I'm not much help but I mixed it up only one time as a quick review of a group of developers that I was going to take a snapshot look at. I ended up with thinner negs than expected and out of the half dozen or so things I tested, I just didn't get back to it. I ended up usiing Ryuji's DS-10 instead and then got back into traditional emulsions and enjoying the easy results of PyroCat syle developers with those films. Would you keep us posted as to your experiences as this was one I was intrigued in checking out again?
FX-37 is a very fine developer, much like FX-39. Fine, sharp grain, more acute than Xtol, a bit of a shadow speed increase over D76, and tends to make a little bit of a shoulder at a usable density ( Zone X or so ) with TX or TMY.
Like many diluted developers, if the first negatives are too thin, increase the development time until you get the midtone density you want. Then, fine tune the highlights by adjusting agitation.
I have not used FX-37 but have some info on file.
According to Geoffrey Crawley, who invented both, the closest published formula to FX-39 is FX-37.He published a slightly modified FX-37 in Amateur Photographer 15 Oct 05, in which the original benzotriazole and 0.5g Potassium Bromide are replaced by 1.0g Potassium Bromide.This is a move to help the amateur as the benzotriazole was hard to dissolve.He said the omission of benzotriazole has no effect on image quality as it was originally included for the conditions of professional re-use as fog levels can rise.
I played with this again today after noticing an error when I copied the formula to my spreadsheet lists. With the correct bromide levels I found that I had slightly too much density. It seemed very active and I am going to try it at 1:5 to extend the times a bit and also see what the grain looks like. I was very surprised at the very fine grain with slightly more solvent effect than expected (at 1:3). One can't tell until printing of course but this certainly looks promising for filling a speed-capable developer need. I decided to use 1 g/l of bromide and omitted the benzotriazole and had nice, low fog levels.