Testing films (heeelp, please... :)

[talajmente]

Hi guys,

I have obtained a densitometer and playing with it hard. Here's the first results (FP4, 9x12 sheet, Rodinal 1+25, 19.5C, 5mins tube development)

logH D
-3,23 0,1
-3,08 0,1
-2,93 0,1
-2,78 0,11
-2,63 0,12
-2,48 0,13
-2,33 0,14
-2,18 0,17
-2,03 0,23
-1,88 0,3
-1,73 0,38
-1,58 0,45
-1,43 0,51
-1,28 0,6
-1,13 0,68
-0,98 0,79
-0,83 0,9
-0,68 1,03
-0,53 1,15
-0,38 1,26
-0,23 1,38
-0,08 1,5
0,07 1,62
0,22 1,73
0,37 1,85
0,52 1,97
0,67 2,06
0,82 2,15
0,89 2,18
0,99 2,27

The table above seems to be correct; e.g: the speed point is @-2.1, which means ~ISO100, BF+0.6 is @-1.1 (~ISO125), etc.

What blows my mind: okay, it seems that ISO100 films need about 0.1 lux-second to become middle gray, but:
according to any lightmeter, EV0 (~2.5lux), f/1, 1secs is the same amount of light (=2.5lux-seconds)
so, 0.1 = 2.5??? What did I wrong????

Thanks for your help,
- t.

 

[RalphLambrecht]

I don't understand what you're doing here. Please explain in more detail what you did. What is logH (absolute or relative exposure)? How do you get your speed point? Why is b&f+0.6 of any relevance? How do you turn logH into lux-seconds?

 

[Chan Tran]

I don't understand what Talajmente was doing either but the Lux-seconds= 10^LogH. The H (I don't why) but stand for the illuminance at the film plane in Lux-Seconds. So LogH is Logarithm to the base 10 of the illuminance at the film plane in Lux.seconds.

 

[don sigl]

I'm having difficulty understanding this as well. I'm not sure what the negative numbers are. Are these supposed to be densities? If so, they shouldn't be negative. Have you zeroed and calibrated the densitomer?
What densities are you reading? A 21 step grey scale?
Also, I'm not following the whole lux relationship. 1 stop =.3 as a logarithmic function within a D log H curve.

Film speed is established as a slight density above B+F.

Maybe all this is beyond me. You probably have to simplify this before I'm going to be able to understand it.

 

[Chan Tran]

After rereading some documents I have on hand it seems that the data is correct. And 0.1 lux seconds is equal to medium gray with ISO100 at the film plane. I believe what you miss is this.
If a totally diffused surface is receiving an illumination of 2.5 lux, the luminance exits from this surface is 2.5/pi candle.meter^2 if it has 100 percent reflectance. With an 18% gray card the luminane is (2.5/pi)*0.18 which is about 0.14 candle.meter^2.
The illuminance at the film plane with an f/1.0 aperture is then about 0.1 lux when taking into account of typical flare and lens transmittance factor (T stop kinda) and the factor of pi/4.

The factors of 1/pi and pi/4 are related to the conversion between illuminance to luminance and vice versa.
I hope it helps.

 

[Chan Tran]

I have a question. How did you expose your film. How do you know if the neg received the correct amount of exposure?

 

[roteague]

And I have a question. Why are you trying to turn photography into a science exercise? Take pictures, print and enjoy the results.

 

[ronlamarsh]

Film Testing

I have found that doing a little testing is good , so that one can get a feel how the film developer combination performs. I do a quick ISO test (zone 1 expsure that gives first visible step above B+F) then cut that ISO in half. I then shoot a Zone 8 at this ISO and print at my max black(for enlarger/paper /dev combination) and adjust my film developement to obtain a visual Zone 8 on the print. I have found that testing beyond this to be a dead end, because using this method you have matched the film/devpaper/dev/enlarger combination to produce the optimum range of printable densities. As stated by many( including Ansel) film testing and sensitometry can become an end in itself. I find I have more time to make images and print using the above method(recommended by Bruce Barnbaum) than getting mired in a sensitometric maze.

 

[RalphLambrecht]

Don

The negative logH are fine. This assumes the lightest bar to be of one base unit of light. The darker bars are a fraction of the base. That's not the problem.

 

[RalphLambrecht]

Yes, but these are relative lux-seconds. One can't use them to calculate absolute film sensitivity.

 

[RalphLambrecht]

After rereading some documents I have on hand it seems that the data is correct. And 0.1 lux seconds is equal to medium gray with ISO100 at the film plane.

Where do you get this from, and what is medium gray at the film plane?

 

[RalphLambrecht]

And I have a question. Why are you trying to turn photography into a science exercise?

It is so much fun!!!

 

[RalphLambrecht]

I have found that doing a little testing is good , so that one can get a feel how the film developer combination performs. I do a quick ISO test (zone 1 expsure that gives first visible step above B+F) then cut that ISO in half. I then shoot a Zone 8 at this ISO and print at my max black(for enlarger/paper /dev combination) and adjust my film developement to obtain a visual Zone 8 on the print. I have found that testing beyond this to be a dead end, because using this method you have matched the film/devpaper/dev/enlarger combination to produce the optimum range of printable densities. As stated by many( including Ansel) film testing and sensitometry can become an end in itself. I find I have more time to make images and print using the above method(recommended by Bruce Barnbaum) than getting mired in a sensitometric maze.

Any testing beyond 5 rolls or sheets is too much, but that is for another thread.

 

[talajmente]

It is so much fun!!!

well... thank you Ralph!

I don't understand what you're doing here. Please explain in more detail what you did. What is logH (absolute or relative exposure)? How do you get your speed point? Why is b&f+0.6 of any relevance? How do you turn logH into lux-seconds?

Okay, first of all: there're no scientific maze things, only that I've bought that dummy Beyond the zone system book and I want to get something for my 50 bucks...

What I did: set up the enlarger. Put a shutter into it, with a 105mm lens. Metered the light (it was EV5.3 which seems to be 2.5*2^5.3 (98)lux) Put a film sheet on the baseboard with a calibrated step wedge (http://www.danes-picta.com/). With 1/10 secs exposure there's 9.8 lux on the film without wedge. (0.99 logarithmic) And so on... the wedge B+F is 0.1, so there's 0.89 (log)lux-seconds, etc. (The wedge is incrementing in 0.15 steps.)
- I think, it was absolute (I'm not sure, that I understand well, what does 'relative exposure' mean here)
- The speed point (here) is that exposure, where the film density is 0.1 over B+F (zone I.).
- 0.6? Perhaps some kind of mid-gray. I've red somewhere over the Internet that 0.8/exposure@BF+0.1 is the (standard?) film speed. And that, 10/exposure@BF+0.6 is the (relative?) film speed as well. And that, 36/exposure@BF+1.2 is ... errg... uhh... a... film speed... kind of...
- log H is lux-seconds, logarithmic, with base of 10. If you put the data above into Excel, you'll get something like this http://www.kodak.com/global/en/professional/support/techPubs/f4016/f009_0438ac.gif

What I did(understood) wrong... okay, my english skill is over @ this point, so... (I've posted this question to photo.net as well, Dead Link Removed and got a great answer there)

Thanks for your help,
- t.

 

[Chan Tran]

It is so much fun!!!

That's what I was going to say too!

"What is medium gray at the film plane?"
I didn't write it clear enough. Let me try again. If you aim your camera at a surface and set the exposure according to a reflected light meter (assuming that the meter is calibrated for 18% gray, which many are not) then the exposure expressed in lux.seconds at the film plane is what I called "medium gray at the film plane" which I think it's my bad choice of word. In another word with ISO100 film the exposure for the zone V is 0.10 lux.seconds.
I hope I explained it.
I am still learning about light and its measurement but basically I got the info out of the 2 pdf files here:
http://www.largeformatphotography.info/articles/conrad-meter-cal.pdf
Dead Link Removed
I may miss something as the subject if quite hard to grab. But I will keep trying because like photography the science of photography is a lot of fun too.

 

[nworth]

I, too, fail to understand the lux-seconds part of the post. But it looks like the film is just showing greater than proportional contrast. Going fron Zone I to Zone V you expect a 16 fold increase in exposure with normal development. Here the measurement was a 10 fold increase in exposure to give the equivalent increase in density, ie. increased contrast, which probably means overdevelopment. Any other considerations should take that into account.

 

[Ray Heath]

did you actually go outside and capture the sort of images you normally make??

 

[RalphLambrecht]

I am still learning about light and its measurement but basically I got the info out of the 2 pdf files here:
http://www.largeformatphotography.info/articles/conrad-meter-cal.pdf
Dead Link Removed

Thanks for the great links. Lot's to read but looks good at first glance.

 

[Stephen Benskin]

First of all talajmente is only attempting to do sensitometric testing. Unfortunately, using an enlarger isn't going to cut it if you're interesting in actual log-H. If you are truly interested in sensitometry, check out Ebay for a EG&G Mark VI or Mark VII sensitometer. You can get one for cheap. EG&G no longer make them and were really resistent to the idea of calibration last time I checked. Their calibration costs went from $200 to $1200. Still, I picked up a spare sensitometer about a year ago for around $100. If anything, they will produce consistent exposures.


There's a constant known as P. P = 8 or 8 lux at the film plane. Meter calibration is based on this. Plug in the ISO for the shutter speed and that will give you the midpoint exposure for the given ISO.

8/100 = 0.80 mcs
8/125 = 0.64 mcs
8/400 = 0.20 mcs

In other words P is equal to E in the E * t = H equation or intensity times time equals exposure. P is the midpoint intensity.

The basic exposure equation is H = q*L*t / A^2
q = light loss constant - .65
L = luminance in lux = 297 footlamberts (297*10.76) (according to meter calibration equation) in this case L can be thought of as Lg or average luminance. Need to convert
t = shutter speed
A = f/stop

take out the shutter speed and you get

.65*(297*10.76) / 16^2 = 8.11

I've found 8.11 works best with all the equations including the balanced constants equation of K = P/q where K = 1.16, P = 8.11, and q = .65. I believe P = 8 is just a basic rounding.

Put the shutter speed / ISO back in 8.11*1/100 or 8.11 / 100 = 0.081 mcs. Your meter wants to make what ever it points at 0.081 mcs at the film plane for an ISO 100 setting. But, then there's flare which in the real world is adding Hf to the exposure equation.

 

[sanking]

First of all talajmente is only attempting to do sensitometric testing. Unfortunately, using an enlarger isn't going to cut it if you're interesting in actual log-H. If you are truly interested in sensitometry, check out Ebay for a EG&G Mark VI or Mark VII sensitometer. You can get one for cheap. EG&G no longer make them and were really resistent to the idea of calibration last time I checked. Their calibration costs went from $200 to $1200. Still, I picked up a spare sensitometer about a year ago for around $100. If anything, they will produce consistent exposures.

Stephen,

I have an EG&G Mark VII sensitometer and do recommend it highly. However, you can get excellent results with an enlarger if you make sure that the plane of illumination is large enough to give even exposure over the entire film, and if you use a light integrator for your exposures. I have been working this way for a very long time and could make a good case that the enlarger+integrator gives results that are at least as consistent as those of the Mark VII.

As I recall, I paid about $100 for my Mark VII on ebay but it did not come with any of the filters so I had to made those with ND material. I did not have my unit calibrated because I am interested in relative comparisons for practical plotting of film curves and don't need to know the absolute values.

Sandy

 

[Papa Tango]

The great thing about all of this is that there is something for everyone. Does anyone have any pictures of this base hog they are writing about? Mere mortals want to see it...

Guess that I will go out an get a few frames in. Anyone see where I left that film box with the "sunny 16" exposure directions???

 

[Stephen Benskin]

It's pointless to get into a how technical is too technical debate. If you don't want to know, nobody is forcing you to be here. It is, however, rude to mock those who are interested.

For those interested, there's a seminal paper by Connelly.

Connelly, D., Calibration Levels of Films and Exposure Devices, The Journal of Photographic Sciences, Vol. 16, 1968.

Another good one is

Stimson, Allen, An Interpretation of Current Exposure Meter Technology, Photographic Science and Engineering, Vol 6, no. 1, Jan-Feb, 1962.

 

[Kirk Keyes]

I did not have my unit calibrated because I am interested in relative comparisons for practical plotting of film curves and don't need to know the absolute values.

Sandy - do you then use another test to determine film speed? If you are using absolute values, you can do it directly from a test like the first poster is doing.

I also agree that an enlarger can be fine for speed/absolute testing. Just make sure the exposure time is long enough and that the lamp is voltage stabilised (no cold lights here) to ensure that you are getting consistent exposures.

 

[Stephen Benskin]

BTW, ISO speed equation is .8/Hm, so the mcs for 100 speed is .8/100 = 0.0080. Notice how there is a 10x (log 1.0) difference between the speed point and the meter point?

There's also something about the speed point most people don't usually consider. Just because it is the point where speed is determined does not mean that it is the desired point for the shadow placement. Take for instance the Fractional Gradient Method. It's speed point falls approximately 0.29 log-H units to the left of the ISO point when the ISO contrast conditions are followed, yet the pre 1960 version produces a speed one stop slower than what is produced with the current ISO method. In other words, shadow placement would fall two stops to the the right of the Fractional Gradient speed point.

Under the statistically average conditions in which the current ISO method is determined, shadow placement will fall approximately 1/3 stop to the right of the speed point.

 

[Markok765]

you guys dont need to do this: i just use the jone system and compenate exp for the diffrent zones