Exposure Meter testing any affordable devices available,or any shops still doing this

[ic-racer]

Another thing I was thinking about is some analogy to electrical circuits and constants. In an electrical circuit there is a pretty good 'standard' voltage that you get with a diode. Just about all diodes always drop the voltage about 0.7 volts. So, this is used as a voltage reference in many circuits.

Unfortunately the intensity of various light emitting diodes is all over the place, so that is no good as a standard.

Maybe a standard size Sterno can. They are said to produce a constant flame for 45 minutes or so.

 

[Kirk Keyes]

Kyoritsu
The EF series camera testers are made for repair, research,
development and quality control. They include exposure
measuring testers, multiplex exposure and analysis system

Stephen - let's chip in together!

http://cgi.ebay.com/KYORITSU-EV-TES...0?hash=item33511db1b7&_trksid=p3286.m20.l1116

 

[RJS]

I think I would try the Walgreen lamp and see if the results conform to Dr. Henry's. Particularly, obtaining the ND filters he suggests would be useful in any case. If your meter is not linear in response it can't be accurate, and linearity is easy to test.

 

[Stephen Benskin]

Stephen - let's chip in together!

http://cgi.ebay.com/KYORITSU-EV-TES...0?hash=item33511db1b7&_trksid=p3286.m20.l1116

Nice find Kirk! Sure I'll go in on it, but can we keep it at my house?

Of course, testing one's materials is a good idea. The question has to be asked about how valid the tests are. Is a "standard" light bulb going to be consistent enough to produce accurate results? Is the tester considering all of the variables that could influence the result? Take a look at any testing equipment and most of it is to insure accuracy and repeatability. There are feedback systems and regulators that makes something as simple as using a photo detector to read the density of a piece of film into a pricey option. And if you want that to be really accurate, then the price shoots up astronomically. Just ask P.E. about the difference between a commercial densitometer that reads 0.00 with an +- 0.01 error with what Kodak uses.

Part of the precision comes from confirmation of that precision. My sensitometer is designed to give repeatable and knowable exposures, but it has to be calibrated to confirm what those exposure values are, and the device that calibrates the sensitometer has to be calibrated too, and so on. The "standard" bulb may have a particular illuminance that is common for that bulb in general, but what about specifically. How can you verify the actual output of the bulb? Variations in manufacturing occur. Somewhat ironic when you are using it to calibrate the one thing that might be able to verify it. Home voltages vary. How accurate is the voltage regulator and is that accuracy knowable and provable? Potential errors can build up quickly.

Then there's actually doing the tests correctly. Take one of my favorite subjects for ranting, Zone System speed testing. The reason why people tend to produce personal speeds lower than the ISO speeds in a consistent manor is because of a false assumption in the test. And that's not including the question of precision of changing the f/stops in increments of 1/3 stop, or the color temperature of the light source the test was made of, or a number of other potentially influential factors. How many consider the concept of hold time? Most also only do the test once thus making it impossible to determine if that one test is an outlier because of procedural errors. Yet, somehow, many of these people point the finger at the manufacturers as being incorrect with speed testing.

I would estimate that around 95% of those who have read the ISO film speed standard are probably misinterpreting a central tenet which has to do with a fixed density to determine film speed. One of the few disagreements I have with Phil Davis has to do with this one point. I believe he was aware but decided not to deal with the issue in BTZS. Even with all this error, most people still manage to produce good images.

With most amateur testing, are we actually doing it to achieve a better understanding of the materials, or are we really doing it to predicate a false sense of control and understanding?

 

[Nicholas Lindan]

Just about all diodes always drop the voltage about 0.7 volts.

Er, no. It can be anything. Even restricting the discussion to pn silicon diodes the operating voltage of a typical diode can be from 0.3V when carrying a small current of a few nA, 0.7V at 50 mA and 0.85 V at 1A.. Germanium diodes are often operated at .1 volts, as in a crystal radio.

http://tuttle.merc.iastate.edu/ee201/lectures/diodes.pdf

Scroll down to page 8, right hand side graph.

If you clamp the current at a constant value then the voltage will depend on the temperature, and vice-versa. Page 6 in the above course lecture.

High current diodes are often operated at 3 volts and above with currents in the 1000A-2000A range.

 

[RJS]

Stephen, I think you have struck the nail squarely. Certainly all the testing we do does give us a sense of understanding the materials better, and when we reach a point that we are able to expose and develop film with the assurance that most of the time we will get predictable, usable results I think that is all we can realistically hope for. And I don't think it is necessarily a 'false' sense of control and understanding. The kind of precision EK is able to produce is way beyond anything we need. They do need it, and I am delighted they produce materials to the kinds of accuracy and tolerance they do. It lets me know that when I do my crude testing that if it is at variance with EK (or Ilford or Fuji) that I need to go back and find where I have gone astray

 

[Anscojohn]

I-C,
Thanks for the charts. I can see if such a variance were--to use a former example--the lumens in a light bulb, that would be too much variance to make it useable as a standard.
I'm beginning to think it might be best to combine some of the suggestions and just forget the whole notion--light some candles, sit in the warm glo, drink the sterno and say the hell with it all.(vbg)

 

[Kirk Keyes]

I think I would try the Walgreen lamp and see if the results conform to Dr. Henry's. Particularly, obtaining the ND filters he suggests would be useful in any case. If your meter is not linear in response it can't be accurate, and linearity is easy to test.

I just bought a 4 pack of the GE 1585 lumen bulbs. I already have a set of 2.0 and 3.0 ND Wratten filters.

 

[Stephen Benskin]

Repeatable, absolutely important. The use of "false" might be too extreme. "Skewed" or "unfounded" might be better. How many ZS people are walking around happy in the knowledge they are using personal speed that are off by 1/2 stop? They tend to be sure that the personal testing they did confirms their results.

Does it really matter if they are off? Not really. Speeds were stop slower prior to 1960 and lenses had around 1/2 to 1 stop more flare, which would make a difference of up to two stops difference in film speed between pre 1960 and today. And people got good results then as well as now. So, it really doesn't matter in practice thinking you have the "true" speed of the film or not, but it is still an unfounded sense of accuracy.

One of the realities with exposure is flare. No matter how accurate the meter or the exposure placement, the variable of flare and the inability to precisely measure it negates absolute knowable placement. The amount of veiling flare changes with the luminance range and even within the same luminance range depending on the amount and distribution of tones. A small black area surround with a large white area will have a different flare factor than a small white area surrounded with a large black area. Flare affects the apparent luminance range at the film plane changing where the shadow falls in relation to the metered shadow. Since flare is near impossible to measure in the field, so is knowable shadow placement.

 

[RJS]

Kirk, I will be really interested to hear what your results will be. I've been thinking a bit about what you will be doing and I wonder if, assuming you don't get a reading of 15 with your Pentax if you vary your distance so your meter reads 15 you won't have a nearly equally good 'calibration'. If your meter is linear in response. Off hand I can't see why your results shouldn't be quite usable if your set-up reads 15 at really close to 29' 2". This will be VERY interesting (I don't know how to italicize).

 

[Kirk Keyes]

Italicize - you highlight the text and the click the slanted capital I in the upper left of the message box.

 

[Q.G.]

Repeatable, absolutely important. The use of "false" might be too extreme. "Skewed" or "unfounded" might be better.

Indeed.
And even that is a bit strong. All things, even standarized things, are never absolutely the same.
That's not a problem, as long as the variability is small enough. The margins good manufacturers allow in their quality control, the variance in voltage, etc., just like the sun, never the same. But 'same enough' to be useful.

 

[Kirk Keyes]

Also, I have an IR heat filter and I can try it with and without it.

I have some meter testing I did a couple years ago where I used a spectrophotometer and took readings with various meters at various wavelenghts. All dropped off in the UV pretty close to the end of visible wavelengths, and they all did for near IR, but all increased readings as it went further into the IR. I'll see if I can post that data later tonight. I tested a Pentax Spotmeter V (unmodified), a Minolta Flashmeter IV and Flashmeter VI in incident and reflected, a Gossen Starlight, and a Zone VI modified Pentax Digital.

 

[Stephen Benskin]

Indeed.
And even that is a bit strong. All things, even standarized things, are never absolutely the same.
That's not a problem, as long as the variability is small enough. The margins good manufacturers allow in their quality control, the variance in voltage, etc., just like the sun, never the same. But 'same enough' to be useful.

Right. You have just explained much more succinctly than I the difference between good enough to use and the precision needed for testing.

 

[RJS]

In the aerospace industry we used to say "good enough for government work".

 

[Q.G.]

Right. You have more just explained much more succinctly than I the difference between good enough to use and the precision needed for testing.

The pitfall however is believing that the precision in testing can be absolute.
Though much stricter demands need to be satisfied, testing also can not get beyond "good enough".

Another thing to remember is that, when testing a thing that is to be used, eventually the precision that the use of that thing demands, not that of laboratory testing, is normative.
Deviations from what should be that are undetectably small in use do, of course, not matter outside papers in technical journals.
The variance allowed by manufacturers of, say, light bulbs good enough to be used as reference, or standard light sources will take that use into account.

 

[ic-racer]

Er, no. It can be anything. Even restricting the discussion to pn silicon diodes the operating voltage of a typical diode can be from 0.3V when carrying a small current of a few nA, 0.7V at 50 mA and 0.85 V at 1A.. Germanium diodes are often operated at .1 volts, as in a crystal radio.

Am I missing something or am I just zooming in on the "EASY?" From that link you posted, frame #16 reads:

So in the circuit, the forward-biased diode can be treated like a
little battery with voltage of Von = 0.7 V. Easy! And it is a reasonable approximation.

In terms of my original post on the diodes, what I was thinking is that it is too bad I can't just buy an electrical component with a given light output (irrespective of voltage), just like I can buy a diode with a given voltage drop. http://www.reuk.co.uk/Zener-Diode-Voltage-Regulator.htm

 

[ic-racer]

Since flare is near impossible to measure in the field, so is knowable shadow placement.

I'm not challenging that (agreeing actually), but I want to use that statement as a springboard to another one of my exposure methods. I'll just call it the "Flare Method."

It determines the exposure beyond which there is no additional improvement in image quality.

Its very easy. Make a box thats painted black on the inside. Put a hole in the box and put the box in the scene to be photographed. Bracket exposures. Process the film. The lowest exposure that shows (arbitrary 0.1log d) above film base can be said to the exposure beyond which there is no additional improvement by additional exposure.

So, basically flare determines the exposure. Also, notice I wrote nothing about a "light meter" (it is a bracket method) or "measuring flare"

 

[Stephen Benskin]

ic-racer, not a bad idea. I once built a such a black box for testing. The exposure was based on metering off a gray card and opening up 1/2 stop. I then place a variety of cards with different tonal values around the open. If not flare existed, then the box should read 0.00. No surprise to anyone, the white card produced the highest density and the black the lowest.

I do see a small problem with perfecting the technique as a similar scene to the one you tested might have different distribution of tones. Placement would have to be carefully done. Putting it on cement would produce higher flare than on grass.

 

[RJS]

Stroebel in his 7th edition recommends taking meter readings through "an infrared absorbing filter such as the filter recommended for use in enlargers used for making color prints." I've never done that so have no idea how it would effect the Pentax spot meter. I don't presently have such a filter, but if I get one will post whatever it tells me. Perhaps one of you have one in which case I would expect it would indicate if the Pentax has much IR sensitivity. Dr. Henry does not seem to mention IR; at least a cursory look at his book does not reveal such.

 

[Stephen Benskin]

Stroebel in his 7th edition recommends taking meter readings through "an infrared absorbing filter such as the filter recommended for use in enlargers used for making color prints." I've never done that so have no idea how it would effect the Pentax spot meter. I don't presently have such a filter, but if I get one will post whatever it tells me. Perhaps one of you have one in which case I would expect it would indicate if the Pentax has much IR sensitivity. Dr. Henry does not seem to mention IR; at least a cursory look at his book does not reveal such.

I believe the Zone VI modified meter had an infrared absorbing filter as part of the package. I did the TV remote test the other day and got a response from the meter. No way of telling if it would have been more with an unmodified meter.

I need to review the Henry book, did he cover linearity at different color temperatures in his testing?

 

[Kirk Keyes]

I don't presently have such a filter, but if I get one will post whatever it tells me.

Try here:
http://www.surplusshed.com/pages/item/m2927.html

Look around as they have several heat (IR blocking) filters to choose from. And the prices are quite affordable.

 

[Kirk Keyes]

I need to review the Henry book, did he cover linearity at different color temperatures in his testing?

No, and as I mentioned, he did not talk about testing with an IR blocking filter. I dug out my Schott KG-1 IR blocking filter and my Wratten 87 IR filter this morning. My basement is only 20 ft across so I need to do a little math to figure out what the reading would be at 20 ft.

 

[BetterSense]

Here is some info on sun variability. The first graph is all the radiant energy over the course of years. The second shows just the UV (to which light meters will respond) over the course of a day:

I still think a candle is going to have less variability than the sun. And I think different candles will very less than different light bulbs.

Perhaps I'll check all the candles in my home and see how much variability I get.

Did you LOOK at the vertical axis on that first graph? It is scaled! The graph varies by as little as 0.1%! If your graph has a purpose, it's to show how extremely consistent solar lighting is!

As for the second, I have no idea what a "UV index" is, but it sounds like something a weather channel would cook up to scare people into putting on sunscreen.

 

[Q.G.]

I don't presently have such a filter, but if I get one will post whatever it tells me.

They are also in slide projectors. So if you have a slide projector, you do have such a filter.