I believe the request for a scan from me, which I regrettably just came across less than a week ago when this thread sprang back to life, was for information from 'Copying and Duplication'.
I have been busy hopping around on other projects, so the scan of the technical meat of densitometry found in that book has not floated to the top of my things to do list. So hear is the meat of the information from memory
The gist of it was that you shoot a test frame with varying levels of pure grey from white to black. I have used a positive target found in an old Kodak darkroom databook. I have also used a positive test slide made by beselar sometime in the 70's or 80's that has something like a macbeth color chart in the slide. I suppose if you are missing these sources, you could just deliberately under and progressively over expose a shot of an evenly lit grey card.
The goal is to get patches in a negative, or a series of negatives, that will show the black to white subjects are differing densities of grey in the negative.
You take readings with the different colour filters of the denistometer, and plot them on log paper. I had old log one axis, linear second axis paper from earlier electrical engineering courses designing and testing cut off filters. There are web sites that let you specify what you want, and the site generates an image to print off with the scales to suit your data range.
You are aiming to see that the negative density response is linear when plotted on the log scale, and that the different red, green and blue plots are a constant distance apart from each other. My first trials showed that I had too many values on the toe, and that the top end of the curve swept up, and that the spacing between the three traces wan not equal.
Pass two was exposed with a stop more exposure, ie using a lower iso value to meter by, and my c-41 development used a bit less agitaion, and the developer was poured out a bit earlier so the stop went in at exactly 3:15. A plot of the red, green and blue filter values reading off the varying density of the grey patches showed that I had got the readings for the dark positive targets off of the toe, and into the linear range of the plot. The plot lines were also much more linear at the higher end. I just had the inter-plot spacings to iron out.
Since the spacing are on the log axis I knew from reading the scale that I need to cut about .3 from green response.
So I sifted though a stash of gelatine filter squares I have squirelled away, and found a 30cc green. It went into the optical path. In my case it wnet under the neg, to filter the colur of the light source, It just as well could have gone over the image making lens.
Exposed the film on trial 3 , giving a tad more exposure to make up for the light loss of the 30cc green filter, and then processed with care. Read and plot pass 3, and I finally had the old 100' roll of expired but cold stored dupe film response dialled in.
Plots between red, green and blue were evenly spaced, off of the toe, but just, and the white positive values ( densest negative values) stayed on the linear part of the response curve.