1/2000th with a horizontal travel cloth FP shutter?

<p>I recently acquired a 1972 Fujica ST801. This has a horizontal cloth focal plane shutter, with a maximum speed of 1/2000th, and made me wonder how Fuji were able to achieve that.<br>

I thought such a high shutter speed had previously been unachievable with a horizontal travelling cloth focal plane shutters, due to their inability to withstand the shock of such acceleration, along with other timing and image distortion related technical troubles. The solution had been the use of a vertical travelling metal blade shutter, which was not only more robust, but moved along the shortest side of the film making faster speeds possible.<br>

So how did Fuji do it? I managed to find a French magazine advertisement that said the shutter was self-lubricating Teflon. Maybe that's the answer?<br>

Just wondered if anyone had any insight into this? I know some of you will consider this a "modern" camera, but it is over 40 years old.</p>

The flash synch speed of the ST801 is just 1/60, meaning the shutter doesn't travel that fast at all. Not faster than in any other focal plane shutter 35 mm camera offering 1/60 X-synch speed, no matter what the top speed may be.<br><br>The 1/2000 speed is achieved by reducing the distance between the first and second curtain, the slit width. That is done by controlling the release delay of the second curtain, and to do it succesfully it must be done with great enough precision and the travel speed (and increase thereof) of both shutter curtains must match closely.<br>So yes: timing is important. It helps to do that electronically.<br><br>FP shutter image distortion is a problem that cannot be solved mechanically except reducing the time the slit travels across the frame, so reducing the travel distance (vertical in stead of horizontal traverse) helps.<br>But this is only a problem with fast moving subjects.<br><br>Hasselblad's 2000 and (most) 200 series focal plane shutter cameras also have a 1/2000 top speed, with both metal foil (2000-series) and cloth shutter curtains (200-series) travelling quite a bit further than 35 mm camera's focal plane shutters. So these shutters reach a higher speed (roughly twice as fast). The old (1940s) 1000-series focal plane shutter Hasselblads had steel foil shutter curtains (top speed 1/1600 and later 1/1000), some of which eventually did indeed buckle due to the stress of coming to a sudden halt. The later corrugated titanium foil and plain cloth shutter curtains did not have those problems. So the in comparison small (makes the curtains stiffers, stronger), lighter (less inertia) and slower shutter in the Fuji should not have a problem with travelling at the speed it does.

<p>Nikon F2 (and later F3) had horizontally traveling shutters (albeit not cloth) that achieved 1/2000s (and 1/80s flash sync speed).</p>

<p>Rolleiflex SL66 did 1/1000s and 1/30s flash sync with a cloth shutter - pretty impressive for 1966.</p>

I may be wrong but I think the Speed Graphic shutters achieved 1/1000 sec (and traveled 5 inches).

No matter the shutter speed the shutter curtains always travel at the same velocity, The trick to shorter shutter speeds is

how precisely the gap between the first and second curtain is regulated.

Some day I am going to take my old Speed Graphic cloth curtain focal plane shutter out and take a photo of it laying out on the ground. Then people could see how a focal plane shutter works. The focal plane shutters in smaller cameras consist of two curtains that overlap to form different size slits. A Speed Graphic shutter consists of one long cloth, like a window shade, with different size slits cut in it along the way.<P>

 

Roll the curtain part way and a slit appears that is as wide as the film; roll the curtain some more and a 1-1/2 inch slit appears; roll it some more and a 3/8 inch slit appears and finally one more turn and a 3/32 inch slit appears. (All sizes measured approximately.)<P>

 

When released the curtain rolls up at a continuous speed. It is the size of the slit passing over the film that determines how much light and exposure the film gets.<P>

 

1-1/2 inch slit = 1/30 sec.<BR>

3/8 inch slit = 1/125 sec.<BR>

3/32 inch slit = 1/500 sec.<BR><div></div>

<p>The first Leicaflex - the one with the external meter - also had a 1/2000 top speed. Flash sync was 1/100 if I remember. How long a mechanically timed shutter would retain its accuracy at the top speed is another issue altogether. It was not uncommon for the 1/1000 speed to be a half stop slow. And it was another Fuji SLR where the top speed was 1/750.</p>

<p>My quarter-plate Ensign Reflex (about 1930) has 1/1000 second. It doesn't synch flash at all though.</p>

My explanation - the bit about the Hasselblads - has a flaw, an omission. I forgot to mention their X-synch speed.<br>As i mentioned, that shutter curtains travel faster can not be learned from the top speed, but by the X-synch speeds. The Hasselblad's 1/90 is slightly faster than the Fuji's 1/60. Combining this with the fact that in this time their curtains have to travel a greater distance to clear the film gate tells us that their curtains travel roughly twice as fast.

<p>I notice my OM-4Ti has 1/2000 with cloth curtains. This shutter is electronically timed though. Not a classic, quite modern and, at least officially, made up to 2002 although I doubt very few were made those last few of years of OM system production.</p>

<p>Thanks for the explanations - especially <strong>Q.G</strong>. I hadn't realised (with an "s" because I'm English) that the speed of the of the shutter is determined by the curtain gap, but it makes sense when you think about it. We hear a sort of cluck/click when a shutter trips, with the space between the clunk and the click being longer and shorted depending on the shutter's open time. I suppose the clunk is the first curtain, and the click the second.</p>

<p>I have never seen one myself but there was supposed to have been a Mamiya 2000 DTL with a cloth horizontally running shutter and a top speed of 1/2000. It would have been mechanically timed. The Olympus OM high speed flash synch was accomplished by having the electronic flash pulse as the slit traveled across the film plane. With a moving subject this could produce odd results. The set-up was suited more to macro/close-up work because the working distance was limited. Konica also made an early SLR with a top shutter speed of 1/2000 but that was with a vertically running metal shutter. My Canon F-1, F-1n, F-1N and Nikon F2 cameras all have 1/2000 top speeds but with horizontal titanium shutters. My Nikon FE2 has a top speed of 1/4000, my Pentax K-x DSLR goes up to 1/6000 and my Nikon N808, N90S and FR90X cameras all go up to 1/8000. These all have vertical shutters. </p>

<p>This is a good informative post....well done all. Thinking about top speeds, my Minolta AL has a leaf shutter with a 1000 top speed, or so it's marked, but they really struggle to achieve 1/750th...still impressive for a leaf shutter.</p>

<p>Never having taken apart a focal plane shutter, I always believed that there should be a coupling between the two curtains to be sure that they run at the same speed. Not just the release time, but some system to couple the speeds. If there is, it would be fairly easy to make the high speeds work, if not, not so easy.</p>

Again, my memory of the Speed Graphics is that on some models you could vary the curtain speed as well

as the slit width giving you some duplicate shutter speed settings and if the lens had a leaf shutter there

were 3 ways of getting some shutter speeds. I think the leaf shutter was generally preferred because it

could use standard flash bulbs at any speed and because it avoided the distortions of moving subjects.

<i>"I always believed that there should be a coupling between the two curtains to be sure that they run at the same speed."</i><br><br>Focal plane shutter curtains speed up all the time, from when they are released to when they have crossed the film gate (and come to a sudden stop). Having a fixed slit width (by coupling first and second curtains) would mean that with those accelerating curtains, the time the slit exposes the film decreases as it traverses the film gate at increasing velocity.<br>This increase in speed however is 'automagically' compensated when curtains are not coupled by the second curtain lagging behind, i.e. by allowing the slit to widen as the speed of the curtains increases. As long as both curtains show the same rate of acceleration, it will be fine.<br>Now if curtains could speed up to 'terminal velocity' before they uncover the film gate and then continue to traverse the film gate at that constant velocity, coupling the curtains, fixing the slit width would work. But that would require a quite intricate mechanism, both to couple the curtains and (most of all) to make sure the travel speed is constant across the frame.

<p>Very good topic and posts.</p>

<p>The accuracy of a shutter at it's top speeds is a design and manufacturing quality issue.</p>

<p>This post brought back memories of servicing the original Contax RTS and it's associated Yashica FR counterparts. As it turned out, all shared the same rubberized cloth focal plane shutter, even the lowly Yashica FR II, a dumbed-down auto only model made to occupy the bottom of the line. The first couple of those I serviced surprisingly tested out on the digital shutter speed tester at a reliable and consistent 1/2000 second with the light meter pegged to the top. Pretty good for the bottom of the line.</p>

<p>The higher end FR I did the same in auto mode. All of this was unadvertised, probably to avoid cannibalizing sales of the original RTS. Of course later models of the RTS had metal foil shutter curtains.</p>

<blockquote>

<p>"How long a mechanically timed shutter would retain its accuracy at the top speed is another issue altogether. It was not uncommon for the 1/1000 speed to be a half stop slow" <em><strong>Professor K.</strong></em></p>

</blockquote>

<p>True, but the <strong>big exception</strong> were the Leicaflex shutters. <br /> Those ultra high quality horizontal traveling rubberized silk curtain bad boys, seem to <strong>always</strong> be well within tolerance...</p>

<p>BTW: Pictured below is the amazing tank like, <em>Minolta V2 "Rocket"</em><br /> It achieved it's ultra high speed accomplishment by <strong>starting the shutter out</strong> at an effective f/8 opening (Red 'aperture' Zone <strong>2,000th</strong> & Blue 'aperture' Zone 1,000th).<br>

Read one of the Pnet threads about it: <a href="/classic-cameras-forum/00Tt3f">Minolta V2</a> <<< click>>> <a href="http://static.photo.net/attachments/bboard/00T/00Tthk-153123584.jpg">V2 Features</a></p><div></div>

<p>The exposure speed as Q. G. pointed out is a simple function of curtain opening and speed it is drawn across. As some articles point this can lead to distortions of length at lower speeds in verticle shutters and other distortions in horizontal shutters. Insignificant usually especially at higher speeds but it is possible to have differences from camera to camera. For static objects there should no difference. For moving subjects the result at 1/100 from camera A could be different from camera B. </p>

That distortion (for those who didn't know already) is caused by the frame being exposed in whatever time it takes for the slit to traverse it, one side first, the other side last. So the image of a subject moving fast enough will have moved noticeably between the time part of it is projected onto the first bit of film revealed by the slit and the time it is projected on the last bit of film of the frame.<br>Assuming the same subject and same focal length lens, the only difference regarding this distortion between two cameras is in how long it takes the slit to cross the frame. A camera with a slower shutter will give the subject more time to move than a camera with a faster shutter.<br>And that speed of the shutter is not the shutterspeed, which (as was said earlier) also depends on the width of the slit formed by the two curtains.<br><br>The shutterspeed however plays a role: too slow and you will not see the distortion because it is hidden in the blurred image recorded. So a good example of focal plane shutter distortion, in which the image of the moving subject is not blurred beyond recognition, but either stretched (when the image of the subject moves in the same direction as the curtains) or shortened (when the image moves in the opposite direction) or skewed (when the image moves in any other direction), requires both a relatively slow speed shutter and a fast shutterspeed.<br><br>If not only the subject moves but the camera too, panning to follow the moving subject, you get the <a href="https://phototechsf.wordpress.com/2012/01/24/race-car-by-jacques-henri-lartigue/">classic distorted image Lartigue captured of a car speeding across the frame along a road lined with what presumably are telegraph poles and some spectators. The inspiration for cartoonists' way of drawing fast moving cars and such</a>.

<p>While we are at it might as well mention that the above discussions are dealing with the math in a perfect world that assumes everything is moving at a constant speed. Of course the will be acceleration of the shutters in the beginning and probably deceleration at the end of travel and with a variation of size of curtain opening. This could lead to variations in exposure from one side to the other. Minimizing this effect is one aspect that would addressed by the quality and design of camera. Also if becomes a factor as the camera ages and what condition it is in. These parameters can also vary at any point in the shutter travel again based on quality, condition and maintainance.</p>

It had been mentioned that curtains do not travel at constant speed, Donald. ;-)<br>Also that the ongoing acceleration and resulting widening slit do not need minimizing or compensation by careful design and manufacture, because the latter compensates for the former automatically.

<p>Certain Fujicas such as the ST605 and ST605N had a marked top speed on their shutter dials of 1/700, not 1/750, as was previously stated.</p>

<p>I just went back and read the previous post and I follow that. Makes things easier. But as you said Q.G. that is assuming both curtains are properly tensioned. In the case of earlier Zorkis and Feds, the only ones I have worked on (I know they are not in the range of 1/2000), adjusting the tensioning is simple if not primitive. There are 2 spools that are tensioned independently. How well they were calibrated at the factory many times depended on what day they were made in the Soviet Union and over long period of time the performance can change</p>

<p>Lartigue once said in an interview (he was nearly ninety) that he knew exactly what he was doing, and that the result was carefully planned.</p>