I know, I know, I haven't made my first print yet but I am also pondering larger prints than my 8" roller transport will accommodate. I am thinking occasional 16x20 or perhaps 20x24. Since it would be used infrequently, I am leaning toward a Unicolor drum or two with the agitator base. It seems to me that the drift-by method of temperature control would be ideal in this application. What I would like to ask is: Will it work?
Yes it will work fine. I've done it many times. Just make sure to do a prewet with water of the same temperature as your developer. Also, RA-4 works just fine at room temperature. It just takes a bit longer.
May I assume that tests with water entering the drum at a positive temperature off-set vs. end of process step temperature should average out to the ideal process temperature?
The prewet is not so much for heat loss purposes but for evenness of development. If you were to pour warm developer onto cold paper in the drum, you will get streaks (uneven development). By doing a prewet you avoid streaks. Unlike color film, with RA-4 process the temperature is not very critical. Don't sweat over it.
I was never very confident with the nomograph presoak methods that were advocated with many of the kits back in the 70's and 80's. What temperature did the developer climb to and what was the effective temperature over the course of the developer cycle? Measuring the temperature of the small amount of developer at the end of the developer cycle, especially if using the minimum in an 8x10" drum, was very difficult.
I'm now using a method that I believe is about as accurate as you can get when using a "drift-by" technique. I use a Simmard (raised rib) 16x20" drum. I use 2 litres of presoak water for 1 minute. It is poured in at 103.5deg. F and exits at 97.5deg. F. The developer (1 litre) is poured in at 97.5F. At the end of the dev. cycle the developer is collected for replenishment. The developer exits the drum at 92.5F - hence a 95deg. F average temperature which is the recommended temperature for the brand I use. I should point out that I'm using a developer designed for tabletop equipment that wasn't totally convertible from the EP2 to RA-4 process. (My developer time is 105 seconds versus the normal 45 seconds for today's RA-4.