fly911,
"In other words, the tail rotor does not go through the main transmission (MRGB)"
There's always exceptions, eh? Even on that Allison model 250 you show, I suppose I could argue that the full engine torque passes through part of the transmission gear train. The same issues would still be relevant, the only difference being that the affected gears happen to be within the engine case and not the MRGB case proper. Regardless, I'll choose my words more carefully next time.
victor pappa,
"All the training manuals state clearly however that the tQ function is not to protect the engine but the MGB despite the fact that for all the tQ reading is taken on the engine and not the MGB."
The engine and gearbox guys always point their fingers at each other when it comes to the issue of what hardware is the cause of the torque limit. The gearbox guys will say the torque limit protects the long, slender power turbine drive shaft. And the engine guys will claim that their engine can easily make more torque than the gearbox can ever handle. The truth is it's usually a little of both.
The torque meters themselves are almost always supplied as part of a turboshaft engine, since they must provide feedback to the FADEC. The old method of torque monitoring was via pressure measurement in a hydrostatic circuit as you describe. The mechanism was actuated via axial displacement of a helical gear or helically splined shaft. This system was originally developed for monitoring BMEP on large aircraft piston engines. Newer engines use electronic systems, usually employing hall effect sensors and trigger wheels on the engine output shaft that measure torsional displacement.
Regards,
riff_raff