[QUOTE=Mike Wilde;1423071]There is no interconnection in the tanks piping systems, so it must be at the tank level.

My guess is that the spray bar in some manner is mis aimed, or has too many outlets plugged with cruddies, so that the pump is squirting liquid out where the piping joins the spray bar.

The public darkroom comment makes me suspect that the machine needs a good cleaning, roller scrub, and spring and spray bar inspection. I had good sucess cleaning the spray arm with a longe scrubby thing marketted at claning glass baby bottles

A transfer rack clean off also would help.

Maybe the overflow of the developer tank is somehow plugged, and the overflow of tank 1 is being washed into the next tank, or event the third tank along the end areas if the machine is not mounted on a levellled surface?.

I find myself curious how they run theirs. The first tank is usually deevloper, because it has the most accurate thermostat ( or at least mine did until it went flaky.)

The last tank I thought would be blix, with a water bath of stop between them. Maybe they run tank 2 as blix.

These are all excellent suggestions. However I'm sure that the school is running the blix in the second bath as this is how the majority of people set up this machine, especially if they have the wash module. However, I'm curious as to how the op has determined that it's developer that's flowing into the second bath. Given that the unit is in a gang drkrm, they most likely have it set up with the replenisher unit. If the sensor bar is grimey and the developer replenisher feed is clogged (much more likely than the blix replenisher feed being clogged) then the second bath would non-stop replenish and the developer bath would just evaporate - a scenario that would match the op's. More careful observation is needed.