I made a speaker with a 2" diameter voice coil made with 26 gauge wire that measures a resistance of 1.5 ohms at 50hz.
It's based off the design of headphone drivers and I have it suspended on a 8" diameter cellophane diaghragm centered over a magnet.
I wasn't too sure what the sensitivity was, so I hooked it to the output of my computer speaker amplifier and played a 50hz tone at 1w input (right at the .8 amp current limit before it shut off) and I got 60dB at 1w/1m, so you can see how insanely inefficient this thing is.
It's very demanding for current at such a low impedance and sensitivity, so that is a big problem.
After hooking it into my bedroom's receiver, I got 20w from it at the highest level before shutoff.
73-74dB is all I got out of the thing at 50hz. This speaker actually puts out mostly bass for some reason (I would think the inductance is pretty low..?), which might be the extremely low fs. Excursion was slightly over 1mm with 20w input.
When I get the amp for the subwoofer system, which tested 4500w at 50hz at 1-ohm in parallel-mono.
If the driver could handle 4000w of input without tearing the cellophane or overheating of the voice coil, then I'd assume over 90dB of output at 50hz.
I might try it just for fun and to prove my homemade speaker works if you give it enough power.