PDA

View Full Version : Shutter speeds on new Copals versus old Compurs



Cor
3-Feb-2011, 07:10
I have literally this mixed bag of lenses when I go out with my 4x5. Some are mounted in Copal (Press) shutters, other in older Compur shutters.

The Copals have the modern range of shutter speeds: 1, 1/2, 1/4, 1/8 etc.

The Compurs run the older set: 1, 1/2, 1/5, 1/10, 1/25, 1/50.

I sometimes have problems to translate the readings of my spot meter to the Compurs.

Ok 1 and 1/2 no sweat, and faster than 1/60 I almost never use. But a reading of 1/4 sec: using 1/5 sec is a bit too short than, and than what about a reading of 1/8; put the Compur on 1/5 or 1/10 ?.. and so on..

I am a bit ambivalent on this though, realizing that the shutter speeds are not exact anyway, and I shoot only B&W only, so there is leeway. Usually I choose the longest speed, a bit over exposure won't hurt, underexposure can.

How do you folks handle above issue?

Thanks,

Best,

Cor

Vick Vickery
3-Feb-2011, 07:33
Guess I'm too old to worry about it...for years I've been just adjusting my aperature a little to compensate for the difference; in practice it works fine: if your meter tells you 1/60 at f16, just use 1/50 and stop down a hair past f16 toward f22. You're talking about way less than 1 stop anyway...my metering isn't that accurate to start with! :)

Cor
3-Feb-2011, 07:36
Good point, Vick,

Forgot to mention that, I also use that method, but with my beloved 120 Angulon in a Compur the f-stop scale spacing gets progressivly narrower (logarithmic?) so moving up or down 1/3 of a stop is a bit difficult and isn't very precise I guess.

Cor


Guess I'm too old to worry about it...for years I've been just adjusting my aperature a little to compensate for the difference; in practice it works fine: if your meter tells you 1/60 at f16, just use 1/50 and stop down a hair past f16 toward f22.

BrianShaw
3-Feb-2011, 07:37
For B+W and C41 use the nearest and don't worry. For E6 correct for the difference with a little more or less aperture.

Leigh
3-Feb-2011, 07:40
The actual shutter speeds should be the same; only the labels differ.

The real speeds are:
1
1/2
1/4
1/8
1/16
1/32
1/64
1/128
1/256
1/512
1/1024

Each of these differs from its neighbors by exactly a factor of two, as required by the law of reciprocity.

You should be able to map either set of labels into that sequence to see which corresponds to which.

- Leigh

Cor
3-Feb-2011, 08:19
I am not sure I am following you on this one, I understand that some rounding off takes place, so the "new" set is more consistent than the old one (ie from 1/10 to 1/25), my question was more a practical one, when your meter gives a reading of 1/8 seconds, go for 1/10 or 1/5..


The actual shutter speeds should be the same; only the labels differ.

The real speeds are:
1
1/2
1/4
1/8
1/16
1/32
1/64
1/128
1/256
1/512
1/1024

Each of these differs from its neighbors by exactly a factor of two, as required by the law of reciprocity.

You should be able to map either set of labels into that sequence to see which corresponds to which.

- Leigh

Frank Petronio
3-Feb-2011, 08:25
Just knock the aperture a hair one side or the other if needed.

Frank_E
3-Feb-2011, 08:27
I was concerned about my shutter speeds so I purchase a tester of ebay and was suprised in the variation of the real speeds from the actual
I would think that is the best solution find out what the shutters are actually experiencing and use those numbers

Leigh
3-Feb-2011, 09:07
when your meter gives a reading of 1/8 seconds, go for 1/10 or 1/25..
You definitely go for the closest one, so in this case 1/10.

To use the list that I gave previously, print it out and write the "modern" shutter speeds next to the real speeds. For example:
1/8 is shown as 1/8
1/16 is shown as 1/15
1/32 is shown as 1/30
1/64 is shown as 1/60
1/128 is shown as 1/125

Then do the same for your Compur shutter, writing one speed per line starting at 1 second. This will show you which Compur labels correspond to which real speeds.

- Leigh

Cor
3-Feb-2011, 09:24
You definitely go for the closest one, so in this case 1/10.

To use the list that I gave previously, print it out and write the "modern" shutter speeds next to the real speeds. For example:
1/8 is shown as 1/8
1/16 is shown as 1/15
1/32 is shown as 1/30
1/64 is shown as 1/60
1/128 is shown as 1/125

Then do the same for your Compur shutter, writing one speed per line starting at 1 second. This will show you which Compur labels correspond to which real speeds.

- Leigh
Ok

Btw I made a typo, 1/25 should read 1/5, sorry about that

Cor

Doremus Scudder
3-Feb-2011, 10:01
The older shutter speed sequence (1/5 - 1/10...) almost exactly one-third stop different from the newer. Here's the run down

1/5 sec. is one-third stop less exposure than 1/4 sec.
1/10 sec. is one-third stop less exposure than 1/8 sec.
==== there is a hole here around 1/15 sec ===
1/25 sec. is one-third stop more exposure than 1/30 sec.
1/50 sec. is one-third stop more exposure than 1/60 sec.
1/100 sec. is one-third stop more exposure than 1/125 sec.
1/200 sec. is one-third stop more exposure than 1/250 sec.

While this is approximate, it is way close enough. Most meters have intermediate markings in one-third stop intervals which make figuring out which aperture to use easy. I like to think of the intermediate values as "+" or "-" of the shutter speed or main aperture number- You can think of 1/5 sec as "1/4+" and 1/50 as "1/60-" etc. Apertures work the same, e.g., I might expose at 1/4+ sec. (i.e., 1/5 sec) at f/32- (i.e., one-third stop open from f/32).

When in doubt and you're shooting black-and-white, expose more. Erring 1/3 on the side of overexposure cannot hurt.

Hope this helps,

Doremus Scudder

Leigh
3-Feb-2011, 15:55
1/5 sec. is one-third stop less exposure than 1/4 sec.
While the statement may be correct as regards 1/5 v. 1/4 second (I haven't calculated the error in stops), it's not relevant to shutter performance.

The very simple point that I'm having considerable difficulty making is that the numbers on the shutter scale DO NOT indicate the true speeds.

When you select the 1/5 second position on the shutter dial, you actually get 1/4 second. That's the way the shutter is designed.


- Leigh

ic-racer
3-Feb-2011, 16:18
My Compur service manual indicates a factory tolerance limit of +/- 15% on the low speeds and +/-20% on 1/250 and 1/500.

You can safely round the speeds up or down if the shutter is 'in-spec' but likey it is 'out of spec' and if you are concerned enough about it to care, you should check the speeds with a shutter speed tester.

BrianShaw
3-Feb-2011, 16:51
The very simple point that I'm having considerable difficulty making is that the numbers on the shutter scale DO NOT indicate the true speeds.

I understand your point and suspect other do also, but have never heard that before. By design, I suppose a timing cam can be cut to get ANY speed - either that indicated on the dial or what you call true (I would call that "theoretical").

But all this is interesting talk and I thank you... but the only "true" shutter speed of practical consequence is that which the shutter actually produces at the time it is used.

The real point, I suppose, is that 1/3 stop difference rarely makes a significant difference.

John Koehrer
3-Feb-2011, 18:07
The real speeds are measured in milliseconds(ms) not fractions.
The indicated speeds are rounded to an easily identified number and the error present in any shutter eats it. It means nothing unless you want to argue about nothing
As an example 1/4=250ms 1/5=200ms, 1/10=100ms 1/8=125ms 1/25=40ms 1/30=33ms 1/50=25ms and 1/60=16ms
In what perfect world do these differences exist?

JimL
3-Feb-2011, 18:45
While the statement may be correct as regards 1/5 v. 1/4 second (I haven't calculated the error in stops), it's not relevant to shutter performance.

The very simple point that I'm having considerable difficulty making is that the numbers on the shutter scale DO NOT indicate the true speeds.

When you select the 1/5 second position on the shutter dial, you actually get 1/4 second. That's the way the shutter is designed.


- Leigh

Wouldn't the shutter be designed to give you 1/5 sec. +/-15%? What you actually get, with a properly working and adjusted shutter, would be anywhere in this range.

Cor
4-Feb-2011, 02:07
Thanks for al the feedback, especially Doremus for a neat way of approaching..on one hand this discussion is perhaps moot because in practice the difference are ruled out by shutterspeeds beeing off, metering errors, perhaps even F stop errors, on the other hand I have this exactness streak (beeing a scientist) and nerdiness which is part of the attraction to LF.

Learned a few things again!!

Best,

Cor

Doremus Scudder
4-Feb-2011, 03:06
Leigh,

You are correct when you say that actual shutter speed times in milliseconds are the way to properly determine the correct shutter settings. This is precisely what I do. I use the attached shutter-speed table to compare measured shutter speeds to the ideal, rounded to the nearest 1/3-stop. I then make a sticker for each shutter tested and attach it to the lensboard. I use the "+" or "-" signs to indicate slower or faster speeds in 1/3-stop increments from the standard speeds as I described above (e.g., 1/4-, 1/30+ etc.).

I realize that there is error in the system. All the more reason to try and be accurate. If there is a 1/3-stop error in the systems and I only try to get within a 1/3-stop tolerance, then my margin of error doubles. I think 2/3-stop error is significant enough to try and avoid. Especially, since it may really be significantly more than that.

So, trying to get to the closest 1/6 stop or so means my exposures have a good chance of really being only about 1/3-stop or a little more off. If I'm within a half-stop or so, then I'm happy, but one simply won't achieve even that accuracy unless you try to be much more accurate than that, due to the errors in the system over which we have no control. I can control the aperture lever accurately, so I do. I can also get my shutters tested and come up with a good idea of where the true speeds are (FWIW, I have the shutters tested at f/22 and fired several times and then take an average; this gives me the speed at an aperture very close to my normal taking apertures).

Here's an example:

Ektar 203 f7.7
Carol Flutot, March 2009
Av. Shutter Speeds at f/22
Marked = Tested = Actual (rounded to nearest 1/3 stop)
1 sec = 880ms = 1 sec
1/2 = 449ms = 1/2
1/5 = 208ms = 1/4 +
1/10 = 114ms = 1/8 +
1/25 = 42ms = 1/30–
1/50 = 22.6ms = 1/60–
1/100=16.6ms = 1/60
1/200=6.3ms = 1/125+
1/400=5.2ms = 1/250–


(I hope the lack of formatting available allows the above to be understood.)

As you can see, the odd, older speeds (1/5, etc.) are very accurate and very close to the stated speed on the shutter, i.e., 1/5 marked is very, very close to actually being 1/5 sec. etc.

The document I mentioned earlier with shutter speeds in milliseconds in 1/3-stop increments is attached for anyone interested.

So, to the OP: If you really want to be accurate, and not just clear up confusion about how to translate the old shutter speed sequence to a meter marked with the modern ones, you need to have your shutters tested and do the above.

Hope this helps.

Doremus Scudder

Cor
4-Feb-2011, 05:25
Thanks for the handy document, Doremus!

It's indeed time that I should test my shutters, I have this soundcard testing device somewhere.

best,

Cor

BetterSense
4-Feb-2011, 07:21
The very simple point that I'm having considerable difficulty making is that the numbers on the shutter scale DO NOT indicate the true speeds.

When you select the 1/5 second position on the shutter dial, you actually get 1/4 second. That's the way the shutter is designed.

I would like to know what you are basing this conclusion on.

Leigh
4-Feb-2011, 07:25
I would like to know what you are basing this conclusion on.
The law of reciprocity (look it up).

- Leigh

Leigh
4-Feb-2011, 07:37
(FWIW, I have the shutters tested at f/22 and fired several times and then take an average; this gives me the speed at an aperture very close to my normal taking apertures).
Obviously film can tolerate significant errors in exposure and still produce satisfactory results. That's particularly true of negative materials, because adjustments can be made in the printing process. Positives (slides) are less forgiving.

My point about nominal vs. marked vs. actual speeds is more an academic exercise than a real concern. The power-of-two progression is required for exacting work, but obviously the real speeds are what must be used with real shutters.

Your Ektar shutter is quite a bit off. I understand that it's being tested at f/22. Many photographers prefer to have shutters tested at the common working aperture. It makes more sense as far as getting accurate exposures in your particular system.

But checking a shutter to manufacturer's specifications requires that the aperture be full open. The blade transition time becomes significant with leaf shutters at high speeds. That's why a Copal 0 will go to 1/500th, while a Copal 1 will only do 1/400th, and a Copal 3 only 1/125th.

Anyhow, thanks for the comments. Good shooting. :D


- Leigh

Jack Dahlgren
4-Feb-2011, 09:37
Obviously film can tolerate significant errors in exposure and still produce satisfactory results. That's particularly true of negative materials, because adjustments can be made in the printing process. Positives (slides) are less forgiving.

My point about nominal vs. marked vs. actual speeds is more an academic exercise than a real concern. The power-of-two progression is required for exacting work, but obviously the real speeds are what must be used with real shutters.

Your Ektar shutter is quite a bit off. I understand that it's being tested at f/22. Many photographers prefer to have shutters tested at the common working aperture. It makes more sense as far as getting accurate exposures in your particular system.

But checking a shutter to manufacturer's specifications requires that the aperture be full open. The blade transition time becomes significant with leaf shutters at high speeds. That's why a Copal 0 will go to 1/500th, while a Copal 1 will only do 1/400th, and a Copal 3 only 1/125th.

Anyhow, thanks for the comments. Good shooting. :D


- Leigh

Your error is in the assumption that shutter mechanisms follow a linear series x/2, x/4 etc. Just because you believe that a power of two progression is required does not mean it is being supplied. In fact mechanical shutters are decidedly non-linear as any shutter test will show - despite the best efforts of designers and manufacturers over the years.

I have no idea how you justify extrapolating from your erroneous assumption to the conclusion that shutter manufacturers knowingly mis-labeled their shutters for decades.

Leigh
4-Feb-2011, 09:57
I have no idea how you justify extrapolating from your erroneous assumption to the conclusion that shutter manufacturers knowingly mis-labeled their shutters for decades.
I've repaired leaf shutters, primarily Compur, for about 35 years. The ones I've measured follow the power-of-two progression pretty accurately (disregarding high-speed errors).

Do you repair shutters?

On what knowledge are you basing your criticism?

When a user wants to change aperture, for example, by two stops, he changes the shutter speed by two steps to compensate. The labeling means nothing. The user expects that the two-step change will result in a 4x change in speed.

In fact many of the old Compur shutters were EV-locked, so changing either aperture or speed would move the other adjustment equally, to provide constant exposure. You had to push a lever to disengage the interlock if you wanted to adjust aperture or speed independently from the other.

Presenting labels that follow common convention does not constitute "mis-labeling". Remember that shutters existed long before light meters. The modern speed sequence is an attempt to make the labels agree more closely with the actual values.


- Leigh

BrianShaw
4-Feb-2011, 13:58
Leigh,

I hesitate commenting since this is becoming a rather "testy" discussion of largely theoretical issues, but let me preface by saying that I have less experience in terms of years than you but I also repair shutters and am basing what I write based on repair manuals that I reviewed this morning.

Summarized from the Graphex shutter repair manual:
Graphex shutters (1, 2, 5, 10, 25, 50, 100, 200, 400 progression) have documented operational speed specs that average the speeds indicated on the dial. This is true throughout the operational range.

Summarized from the Synchro Compur repair manual:
Synchro Compur 0 and 00 MXV and XV shutters (1, 2, 4, 8, 15, 30, 60, 125, 250, 500 progression) have documented operational speed specs that average the speeds indicated on the dial EXCEPT 250 and 400 has a much larger tolerance.
- and -
Compur 000-X seems to be spec'd true to the speeds indicated on the dial at 30 and 60, but almost 1 ms faster nominal speed at speeds over 125.

It seems to me based on these data points that the shutter speeds by a couple of major shutter manufacturers was intended to be nominally true to the scale on the shutter, not the "logical progression based on reciprocity".

Am I missing something? You seem to be saying that despite these documented data points the shutters you've seen tend to run in accordance with the logical progression rather than the speed scales. I'm not doubting your experience, just curious.