Second Curtain Release -- Almost Random

[ic-racer]

I'm working with a stubborn camera in which the 1/1000 speeds are crazy fast and all over the place.

The camera has cloth horizontal focal plane shutter. A spinning cam on the first curtain gear disengages the second curtain, via the closing curtain latch, as the cam spins. It does this by pressing the cam follower into a pin on the closing curtain latch.

I'm trying to track down the cause of the problem.

We know that the faster the first curtain travels, the quicker the second curtain will release. So, I wanted to see if the crazy results were from randomness in the opening curtain speed.

So, I plotted the second curtain delay after first curtain release (should be 1000 microseconds) against the speed of the first curtain.

Correlation (R^2) was close to Zero, so, randomness in first curtain speed is not causing this. In fact the curtain speeds are pretty stable, clustered over an appropriate narrow range.

 

[Nicholas Lindan]

Well, from looking at the scatter of speeds, I would suspect either:

1. Gummed up lubricant (or, possibly, insufficient lubrication)
2. Loose part

Heck, those are the usual suspects in most shutter problems.

 

[ic-racer]

The camera has been disassembled cleaned and lubed. The curtains do have pretty consistent speeds.

Things I was thinking:

1) Yes, something loose
2) The second curtain latch, as is common, does not engage the second curtain spinning gear until the shutter button is depressed. Is there some interaction here? Does pressing the shutter button a certain way cause the latch to engage better or different (so it releases different). I checked and it seems the latch is always all the way over fully engaged by the time the first curtain is released.
3) The axle of the second curtain latch not free moving? I re-lubed this area with no change.
4) This is the best it is going to get with the camera design and aged components (give up on 1/1000, use 1 sec through 1/500.)

 

[ic-racer]

Here is another graphic representation of the data. This shows the slit widths at 1/1000. Again, all over the place.

 

[Andreas Thaler]

The scientific approach to DIY

 

[Nicholas Lindan]

4) This is the best it is going to get with the camera design and aged components (give up on 1/1000, use 1 sec through 1/500.)

I vote for #4. "Fix anything for long enough and you will really break it."

 

[4season]

What did you use when cleaning and lubricating the fast-moving parts, including shutter curtain drum, rollers, and mirror box?

 

[ic-racer]

What did you use when cleaning and lubricating the fast-moving parts, including shutter curtain drum, rollers, and mirror box?

To clean, I flush with denatured ethyl alcohol. I don't use isopropyl alcohol because it is not easy to find pure, without water mixed in.

After that, I use either actual Singer brand sewing machine oil, Triflon oil or RC car bearing oil. Which one? Pragmatic answer; each container has a different applicator. I choose the applicator to match the surface to be lubricated.

I use grease on the parts that bang together or slide against each other. I use either ceramic grease or moly.

This particular camera got cleaned and lubed then multiple times over a 4 week period. With the hope the timing would eventually even itself out. But, no, it didn't happen.

At this point, I have plenty of working cameras, I'm more interested in the answer to what is causing the random behavior. Everything has an answer, the challenge is finding it.

 

[4season]

Singer brand sewing machine oil

FWIW, several years ago I had a request from my mother to buy sewing machine oil, so I visited a specialist shop. I was specifically advised to discard any vintage green and white cans of Singer-brand oil I might have, and sold a replacement in a plastic bottle. Where the Singer oil had been amber-colored, the new oil was colorless. I don't remember the reason for the recommendation, but I do know that some lubricants have an expiration date, especially those for wristwatches.