I've had the chance to twiddle the friction constants in a turbine jet I modeled. You can, in theory reduce the second motor friction constant down to about 0.28 and get less internal engine friction.
My problem with the jet was increasing the rotation moment of inertia to simulate the rather long spool-up and spool downs of a true turbine engine, and of course the rather small outlet size of the engine was modeled too. I had to make 2 PAR entries for the engine with LEFT entered in the second one to counter the extreme gyroscopic forces of the first RIGHT rotating engine. The problem with reducing the friction is that the engines, once spooled up, are reluctant to spool back down again. I ended up having to come in hot from the north of the larger valley in the Switzerland .scn and hold a high alpha flare nearly the entire length of the scene just to land. So I've modeled BIG spoilers on this model.
Remember we're just spinning electrons here, the constants for FMS PAR file have a huge amount of flexibility and you can get over 22Kgs. of thrust out of a theoretical 100 cc engine with a 5" diameter, 6 bladed propeller disc that has over 40 grams per second of fuel fed into it. There you go, virtual internal combustion jet turbine, just add proper .WAV file and go shred some atmosphere.
Lewis