Friday, June 24, 2005

More Spinning Things

The PT6A engines drive three-bladed, constant-speed, full feathering propellers. Both rotate clockwise, viewed from the rear. A typical cruise setting of 76% Np is equivalent to 1672 rpm. A tachometer on the side of the reduction gearbox gives a cockpit indication of propeller speed.

Airplanes that work properly are really fairly easy to fly, and anyone can learn to fly straight, turn, and go up and down in an hour or so. The taking off and landing bit can be a bit trickier, but really a few hours and you can do that, too. The hard part is continuing to fly the airplane when things stop working, so most of learning about a new airplane is learning how things work, and what they are likely to do when they stop working.

If you have two engines, and one stops working, the non-working one contributes only drag to the operation, holding back its side of the airplane, and the working one continues to pull its side forward, causing the airplane to yaw towards the dead engine.

To make things worse, Murphy's Law ensures that any non-workingness occurs at the most inconvenient moment possible. For an engine failure, such a moment is when you are using the engines to climb at a low speed, so the axis of rotation of the propellers is angled up relative to the oncoming airflow. This means that the plane of rotation of the propellers is not perpendicular to the oncoming air, and thus the downgoing propeller blades meet the air at a steeper angle than the upgoing ones. The steeper angle of attack means more pulling force for the propeller on the downgoing side. That's the right side for both propellers, but the closer-to-the-middle side for the left engine and the further-from-the-middle side for the right engine. That means that if the left engine fails, the remaining engine has more turning moment than would be the case if the right engine had failed. For this reason, the left engine is called the critical engine.

To reduce the drag of the dead engine, its propeller can be feathered. That means that the blades can be turned parallel to the oncoming air. Normally in flight, oil circulating through the propeller hub holds the blades at an appropriate lower angle, opposing both a set of rotating counterweights and the feathering springs in the hub. When the pilot selects the propeller control to FEATHER, by pulling the propeller lever full aft, a pilot valve cuts off the oil supply to the hub and the pressure drops. If the propeller is rotating, counterweights drive it to the feathered postion, of 87 degrees to its plane of rotation. If the propeller is not rotating, the feathering springs alone will do the job.

Later models incorporate an autofeather system. If the autofeather system is selected on, both power levers are far enough forward that the engine would normally produce at least 86 to 88 percent NG, and the torque of one engine decreases to 11 psi for at least two seconds, the autofeather solenoid valve opens and oil is dumped from the hub, allowing the feathering springs and counterweights to drive the propellers into feather. The pilot has to be careful not to pull back the power lever of the failed engine--the normal reaction-- because that would inhibit the autofeather mechanism.

To test the autofeather mechanism, the pilot (on the ground) sets up all the pre-conditions for the autofeather to work: both engines running, autofeather selected on, power levers moved up to 88% NG, torque on both engines indicating 20 to 30 psi, and waits for the the amber ARMED light to illuminate. Then the pilot lifts one of the AUTOFEATHER TEST switches, which cause the system to not notice if a power lever has been pulled back. Then the pilot pulls back the power lever on the same side as the test switch and waits two seconds. The propeller should feather. The pilot then repeats the procedure for the other propeller.

That is about one twentieth of what I have to know about the propeller system. This system makes me wish it were a jet.


Anonymous said...

That's pretty clever stuff. Oh, and "non-workingness"? Too much time in some rubbish chat-room, I reckon

Greybeard said...

You mentioned the ALWAYS in effect Murphy's law. Here's an example:

I was taking multi-engine training in an Army T-42, (B55 Baron). In cruise, the instructor failed and feathered the right engine. When we tried to unfeather the engine, it wouldn't come out of feather.

Then the alternator failed on the left one!
The battery in the Baron is a pretty weak affair. We had to lower the gear manually, (a gazillion turns of a handle that requires the flexibility of a contortionist to even get at it) and then the radio didn't have enough juice to contact the tower for landing.
A "no radio", single engine approach.......then on rollout we couldn't use differential power to clear the runway and had to have a tow!
Thank God we were VMC!
Twins sure are interesting!