FED 2: can't make equal exposure times across frame for all shutter settings

[e_asphyx]

It's the second FED 2 I got and I still have the same issue: after a complete cleanup and oiling I can't adjust the shutter to expose the frame evenly across all settings. With a properly adjusted 1/25 setting the situation becomes progressively worse as the shutter speed increases. At 1/100 setting the leading frame edge gets almost a full stop less of an exposure. Replacing my usual Li grease with a thin oil doesn't change much. Over/undertensioning from the start (like 30 or 48 ms instead of 40 for "1/25" setting) doesn't help either.

I made a table to summarise the data. Step = (F+S)|n / (F+S)|n-1



Initially I suspected too thick lube in the timing mechanism (the image is taken from an old Leica manual but this part is almost identical) as it's basically the only part where two drums experience a mutual friction so the could potentially "help" each other but it's not the case.


I have a DIY shutter meter which consists of a photodiode and a photocurrent amplifier connected to an oscilloscope.

 

[TenEleven]

This usually means that the differential between the two drum tensions is wrong.
Unless I am severely mistaken there should be also two drums to tension for the shutter in the Fed-2.

If the leading edge gets more exposure then the second curtain is under-tensioned. The way I do this (for Canon LTM rangefinders, but again this should be analogue) is make sure that the shutter does not fade at any speed. And then bring the tension down for both drums simultaneously (e.g. 2 clicks first, then 2 clicks second) until the shutter runs at the right speed with minimal tension required.

Edit: I am however far from a knowledgeable camera tech, and my goal is just to keep my cameras working. Let's see if someone else knowledge can chime in.

 

[e_asphyx]

This usually means that the differential between the two drum tensions is wrong.
Unless I am severely mistaken there should be also two drums to tension for the shutter in the Fed-2.

If the leading edge gets more exposure then the second curtain is under-tensioned. The way I do this (for Canon LTM rangefinders, but again this should be analogue) is make sure that the shutter does not fade at any speed. And then bring the tension down for both drums simultaneously (e.g. 2 clicks first, then 2 clicks second) until the shutter runs at the right speed with minimal tension required.

Edit: I am however far from a knowledgeable camera tech, and my goal is just to keep my cameras working. Let's see if someone else knowledge can chime in.

What do you mean by fading here?

 

[rick oleson]

Fading is when the exposure is inconsistent from beginning to end of travel, basically what you’re describing. At the fastest speed you’ll have no exposure at all at one end of the frame.

Adjust this at the fastest speed. If it doesn’t fade there, exposure will be even at the slower speeds.

 

[Valkir1987]

The shutter curtains are interlinked at speeds above 1/30 (or 1/25, depending on the model). With some patience you can get the tolerance within 0.5 EV. The Fed-2 shutter differs from the Leica, where the curtains can move freely. The drag of the Fed-2 will not be noticeable on most films, unless you plan to shoot slides and low ISO films. A tolerance of 70% percent is even mentioned in the Zenit-E service manual. Equal exposure of possible, but if requires changing the drum sizes and repositioning the shutter curtains. You can read more about it here:

The fatal flaw in the Soviet Leica shutter

1/50 (or 1/60) and 1/500 can give almost equal exposure in my experience. 1/100 (1/125) and (1/250) are the 'worst'. If equal exposure is required you can avoid these speeds as a workaround.

 

[e_asphyx]

The shutter curtains are interlinked at speeds above 1/30 (or 1/25, depending on the model). With some patience you can get the tolerance within 0.5 EV. The Fed-2 shutter differs from the Leica, where the curtains can move freely. The drag of the Fed-2 will not be noticeable on most films, unless you plan to shoot slides and low ISO films. A tolerance of 70% percent is even mentioned in the Zenit-E service manual. Equal exposure of possible, but if requires changing the drum sizes and repositioning the shutter curtains. You can read more about it here:

The fatal flaw in the Soviet Leica shutter

1/50 (or 1/60) and 1/500 can give almost equal exposure in my experience. 1/100 (1/125) and (1/250) are the 'worst'. If equal exposure is required you can avoid these speeds as a workaround.

I don't get it. My understanding was that the travel speed of any curtain should depend on a spring tension only. The drum and rollers despite sharing the same axle shouldn't interact in ideal situation. Thus their diameters and angular speeds shouldn't matter.

 

[Valkir1987]

I don't get it. My understanding was that the travel speed of any curtain should depend on a spring tension only. The drum and rollers despite sharing the same axle shouldn't interact in ideal situation. Thus their diameters and angular speeds shouldn't matter.

If you wind the shutter and carefully hold the middle of the drum where the second curtain sits with your finger, you will notice the first curtain won't travel further than a set point if you release the shutter from speeds 1/100th and onward. I know this fact is counter intuitive, it took me a long time to figure out myself.

I can make a video to point out the differences between a Leica and a Fed shutter if you like on Friday.

Fed, Zorki and Zenit cameras are not the only ones to have interlinked shutter blinds, some Exaktas have it too.

 

[e_asphyx]

I can make a video to point out the differences between a Leica and a Fed shutter if you like on Friday.

Yes, please

 

[Ko.Fe.]

It is only complete if both drums are disassembled and gunk is washed out with camping fuel from springs, gun oil is applied.
Have you done so?
If not, the rest is just useless.

 

[e_asphyx]

It is only complete if both drums are disassembled and gunk is washed out with camping fuel from springs, gun oil is applied.
Have you done so?
If not, the rest is just useless.

Of course I did

 

[Valkir1987]

Here is an example, the shutter is set to 1/100th. The first blind won't travel any further when the second curtain is held is position. It drags the second curtain forward when it is slowly released.

 

[Ko.Fe.]

Of course I did

In this case, it might be re-assembly issue. I have to deal with light baffles to make sure curtains are moving freely.

 

[e_asphyx]

Here is an example, the shutter is set to 1/100th. The first blind won't travel any further when the second curtain is held is position. It drags the second curtain forward when it is slowly released.

I didn't expect this. While figuring out the shutter operation I took it apart completely multiple times but I still have no idea what part is responsible for this interlocking action.

Update: I got it. I haven't noticed that the selector indeed catches the latch tab!

 

[Brambling]

Please note - the diameters of the curtain drum and the tape pulley are different, the angular velocity on them is the same, but the linear velocity is not. There should be mutual compensation.
The thickness of the curtain and the ribbon cover is also not the same and is strictly defined.
the speed of movement of both the first and second curtains changes at the beginning and end of the movement relative to the frame window.
At the same time, the width of the gap between the curtains changes.
Several actions occur here at once; it is not entirely correct to measure only the speed of movement of a single curtain.

 

[e_asphyx]

Please note - the diameters of the curtain drum and the tape pulley are different, the angular velocity on them is the same, but the linear velocity is not. There should be mutual compensation.
The thickness of the curtain and the ribbon cover is also not the same and is strictly defined.

Yes, I thought it serves just one purpose, to keep curtains from rubbing against each other

 

[e_asphyx]

By the way I have a spare selector so I can run an experiment by filing down the catch leaving just a cam to allow it to pass above the latch tab

 

[Valkir1987]

By the way I have a spare selector so I can run an experiment by filing down the catch leaving just a cam to allow it to pass above the latch tab

It is an interesing experiment. I tried this before with a Zorki 4, but ended up having 1/300th as the highest speed. I you want to proceed with the original coupling, use a light lithium based lubricant (the grease) as you mentioned before. The shutter parts have a more loose fitting and were designed to be used with light grease. It also has a damping effect on the movement of the curtains. Tension the second curtain first. (the higher the tension, the more force on the release hook. Tension also increases friction and wear. It should close swiftly without too much force.
Then, adjust the first curtain and increase the tension until its exposure is equal at the highest speed, then check 1/100th and 1/200th accordingly. You might be able to get it within 60 or 50%.

 

[e_asphyx]

It is an interesing experiment. I tried this before with a Zorki 4, but ended up having 1/300th as the highest speed. I you want to proceed with the original coupling, use a light lithium based lubricant (the grease) as you mentioned before. The shutter parts have a more loose fitting and were designed to be used with light grease. It also has a damping effect on the movement of the curtains. Tension the second curtain first. (the higher the tension, the more force on the release hook. Tension also increases friction and wear. It should close swiftly without too much force.
Then, adjust the first curtain and increase the tension until its exposure is equal at the highest speed, then check 1/100th and 1/200th accordingly. You might be able to get it within 60 or 50%.

It worked! This is the selector part from FED 2.

And this is how it must be modified.

As a consequence "1/100" became about 17 ms (~1/60 sec) and "1/250" is now about 10 ms (1/100) but perfectly even. The height of the contact area between the selector cam and the latch is now less then 1 mm which may cause excessive wear. I may try to make a new slightly thicker latch from a brass stock and solder a brass shim to the selector itself.

Update: the scale 1/25 – 1/50 – 1/100 – 1/250 – 1/500 now turned into 1/25 – 1/35 – 1/60 – 1/100 – 1/150