Help with Kodak Flash Supermatic Shutter?

[Denverdad]

I am in the process of overhauling a Kodak Monitor Six-20 with a Flash Supermatic shutter, trying to get the camera ready for 620 camera day which is coming up in a little over a week. Today I was so pleased to finally get the shutter back together after a long and rather tortuous CLA (torturous because it is a new shutter to me and I don't yet have a manual). Everything seemed to be looking and sounding great, and in fact could tell that the slow speeds were right on the money. But unfortunately my shutter speed test showed that the fast speeds are slow, falling behind by as much as 1-1/3 stops at the top end. Any suggestions? Is there something which is perhaps not engaging properly, or maybe some adjustment I have missed?

For reference, here are my numbers, and also a plot showing the error in stops (measurement resolution is 1/6 stop):

Thanks.

Jeff

 

[Tom1956]

Yeah, typical. I don't think they were right for more than a week after they left the factory. Annoying, isn't it? That high speed spring isn't worth a damn. What we have here is wear introducing just enough slop that parts shift laterally in their movements, and additively they foul up everything. Springs get weak, etc. Good luck with that.

 

[Denverdad]

Well that's not encouraging to hear!
Do I at least understand correctly that the high speeds on this shutter result from the pallet being disengaged from the low speed escapement, but with the escapement gears still engaged? That's what it looked like to me. Also, can you confirm - is the high speed cam designed to come into play ONLY at the highest speed?

As for springs, I have had good luck with very simple shutters bending or re-making them to restore (or even increase) speed. But it could be trickier with a "real" shutter since it is not just a single speed I am trying to get right.

 

[Tom1956]

The high speed spring come into play only at 1/400. I can't clearly recall the configuration, but if incorrectly installed the "peg" that catches the leg of that springs will pass by it and not catch-and-engage. If your shutter is a Wollensak-style rim set, then filing of the step cam is a remedy. But to me, it is a false remedy IMO. That is to say if you have to resort to that, then the shutter parts are obviously worn enough that "lateral" shift in the motion of parts is additively causing slow movement. In other words, there is only so much you can do. I'm convinced that an iris shutter is one of the poorer inventions of man, but they exist only because there IS nothing else. I think they wore out after the first 5 rolls of film, and have been inaccurate ever since. Or something. I think an iris shutter would be a watchmaker's nightmare. You can spend weeks in experimenting, bending new springs, re-fabricating parts worn by the measurement of a germ's hair, and end up in an asylum before making one behave the way it's supposed to. Sometimes you have to say "it is what it is". Try not to get too annoyed when you have to open it up again for the 40th time.
Remember, I'm no shutter expert by a long shot. I'm just a fellow who refuses to be just a regular boob.

 

[Tom1956]

Be sure the bottom of the HS spring is not between the 2 parts, but against it's appropriate notch. even then, there is no 1/400. Probably never was. Top speed will never be more than 1/250 unless you put a turbo supercharger and a rocket engine on it. Be glad to get 1/200 and below The bitter truth. Congratulations. 1/400 is now your 1/200th or 250th if you're lucky. And 1/200th is your official 1/150th. When you've had the shutter apart for umpteen dozen times you will realize this.

 

[Denverdad]

Tom, thanks for the information, suggestions, ...and empathy! I reluctantly removed the lenses from the shutter last night and tore into it for a follow-up inspection. One thing I saw was that the high speed spring and cam appear to be doing just what they are supposed to - contact is made just above the 200 mark on the cam ring thingy (only at 400 does the boost kick in), and the spring is being wound normally when this occurs without binding or anything like that. Also, checking the numbers above, the duration at 400 is in fact half of what it is at 200, so that is consistent. Without the high speed mechanism engaging, the speed would have been the same and the error at the highest speed would have continued its trajectory to an even higher value, instead of leveling off. So all in all, I think that part of the shutter is actually OK!

But I discovered a curious thing last night that I hadn't noticed before. At one of the shutter blade locations, there are actually two blades, so it basically appears as if that blade were doubled. There are a total of 6 blades instead of 5, and at the double blade location the adjacent blade actually slides between these two. Is that normal for this shutter? If not (perhaps an assembly error at the factory?) it might explain the slowing of speeds on the high end. Does anyone know? I can take a picture of this later if it helps to explain what I am seeing.

 

[Hatchetman]

I recently had a Supermatic 800 shutter CLA'd by Paul Ebel. It was his opinion that the fast shutter speeds never did, in fact, reach those speeds. Now the shutter is accurate to 1/100. Then 1/200 is also 1/100. 1/400 is actually 1/200 and 1/800 is actually 1/400, which is pretty good. I just need to remember that when using.

 

[Tom1956]

Yeah, 2 is typical, on a Betax shutter anyway. I didn't know they put 2 on the Supermatic. Leave it be, it's no mistake. One is on top of the blade stack and 1 at the bottom of the stack. This is to be sure the shutter really is light-proof when it's closed. Otherwords, light would "leak around" the next lowest blade from the top of the stack without it, and the shutter would be worthless for walking around in the daylight without a lens cap.

 

[shutterfinger]

free to save the jpegs to your hd.
The double shutter blade is to ensure a light tight shutter. A few shutter makers use the system so it is not uncommon.
Slow high end speeds are a result of running the shutter too long between CLA's. A low use or cla done when the shutter was between .3 and .5 stop off usually return to .5 stop off at 200 and 400.
Shutters run until they were 1 stop off to quit working only come back to in tolerance up to 100 with faster speeds 1 to 2 stops off.

 

[Denverdad]

View attachment 89433 View attachment 89434
free to save the jpegs to your hd.
The double shutter blade is to ensure a light tight shutter. A few shutter makers use the system so it is not uncommon.
Slow high end speeds are a result of running the shutter too long between CLA's. A low use or cla done when the shutter was between .3 and .5 stop off usually return to .5 stop off at 200 and 400.
Shutters run until they were 1 stop off to quit working only come back to in tolerance up to 100 with faster speeds 1 to 2 stops off.

Thanks shutterfinger - I was hoping you would chime in!
Very interesting about the double blades - I had never seen that before but I guess the previous shutters I have worked on simply didn't have this feature.
If I am hearing you right regarding neglected shutters (infrequently serviced or unused for a long period of time), it is rather disappointing to know that even with an expert CLA the top speeds might not be recoverable to any closer than a stop or two. It seems to me that many if not most vintage folders that one might buy fall into this category. I take it that this is more than just a case of worn springs which could theoretically be replaced? Is it more an issue of wear to contact surfaces such as pins and cams and that sort of thing?

 

[Tom1956]

I echo that, shutterfinger--thanks. Had seen your handle on occasion, but never connected it with a good knowledge on shutters. Regards.

 

[Denverdad]

I recently had a Supermatic 800 shutter CLA'd by Paul Ebel. It was his opinion that the fast shutter speeds never did, in fact, reach those speeds. Now the shutter is accurate to 1/100. Then 1/200 is also 1/100. 1/400 is actually 1/200 and 1/800 is actually 1/400, which is pretty good. I just need to remember that when using.

Would that be the Synchro-Rapid 800 you are talking about? I happen to have one too, in a Kodak Tourist. For a long time I had resisted buying the top-end Tourist I wanted (the one with the Anastar lens), just because of all the stories about how unreliable and potentially difficult to service that 800 shutter was. But I finally decided to risk it, partly because of the cool factor of the double-ended shutter mechanism (I guess I'm a bit of a gadget freak), and because I really just wanted to see how that Anastar lens would perform. It turns out the 800 on the one I bought is in fact going to need some service - surprise, surprise - so we'll see how it goes! In the end if I could get to the clean sequence you mentioned, topping out at 1/400 I would be happy. It is sounding more and more like that is the best I can hope for with the Flash Supermatic too (topping out instead at 1/200). As long as the speeds are consistent I think I could probably live with that.

 

[shutterfinger]

Had seen your handle on occasion, but never connected it with a good knowledge on shutters

I'm just a good electronic and mechanical technician that picked up a few service manuals and taught myself the shutters. I serviced about a dozen Supermatics, about the same number of Graphex/Rapax, and some Compurs, Ilex, and a few Betax.

I think too importance is placed on a shutter speed tester reading. I have been working out testing on Graflex SLR shutters. I did some work on making a shutter speed tester and posted my findings on both photo.net ( http://photo.net/large-format-photography-forum/0044cW?start=10 ) and Graflex.org ( http://www.graflex.org/helpboard/viewtopic.php?t=6105 ). I recently had an Anniversary Speed Graphic in 4x5 format that I CLAed and it tested that the 1000 speed was around 1/550 to 1/600. I metered an exposed some fresh white calla lilies at 1000, 500, 250, and 125 varying only the aperture. The result was the density on film suggested the variance in density to be .5 stop or less across the exposures.
The film used in my test was Arista EDU UIltra ISO 400 developed in HC 110. The flowers were in bright sun with deep shades.

Take a shutter speed tester readings with a grain of salt until some carefully executed test exposures are made to verify or refute the speed tester results.


Springs weaken with age and use. Shafts, pivot pins, and bushings wear faster with dirt and dried lubrication. Wear results in play which prevents parts from moving freely.

 

[Tom1956]

That's about the way I see it on springs and parts wear, Shutterfinger. I'm not big on actual film exposure tests as my defining standard, as there's enough variable there to mislead the results. Still think lab testing for speeds is best. Problem is ,cutting up computer mice for the IR sensors and sound-card audio programs are not exactly Apollo 11 grade tests. Luckily I now know that a 1947 shutter was a worn out piece of junk in 1948 after the first 4 rolls of film.

 

[Hatchetman]

Would that be the Synchro-Rapid 800 you are talking about?

Yes. In the Tourist. Pretty cool camera.

 

[John Koehrer]

In spite of(in some opinions) being instantly worn out or only being good for
four rolls of film, these things served a great many people for a good long time.

Most leaf shutters typically do overexpose at their highest speeds.
This is caused by the the film getting additional exposure as the blades
open and close.

You can see why if you open and close the blades slowly/manually.
In any case your camera is perfectly good to use as long as you know
the error involved.

 

[Ken Nadvornick]

Most leaf shutters typically do overexpose at their highest speeds.
This is caused by the the film getting additional exposure as the blades
open and close.

This was always my common sense understanding as well. Simplified, if it takes the blades a millisecond to open, and another millisecond to close, that extra transitional "open" time gets added to the nominal shutter speed time.

If the shutter is set for a correctly calibrated 1/500 sec once it reaches full open, then the addition of that 2/1000 sec effectively doubles the expected exposure time to 1/250 sec.

My low-mileage and recently serviced Graphex on my Crown measures: 1/400=1/200, 1/200=1/150, 1/100=1/100, with all the rest near perfect down to 1 sec.

At 1/200 sec I know I can handhold the beast for critically sharp negatives, which is the only true non-negotiable shutter requirement for this camera. And as long as I know the shutter speed deltas, I can simply compensate for them when reading my light meter.

My base sunny-16 setting for Ilford HP5+ thus becomes 1/400=1/200 sec at f/22.

Ken

 

[shutterfinger]

My limited testing showed it takes the shutter blades .1 ms to .3 ms to open and faster close times once the delay mechanism has released the blade controller and the cocking spring returns the blade controller to the closed position.
One must factor in the rise and fall times of the sensor/electronics, the sensor's angle of view, and the fact that a leaf shutter is basically a variable aperture and as the shutter opens and closes there is a small portion of that open time that the shutter opening is 3 stops and more smaller than the exposure aperture resulting negligible exposure if any exposure density build up on the film.

 

[Ken Nadvornick]

Agree with all of those geometry-enforced considerations.

My understanding is that factory calibration means a shutter being open all the way for the nominally marked time. Transition time does not factor in to nominal time. So the shorter the nominal, the higher the percentage is the (fixed) transition of the total aggregate open time.

And since no mechanical shutter can ever open truly instantaneously, then no shutter can ever match the nominally marked times. However, only with the shorter times does the transition time become a large enough percentage of the total time to matter.

My assumption has always been that this is the reason behind marking the highest nominal leaf speed as 1/400 instead of 1/500, as the arithmetic progression would require. Just as a recognition of the practical geometric design limitations.

So I guess what I was really trying to say was that none of this really matters in practical use.

If one's top-end shutter speeds are off by a stop slow, just compensate for it. And as long as I can handhold a 4x5 camera without blurring the negative, I'm good. And the shutter is doing its job just fine.



Ken

 

[Ken Nadvornick]

My limited testing showed it takes the shutter blades .1 ms to .3 ms to open and faster close times...

Was that a typo? The blades actually opened in a measured 1/10,000 sec?

Ken

 

[shutterfinger]

Not a typo.

All the variables is why I take actual exposurers over test sensors.


Cock a shutter set on any speed. Place a microphone connected to a computer with an audio recording program running, against the shutter case. Trip the shutter. The first major peak will be the trip lever moving and the next will be the blade controller reaching its limit. The time between the trip peak and shutter blade controller reaching its limit is the "opening time".

Playing back the recording and listening to it through headphones helps to verify the actual peaks from other mechanical movements which will occur close to the desired ones and distinguish between microphone movement against the case and actual shutter operation.

In my testers I use 1/4 inch foam core to mount the sensor(s) on using a #28 wire guage drill bit (.1405) for the photodiode mount hole then a #52 wire gauge drill (.0635) for the hole for the test light to access the sensor. This helps reduce the side light picked up from the sensors angle of view. One could make a 2 sensor tester with one sensor at the center of the shutter opening, which is standard for speed testing, and the second at the outer edge of the shutter opening which would be a more accurate test than the microphone method.

 

[Ken Nadvornick]

Not a typo.

Wow...

Then that makes the shutter blade transition time almost as short as the shortest xenon tube burst in some electronic flash units.

For example, at 1/10,000 (3/30,000) of a sec it's just a tiny bit longer than the shortest duration in the Vivitar 285HV's auto mode flash duration range of 1/30,000 to 1/1000 (30/30,000) of a sec.

This is actually an order of magnitude faster than I have read about previously.

Ken

 

[Denverdad]

Shutterfinger, can you fill me in on lubrication requirements for the sliding surfaces of the Flash Supermatic; either via the manual pages or just from your own experience if you prefer?

The ones I question are the setting lever with sliding contact to the case and the mechanism plate, and the edge of the mechanism plate along which the one coil spring slides (don’t know what to call this spring, but it is the one which is attached directly to the setting lever). Finally, what about the blade controller – should it be completely dry?

Thanks.

 

[shutterfinger]

The ones I question are the setting lever with sliding contact to the case and the mechanism plate, and the edge of the mechanism plate along which the one coil spring slides (don’t know what to call this spring, but it is the one which is attached directly to the setting lever).

Use a thin film of Texaco Unitemp RCX169 grease (I use white lithium grease) to the main drive assembly where it engages the stop stud on the setting lever, on the main drive stud, on the latch at the point it contacts the latch spring, and on the latch where it contacts the retarding sector stud.
I also put a trace on the main setting spring where it contacts the main drive assembly when the shutter is set (cocked).

what about the blade controller – should it be completely dry?

Aperture and shutter blades as well as the shutter blade controller should be dry and free of oil or grease as these lubricants act like glue when on these parts.
Extra fine powdered graphite as well as other powdered dry lubricants work well and improve the shutter's operation.

 

[Denverdad]

Just an update - I overhauled the shutter again, this time disassembling it more completely and doing a more “proper” CLA than before. As you can see from the attached before-and-after data, I was able to improve the timing. This comes partly just from adding lube where it is helpful, while keeping it away from where it doesn’t belong! (thanks for the info shutterfinger) But I also reworked some components, including some deburring and sanding, and also shortening the main spring, all to help get the speeds up. I have to admit that I applied grease in a couple of additional places just to make things smoother, for example on the underside of the aperture pointer ring (and yes, you do need to be extremely careful not to let any of that grease migrate to the blades themselves or any part of the blade assembly!)


I feel I have a much better appreciation and understanding of the workings of this shutter now, including where things can go wrong over time that may account for them tending to run slow. Oh, and I also have a more detailed version of this data which explores how repeatable the shutter speeds are (sometimes not very!), and the effect of aperture on the effective timing, if anyone is interested.