I solved the DIY densitometer problem. Here's how. I am fortunate to have two UV dose meters: one that controls my UV exposure unit (just set the dose and walk away), and a small battery-powered UV meter designed for industrial use. The small meter is round and flat, like an oversized old-fashioned pocket watch, with the LCD on one side and the sensor on the other (you can get one on eBay -- search for "UV-Integrator"). I placed small meter inside my exposure unit with the sensor facing up toward the bulbs. I placed my negative on top of the small meter's sensor, emulsion-side down. I exposed for a constant dose of 800 units (about 15 minutes). With shorter exposures I could not read densities over 1.5. Anyway, I exposed each of the steps of a 21-step Stouffer 4x5. For each exposure, I recorded the reading on the small meter and the (known) Stouffer step density, then reset the small meter to 0 for the next exposure. This gave me 21 data points. I fit an exponential curve to the 21 points. It was a beautiful fit with R=0.999. This is my calibration curve. Now I can take a sample negative, expose it in exactly the same way, and use the reading from the small meter to find the corresponding density on the calibration curve (I actually use the equation of the best-fit curve). The only drawback of this solution is the length of the exposure -- fifteen minutes for one density measurement! This is not a good general solution, but it works for me. The samples that I'm measuring are small pieces (say 2cm x 2cm) of solid black digitally printed on transparency film. No "maneuvering" is required to position the sample over the sensor.