After an extensive troubleshooting process, the problem was with the wire connecting the fan to the controller board under the printer. The cable was wired incorrectly (in the factory) so that the positive output on the board was connected to the negative (or ground) input of the fan and vice versa.
This would short out the fan motor (killing the fan) and the burn out the Darlington on the controller board when the fan was switched on for the first time. Before you connect another fan, you will need to correct the polarity of the connector wire and possible replace the Darlington.
From your multimeter readings I suspect that your Darlington may be ok, but you will need a new fan.
Here are the checks that support suggested to check the Darlinton:
1. Check for continuity between the middle leg of the Darlington and the right hand pin of the fan connector on the board (right hand pin when you are looking at the connector from the "outside" of the board - i.e pin furthest from the Darlinton). There should be continuity.
2. Check for continuity between the other pin of the fan connector and the left pin of the connector for the big (centrifugal) fan. There should be continuity.
3. With the fan disconnected from the board but switched on via the software, there should be 19V between the pins on the board.
4. With the fan disconnected from the board but switched on via the software, there should be 5V between the middle leg on the transistor (GND) and the +5V pinout on the board near the LED header. If you get continuity in step 1 above, it is easier to use the right hand pin of the fan connector rather than the middle pin of the Darlinton to test here.
If the above all check out, the Darlington should be ok and it's just a case of swapping the pins in the connector wire and installing a new fan in the print head.
I hope this sheds some light on your problem.