My experience. I recently dissolved a 5mm or 6mm tap from what was to be a repaired brass feed screw nut. The old nut had been bored out and plugged with a solid brass plug before boring and threading and I was going to securely pin the parts together with a brass threaded screw as well as loctiting the assembly.
The tap was about 20mm, perhaps more, through the hole before breaking, and total thickness was around 25mm. Took some time to dissolve, needed as concentrated a solution as practical and required to be kept very close or at boiling point for any decent reaction rate. Afterwards I wished I had stuffed the reaction vessel in an old pressure cooker, to raise the temperature a lot more!
At one point I experimented a bit and used some small pieces of pure copper – to ensure as free a flow as possible of solution past the broken tap – to support the nut away from the vessel. Solution turned decidedly blue, so the copper chips were removed. Eventually the tap was dissolved virtually completely from the brass . Slow but effective.
So, it definitely works for dissolving steel from brass items. I would use it again, so am keeping an eye out for a sensibly priced alum supply. Small quantities on epay are expensive (used by hobbyist dyers, in particular). It is not aluminium sulphate (as reported on one thread). It is the double salt potassium aluminium sulphate (usually) – with an awful lot of water of crystallisation.