to come) but this is a recap of the repair and refurbishment. Fortunately, the 1943 Ektra repair manual is available though Pacific Rim: (warning - 500MB dowload)
https://www.pacificrimcamera.com/rl/01346/01346.pdf
Cleaning the optics and aligning the rangefinder were fairly uneventful other then being tedious. I disassembled the lens to clean the old grease off the helical and re-lubricate. It's much smoother now and collimating the focus with the body was quite easy.
The optics in the top cover were also cleaned, this was somewhat tiring as the Ektra has five glass optical lenses which make up the variofocal viewfinder and the rangefinder has five prisms and six lenses and my ADHD won't let me leave a single speck of dust visible in any of them. adjusting the rangefinder is easy with one screw doing halving and another doing the horizontal alignment.
The tricky part has been the Ektra's shutter, mainly understanding the mechanism, since it is similar in some ways to other shutters but it has key differences. I am not aware of any other resource for understanding the shutter either, so other than the above repair manual (which is largely text based and has few diagrams) I am flying into uncharted territory.
What we know is that the Ektra uses a three drum shutter, much like a Leica, but differs mainly from a Leica shutter in setting the gap between the curtains. On the Leica, the gap is made by delaying the release of the trailing curtain. On the Ektra, the gap is made on the curtain drum before any curtains release.
So on the Leica, the chain of events is that the shutter is wound with the curtains overlapping, then they stop. The leading curtain is released, then there is a delay and the second curtain is released creating the gap where they both run to the resting position. In this design each curtain is tensioned separately and the tension on each curtain determines the curtain velocity.
On the Ektra, there is a gap between the curtains as they rest. When winding begins, the leading curtain needs to pick up immediately and close the gap with the trailing curtain so that the film in the film plane is not exposed as the shutter is wound. Once the leading curtain clears the film aperture the leading curtain stops moving. The trailing curtain continues to move and wind around the center of the curtain drum setting the gap. The gap is 0.030" at 1/1000 and about 1.375" at 1/25. The shutter is thus wound and the sear catch releases both curtains simultaneously, running back onto their rollers. In this design, each curtain roller can be tensioned, but the roller drum rigidly couples the two curtains and the differing inner and outer diameters of the drum regulate the curtains so that the gap starts out narrow and widens as the curtains accelerate to give an even exposure.
The Extra's operation necessitates a complex shutter drum, and this is evident in the following diagram. The center portion which holds the narrow trailing curtain is rigidly coupled to the winding mechanism and speed throw-out. The upper roller for the wide leading curtain is coupled to the center portion of the drum and thus the winding mechanism by two different mechanisms simultaneously, there's a backlash spring on the center spindle which allows the wide curtain to stop and the narrow curtain to continue to wind, but not change the spacing when released. There is also a toothed gear and pawl which limits rotation in one direction only until the drum rotates and centrifugal force pushes the pawl outwards which allows the gap to change once the leading curtain is fully on it's roller, closing the gap and allowing the shutter to blackout when rewound. The lower roller for the wide leading curtain is coupled to the center drum by means of a spring which allows the roller to rotate independently but under spring tension.
The manual indicates the Ektra shutter went though at least one revision, as it lists a "early" and "late" version when it was published in 1943 with the changes being how the lower roller is coupled to the drum and the late version having an additional ribbon roller. FWIW, my camera is SN 3724 and has an early style shutter. Serial numbers are reported to have started at 1000 and production was estimated at 2000 to 2400 examples. Make of that what you will.
Understanding the operation of the shutter I still had to take it apart to clean the bearings and thus remove the curtains to do so. Lots more intricate work. Cleaning the bearings necessitated removing tension of the rollers, so I notated the number of turns on each roller for later, but preliminary settings are also listed in the repair manual.
With the shutter cleaned and reassembled the next question was calibration. The repair manual lists tolerances for each shutter speed, but with both roller tensions, the curtain slit width and the speed control throw-out all having control over the shutter speed and being interactive controls, the need to measure the effective shutter speed and its evenness is imperative.
Kodak referenced the "Ektra shutter speed tester" in the repair manual. I have little other details on it, but the tolerance list would suggest that it output a singular or average shutter speed in milliseconds, and I believe it to be the apparatus shown below:
I don't have such an apparatus at my disposal, and likely none exist at this point in time. In the past, I have used a photo-diode and bright light to output a signal to a program like Audacity to measure shutter speed. This is fine for a leaf shutter, but a focal plane shutter requires at least two measurements to ensure that the shutter is traveling evenly over the film plane. Some modern focal plane shutter testers use 3 to 5 sensors, but professional models like those made by Kyoritsu can cost hundreds to thousands of dollars.
With that in mind, I built my own shutter tester to test the Ektra's shutter speeds. It uses lasers and two photo-transistors biased as photo-diodes. The sensors are mounted about 30MM apart giving an effective speed based on gap width and velocity at the start and end of the shutter's travel. The key attribute is that each sensor is under an aperture of only 0.030" which is the narrowest gap of the shutter, which gives a point estimation of shutter speed. The signals are output though a stereo connector to my computer running Audacity which allows me to compare the signals and given the specs of my sound card, This allows me to resolve the shutter speed down to 0.1 ms with an accuracy of +/- 0.02 ms (1/1000 is 1 ms).
With this in hand, I've been able to calibrate the Ektra's shutter to its original tolerances and reassemble the camera. I've already shot some test rolls, including a roll of Provia 100F to confirm even shutter travel and will develop and scan tomorrow.
