Just for fun, I tried to model your Turnigy 3548(800) and compare it to an AXI2826-12.
I used the free DriveCalc program. It actually has both motors in its database, which means someone made some measurements on the motors and inputted them into the database.
What I did was set them up with an APC11-5.5 TE prop--because the 11.5.5 prop nominally draws the same current statically as the APC12-6 TE prop. Then I tried to get your observed inflight RPM (~9500 rpm) by adjusting the "voltage". I'll post two png graphic files of the two motors in this config. What is nice is DriveCalc tells you some useful info on the motors---more than you get from the manufacturer usually. I note the Turnigy is slightly smaller than the AXI 2826---maybe the 2820 would have made a better comparison.
I do note that Drivecalc claims the effective kV for the Turnigy is 702 rpm/V (the "n
s" parameter. Compare to the 800 rpm/V manufacturer number.
The green dots on the curve show the 9500 rpm points. The blue line is rpm vs current, the greenline is the motor efficiency, and the red line in the OUTPUT power (to the prop). If you look at the set of numbers on the lower right hand side of the window (just above the reliability sliders), the values for the green dots are given, including the power input (what we normally measure). I note the AXI claims it needs 394 watts to the Turnigy 440 watts to turn the prop at this rpm. I also note the "dynamic" resistance for the AXI is 64.6mOhm and the Turnigy is 166mOhm, a considerable difference. How much of that is due to the smaller size of the Turnigy and its lower kV (more winds) and how much to the winding packing efficiency--I don't know.
But anyway, from this you can see that the Turnigy needs more power than this AXI, and that extra power is basically turned into heat.
They didn't have Roger's original motor, so that's why I chose the AXI, which I bet is larger than the Turnigy, but smaller than the original motor.
Enjoy!!
Note after---Added my Scorpion SII3020-780 plot. This is the motor I use on the Vector. It is pulling 394 watts to turn the prop (coincidently the same as the AXI.)