Isn't Kodak Flexicolor RA bleach replenisher for paper? Do you mean Flexicolor Bleach III Replenisher #6600258?
No, Eric is referring to #66000274, which is the "Rapid Access" version of the C-41 bleach, which requires just 1:00 for bleaching color negative film (versus 6:30 when using the conventional Bleach III).
The practical wisdom on C-41 bleach is that it can be used many, many times over, contrary to what the manufacturer recommendations are (e.g., where Kodak says the Flexicolor Bleach III has twice the capacity of the developer). People report using it almost indefinitely, but I can't attest to that, as I do what you do -- use it twice and dispose. That's probably wasteful on my part but I just don't process enough C-41 film to worry about bleach economy (I usually end up being more worried about shelf life of unused concentrates).
One thing worth investigating is if you're using more chemistry than is really necessary per roll. You report following the Flexicolor instructions regarding chemical capacity -- e.g., three rolls (80 in
2 each) per liter of solution, or ~330 mL of chemistry per roll of 135-36 or 120. If you look through these forums, you'll find that the question of minimum Flexicolor chemistry required to achieve correct development has been discussed many times over, though with resolution coming mostly in the form of anecdotal reports only. The general-ish consensus seems to be that the Jobo minimum requirement of 170 mL for a 120 roll in a 2500-series tank (which is based solely on the threshold to achieve submersion) appears to also be adequate in terms of developmental capacity (assuming an "average" roll made up of Zone V exposures). The only way to quantifiably test if 170 mL is truly adequate for each 80 in
2 of film is to put a 120 film into the tank along with a control strip, but that would require mixing 120 film with 135 (the control strips are short lengths of 135 film), which there's no easy way of doing with a Jobo reel. I tried testing a more extreme version of this about a month ago by developing two rolls of 120 in 270 mL of chemistry in a Jobo 2500 tank (so, 135 mL per roll) with a control strip taped emulsion side up to the inside wall of the tank. In that test, I got less-than-satisfactory results in the control strip. For me, that suggested that 135 mL per 80 in
2 of film probably isn't adequate. (To be fair, the negatives seemed to scan fine; only the control strip indicated a problem.) I'd like to pursue further tests to find the minimum volume required, but for now, my gut instinct is that 200 mL per 80 in
2 is a safe bet.
Regarding developer replenishment, my understanding is that, although not impossible in rotary systems, the frequent agitation and high rate of oxidation in something like a Jobo means that the published replenishment rates (which are typically calculated for dip-and-dunk and roller transport machines) don't apply. I personally use the developer and fixer one-shot, but one could presumably test various replenishment rates with control strips. I'm guessing the replenishment rate would need to be quite high to overcome the oxidation of the developer (assuming, of course, your objective was to stick with the standard 3:15 development time).