The Great Sensitometer Shootout....

ic-racer · Oct 13, 2012

I have a number of sensitometers to compare.
Each is a different construction.

EG&G: Xenon flash
Wejex: incandescent lamp
ESECO Speedmaster: green LED
ESECO Speedmaster: blue LED

I'm going to run some step wedges at different development times and see if they all produce curves that can give reasonable information about development time.

The LED sensitometers are frequently available on ebay and I hope to provide some information on the utility of those devices in making control strips for the B&W home darkroom.

Then null hypothesis here is that the color of light and time of exposure will have very little effect on the films response to changes in development times.

James in GA · Oct 13, 2012

My old toys . Most people just eyeball it. I love how a good meter saves so much time.
And you can't lie to yourself.

ic-racer · Oct 14, 2012

Two strips of Tri-X 35mm film were exposed using the 4 devices.
The first strip was processed for 5 minutes.
The second strip was processed for 7 minutes.

The the resulting negatives were read with a calibrated Tobias densitometer and plotted in Microsoft Excel. The slope of the straight line portion of the curves was obtained with a linear regression analysis.

ic-racer · Oct 14, 2012

Exposure on X-axis, Negative Density on Y axis.
Note: Colors of lines change (thank your Bill Gates).

ic-racer · Oct 14, 2012

Resulting slopes.

ic-racer · Oct 14, 2012

ic-racer · Oct 14, 2012

Conclusion:
The color of sensitometer light does not affect Tri-X response to changes in development time.

michaelbsc · Oct 14, 2012

ic-racer said:
Conclusion:
The color of sensitometer light does not affect Tri-X response to changes in development time.

Well done, sir!

gmikol · Oct 14, 2012

This is great. Thanks for taking the time to to do a controlled test.

Knowing that the type of sensitometer doesn't make a real difference, I might abandon my plans to build my own, and see what I can scratch up used.

Could you post the entire curves (toes and shoulders)?

--Greg

Rafal Lukawiecki · Oct 14, 2012

Thanks, IC. This is very useful, considering my recent barrage of sensitometer-related questions on other threads. I suppose other films should be similar in ignoring the sensitometer light source design, but this is only a hypothesis for now.

Now, I would like either of those, please.

Bill Burk · Oct 14, 2012

Thanks ic-racer,

I believe this shows all the sensitometers you tried are usable for process control. It's better agreement than I expected.

The choice of sensitometer makes a difference in the 2nd decimal place of accuracy.

But if you want to process to a consistent, predictable contrast... Any of these tools will do the trick.

Photo Engineer · Oct 14, 2012

Note that the exposure with the EG&G gives higher speed due to its being "white" light where the LED types are not truly white. Therefore, part of the energy is missing which leads to slower results.

The LED sensitometers are good but will give you what amounts to slower speed.

PE

ic-racer · Oct 14, 2012

Photo Engineer said:
Note that the exposure with the EG&G gives higher speed due to its being "white" light where the LED types are not truly white. Therefore, part of the energy is missing which leads to slower results.

The LED sensitometers are good but will give you what amounts to slower speed.

PE

Thanks for viewing the thread.

Don't read anything in to x-axis variables on those graphs. The x values are just "numbers" or "units" not Lux-seconds. The sensitometers all have different intensities and exposure times and I'd consider the relative right-left position of each curve on the x-axis in the graphs to be random. This was not a film speed test. (That is to come, stay tuned!)

Rafal Lukawiecki · Oct 14, 2012

Do you have a similar, prior experience with a Kodak Process Control Sensitometer? I'm not familiar with any of those, but looking for one.

Photo Engineer · Oct 14, 2012

ic-racer said:
Thanks for viewing the thread.

Don't read anything in to x-axis variables on those graphs. The x values are just "numbers" or "units" not Lux-seconds. The sensitometers all have different intensities and exposure times and I'd consider the relative right-left position of each curve on the x-axis in the graphs to be random. This was not a film speed test. (That is to come, stay tuned!)

I fully recognize that the intensity of these instruments may be far different. I should have made that post of mine conditional in that respect. Sorry.

If you have any degree of congruence in intensity at one selected wavelength or group of wavelengths, then if the speeds differ in the exposed films, then the spectral output of the group of instruments varies.

Since we see that here, then either the absolute intensity of each device varies or the spectral output varies.

PE

ic-racer · Oct 14, 2012

Photo Engineer said:
I fully recognize that the intensity of these instruments may be far different. I should have made that post of mine conditional in that respect. Sorry.

If you have any degree of congruence in intensity at one selected wavelength or group of wavelengths, then if the speeds differ in the exposed films, then the spectral output of the group of instruments varies.

Since we see that here, then either the absolute intensity of each device varies or the spectral output varies.

PE

Relative speed will be tested within each sensitometer. So, if sensitometer "A" shows 2 stops between two films and sensitometer "B" also shows two stops between those same two films then the response of the two sensitometers will be deemed OK for home darkroom film speed comparisons.

I know a few posters have shown some interest in absolute sensitometer calibration, but only about 5 of us here care about that, and I'm not going to address that. For everyone else, they just need to know how much faster or slower some oddball film from another continent is, compared to their favorite film that was just discontinued.

Ideally we'd test a number of films with different spectral responses, but lets see how just two films compare. It could be that the two films response to the LED units are so different that further testing is not needed. If the first test is close, I can keep going with other films.

ic-racer · Oct 14, 2012

Rafal Lukawiecki said:
Do you have a similar, prior experience with a Kodak Process Control Sensitometer? I'm not familiar with any of those, but looking for one.

I have no experience, but I'd be surprised if it behaved any different than the others in the first test.

ic-racer · Oct 14, 2012

I have a lot of information on the construction and method of operation of each of these three units. I'll post more about them when all the results are in. Having said that, I do want to say that the ESECO unit is by far the most sophisticated of the bunch. It actually has photoreceptors in the light chamber that monitor the LED intensity and correct the time based on the intensity. That takes care of any warm up and decay issues with the LEDS. It also addresses any LED intensity changes as the battery ages. As a practical note, the ESECO prints a much nicer step wedge on the film (compared to a Stouffer wedge). The numbers are clearer and it is much easier to align the test patches with the densitometer aperture and still see the wedge number.

Bill Burk · Oct 14, 2012

ic-racer,

Not to put you to more work, but can you do a quick Reciprocity study using the EG&G at different exposure times?

I think it would be valuable to show how different the gradient might be from a home-made "flash" sensitometer used at low light output (for example time might be 1/10,000 second at lowest setting).

I'd do the test but I don't have the acetate filters to balance the different speeds, and I think you made a set...

ic-racer · Oct 14, 2012

Bill Burk said:
ic-racer,

Not to put you to more work, but can you do a quick Reciprocity study using the EG&G at different exposure times?

I think it would be valuable to show how different the gradient might be from a home-made "flash" sensitometer used at low light output (for example time might be 1/10,000 second at lowest setting).

I'd do the test but I don't have the acetate filters to balance the different speeds, and I think you made a set...

Should be easy to do. I think it will turn out OK. The Wejex in the test had an exposure time of 1 sec and the EG&G was set to 1/1000. There was not detectable difference in the way the exposed film responded to development time changes between those two. So, I suspect 1/10,000 would be similar.

Mr Bill · Oct 15, 2012

Rafal Lukawiecki said:
Do you have a similar, prior experience with a Kodak Process Control Sensitometer? I'm not familiar with any of those, but looking for one.

Rafal, many years ago, I used a Kodak Process Control Sensitometer (model 101, I think) quite a lot, so could answer some questions on it. Basically it's a fairly large, heavy machine using an expensive tungsten lamp. (They were expensive because each bulb came with a calibration certificate, listing the mount-position specs, as well as operating amperage at a spec'd color temp.) The sample mounting platten is curved, presumably in an arc around the lamp filament position. Exposure is a single time, ~ 1/5 second(?), via a rotating disc shutter with a cutout sector. They came with a 2-step inconel-coated filter, so you had two density steps to choose from (removing the filter was another option). Finer adjustments were done with gelatin filters in a filter drawer. That's pretty much about it.

In my job, we used it for paper testing, but found the exposure-time limitation too restrictive. We had it modified with a stepper motor to rotate the shutter disk, so we could dial in different times to match the machine printers we used. I'm not sure that one of these sensitometers has much practical use today.

Rafal Lukawiecki · Oct 15, 2012

Thank you Mr Bill, and thank you IC. I seem to have just picked up a Speedlight SL-2 on that site, driven mainly by the thoughts expressed on this thread. I am looking forward to trying it when it arrives, assuming it is in a working order.

By the way, the Kodak unit looks very similar to X-Rite 394, judging from photos, but I do not see the model number on it.

Bill Burk · Oct 15, 2012

ic-racer said:
Should be easy to do. I think it will turn out OK. The Wejex in the test had an exposure time of 1 sec and the EG&G was set to 1/1000. There was not detectable difference in the way the exposed film responded to development time changes between those two. So, I suspect 1/10,000 would be similar.

I am just curious because common advice to compensate for reciprocity failure is to develop 10% longer when using 1/1000 second. So if that advice is based on sensitometry, a 1/1000 curve should have a lower gradient than a curve exposed for 1/100 second (and developed for the same time).

You were on 10^-3? (I usually use 10^-2).

Rafal Lukawiecki · Oct 15, 2012

How do you control the exposure of your ESECO Speedlight SL-2, if I may ask? I was hoping to use it with both 400 and 100 film, and it would be great if I could get the entire set of steps, except for the first one or two, exposed enough to read a density above fb+f. Do you use filters, multiple exposures, or a modification of some sort?

ic-racer · Oct 15, 2012

Bill Burk said:
You were on 10^-3? (I usually use 10^-2).

Tests were done at 10^2, 1/100. Earlier post is a typo, thanks for catching that.

Bill you should be able to test that. You don't need the EG&G line filters, just any ND to get both exposures somewhere on the step wedge to plot a slope.

The Great Sensitometer Shootout....

Privacy & Transparency

Privacy & Transparency