Testing Some New Film (Shanghai 8x10)--How I Do It...

[ic-racer]

I have never used Shanghai film, so before I load it into the film holders, I'll test it for its sensitivity to light and development.
I use HP5 almost daily, so Ilford HP5 will be the comparison film.

 

[ic-racer]

I'll use a Wejex Sensitometer (by Tobias) and Tobias Densitometer pair for the test.

 

[ic-racer]

I will cut the 8x10 sheet into strips to get multiple tests from each sheet. In this case I cut to 120 film size for easy processing. I set the backstop on my paper cutter for 61mm to cut the film in the dark. I make note of the film notch and clip a corner of each strip to identify emulsion side.

 

[ic-racer]

In the dark I expose a strip of Shanghai with the sensitometer and load it on the 120 reel. I also remove a strip of HP5 (in the bulk loader) and expose it with the sensitometer and load that on the other reel.

 

[ic-racer]

Process the film in the usual manner, making sure the developer has stabilized at the appropriate temperature.

 

[ic-racer]

Even before the film has a chance to dry, you can see the new film is about six of the half-stop patches away from HP5 (20-14=6) without any specialized math. That puts the film at EI 50. But I'll use some specialized math anyway...

 

[ic-racer]

The dry film is measured and plotted for density. I already know the values of the step wedge in the Wegex sensitometer to use as the input values for the H&D curves. I use a spreadsheet that finds the film speed using a mathematical approximation of one-third the average gradient. That saves me from having to make a whole family of curves to find one that fits the ASA triangle.

The stepwedge in the Wejex is permanently affixed in the machine to project the numbers with the correct orientation. Just as a reminder, in case one forgets if the emulsion should be up or down when making the readings. (Emulsion UP).

 

[mshchem]

Nice work! Very good information.

 

[ic-racer]

After plugging the values into the spreadsheet, I can view the two characteristic curves and the relative film speeds, compared to Tri-X which I had previously tested and assigned ISO 400 in the software.
The software gives more significant digits than necessary, so EI 400 for HP5 and EI 50 for Shanghai 100 would be the final results.
Again, since I use HP5 almost every day for everything from 35mm, 120, 4x5 and 8x10, I know where to set my meter to get a good exposure.
Now I know the Shanghai is THREE STOPS slower, so I can set my meter appropriately and start exposing film tomorrow. I know this after using less than one sheet of 8x10 film.

 

[guangong]

Very impressive. If you use Shanghai film over a period of time, would it be possible or worth while to monitor maintenance of quality control? Could there be significant variations in speed from batch to batch?
Again, this was a very fine piece of work! Thanks.

 

[Raghu Kuvempunagar]

Nice work!

Were both strips of film developed for the same duration of time? Is the recommended time for both films the same for the developer you used?

 

[ic-racer]

Nice work!

Were both strips of film developed for the same duration of time? Is the recommended time for both films the same for the developer you used?

Yes, for the comparison both films were processed together in the same tank. Since I already know how many minutes to process the HP5 so it prints well with my diffusion enlarging equipment, I can see the Shanghai will need to be processed a little longer to match the same printing contrast.

 

[ic-racer]

Very impressive. If you use Shanghai film over a period of time, would it be possible or worth while to monitor maintenance of quality control? Could there be significant variations in speed from batch to batch?
Again, this was a very fine piece of work! Thanks.

Yes, in a way, but not the best however. When one goes to the hardware store to buy a ruler, one expects it to measure with reasonable accuracy, however, it is by no means an ISO standard. Likewise when I buy HP5 to use as a benchmark for testing other films, I expect the fresh Ilford film to be close to ISO 400, but Ilford is not necessarily free from small variations too.
Realistically, in 30 years of doing this, the major thing that affects film from batch to batch in my darkroom is base fog. Sure one can 'print through' the fog, but the fog does affect the film speed.

One example of this is a batch of HP5 I have from 1984 or so. Every decade or so I test a little bit of it. I'd have to check through the archives, but I believe the last time I tested it the base fog was around 0.65 or so and the speed around 100 using a fractional gradient method.

 

[Mainecoonmaniac]

I ordered a box from Ebay this year and ran some test. It's not bad film. I'm thinking about using it for my 4x5 pinhole camera because it has better reciprocity than Arista EDU 100. I looked at the images shot with Shanghai film on Flickr and it prompted me to try some.

 

[Andrew O'Neill]

I have a few boxes of 4x5 and 8x10. I used the 4x5 for my testing by exposing a few sheets with a step wedge, under the enlarger. Drew up the curves to determine N times and EI's. Tested it out in the field along with 8x10. It's a really nice film. Got a couple of images in my gallery...

 

[ic-racer]

I processed the first negatives, they are in the washer right now. The images were low light, near the end of day (4:00 PM this time of year; I think it was the shortest day). Shadow detail is not what I usually get with HP5, but the images will likely work as they are of water. With 400 speed film the shutter speed is just not right for water. Too slow shutter speed to freeze the water, but too fast to make it look nice.

 

[Raghu Kuvempunagar]

Did you do tests for reciprocity failure? Would be interested in knowing your findings.

 

[Mainecoonmaniac]

If you have an iPhone, this app lists Shanghai as one of the film. Great app.

https://apps.apple.com/us/app/reciprocity-timer/id459691262

 

[Bill Burk]

Can you share the “raw” data, then I could draw graphs and check “delta-x” speed, etc.

 

[ic-racer]

Yes, they are on my other computer but I'll post them so you can put them in your software.

As a review, fractional gradient speeds were difficult to obtain from a hand-drawn H&D curve, so there was a paper in 1956 "Simple Methods of Approximating Fractional Gradient Speeds."

The two methods involved clear plastic cutouts that could be placed over the hand-drawn H&D curve (W-Speed and Delta-X).

W speed meter was in two pieces with a pivot:

 

[ic-racer]

The Delta-X meter was a single plastic cutout:

 

[ic-racer]

Although the Delta-X looks easier to do on a hand-drawn H&D curve, if one has the H&D curve data in computer format, I fount it easier to calculate the W-speed.
For example, the W-speed relies on this calculation of "W" which is obtained by triangulation and can be done with simple spreadsheet program functions.
Looks complicated but break it down. Slope is easy to get. Then the Y-intercept is easy to get ("A"). With "A" one can easily intersect straight up to get "E" by stepping though the dataset until the two surrounding points are found and interpolated. If we know "E" and one-half the slope, the second Y-intercept is easily found at "C." Again the spreadsheet it calculating these Y-intercepts for me.
Once the two Y-intercepts are obtained the key value "W" is found with simple subtraction of "C" - "A."

Once "W" is found the fractional speed is easy math (all the numbers in the equation are empiric constants):

Fractional gradient speed = -0.0948 + (0.0122 *W) - (2.2945 * W^2)

 

[ic-racer]

The tables did not cut and paste nice, so I made screen shots:

 

[Bill Burk]

You developed these to a pretty high Contrast Index. I used a modified Delta-X meter, like the picture you showed but shifted 0.29 to the right. If you have an 0.62 CI curve of HP5 I could double check, but it looks like this works backwards to 0.40 mcs (-0.4 log mcs) for the Wejex. (Do you have an estimate? Is that about what you think it is?)

Exactly three stops difference by Delta-X paper graphing. So the advice I would give matches your advice...

 

[ic-racer]

Nice, thank you for posing.
I actually have never done the math to actually work backwards to get the MCS of my sensitometer but it should not be hard to incorporate this calculation in my spreadsheet. I believe you shift the Delta-x meter to remove the safety factor, yes? I do this in my spreadsheet, likewise shifting the W-speed to reflect current ISO.

The high contrast index of these test strips could be the start of some other epic thread, but it is due to processing only a short strip of 35mm and a short strip of 120 sized negative in a solution volume that would normally process FIVE 8x10 sheets.
My actual gamma, when a control strip is placed on the tail of a 36 exposure roll, processed as I do with seven other 36 exposure rolls in the Jobo, gives me a gamma around 0.75 with the same developer, and temp and time.

When I first got my Expert drum, I also did the above with the Jobo Expert 8x10 drum. That is I put a sensitometer exposure on a full 8x10 sheet and processed it with 4 other fulll 8x10 sheets to get my actual gamma. But I don't see the need to do that experiment again. I'll just extrapolate the gamma from the over developed strip. But this is all open to constructive discussion.