When is a PT-6 like a Barbeque?
October 23, 2019
November 13, 2019
Show all

Overpressure valves: Stall or loose the torque?

The Pratt and Whitney PT6-42 engine makes 850 hp at sea level, and it just keeps dishing out that huge torque even at high altitude where the air thins.

How does it keep dishing out all that torque even in thin air?

Simple: They built the compressor with such a huge pressure differential across the blades that they can ONLY bear the compression load above 90% compressor RPM!

Just like an airplane wing can only bear the weight of the airplane above a certain SPEED without stalling, the compressor can only bear the load of the pressure increase in the engine above a certain compressor RPM without stalling!

Compressor RPM is Ng, measured in percent.

So an airplane might need 60 knots to bear the pressure differential needed to hold the airplane aloft… slow down below that and you stall the wing.

The engine needs 90% Ng to bear the pressure differential across the blades.. slow down below that and you stall the compressor!!

Same thing:

Don’t go below 60 knot stall speed unless you lighten your load  on the airplane.

Don’t go below 90% Ng stall RPM unless you lighten your load on the compressor.

So what is a compressor stall like?

At LOW altitude and power, a compressor stall is a BANG! sound, a sharp BANG like someone hit the engine hard with a hard rubber mallet. You hear it and feel it in the airframe. It doesn’t just happen ONCE, it’s a BANG! BANG! BANG! until you get hat power back to idle and just take the pressure load right off the engine and abort your take-off, as I did just now before writing this.

At HIGH altitude (28,000 ft with high power), the air seems ‘softer’, and it is more of a WHUMFF! WHUMFF! WHUMFF! like someone is hitting the airplane with a big over-stuffed pillow as the fuel is suddenly cut.

So you don’t want a compressor stall because you lose power and it sure feels like you might hurt the engine at low altitude.

So how do we avoid these compressor stalls at LOWER compressor RPM?

Simple: We have to UNLOAD the crazy-high pressure from the high-pressure side of the compressor at low compressor RPM, just like we would unload the wing of an airplane to avoid a wing stall.

We do this by getting RID of that high pressure air on the high-pressure side of the compressor, so that high-pressure air won’t force it’s way backwards through the compressor and stall the blades!

So at low compressor RPM, when the compressor blades can NOT bear a large pressure increase without stalling, we need to RELIEVE pressure on the hi-pressure side of the engine by simply dumping it overboard. Once the compressor RPM gets high and can bear the load, we don’t have to dump the air overboard any more, and we keep that super-high pressure on the hi-pressure side of the compressor and use it for torque and power and SPEEEED!

So how do we control when the high pressure and therefore high load on the compressor is:

1: relieved (as it must be for low-compressor-speed operations)


2: steadfastly HELD (as we want for high-compressor speed, where the huge pressure increase can be maintained, with the resultant increase in torque and power)?

The answer is simple: Overboard ‘dump valves’.

These two valves are:

1: Lo pressure on the left side of the engine, dumping extra air overboard until 80% Ng,

2: Hi pressure on the right side of the engine, dumping extra air overboard until 90% Ng,

each valve lets air flow right out of the middle of the engine at the low Ng to keep the load across the compressor low, but then close up and make the engine air-tight as the compressor speeds up.

Lower the load at low Ng by LEAKING the air OUT of the hi-pressure side of the compressor.

Maximize the load at high Ng by HOLDING the air IN on the hi-pressure side of the compressor.

So what happens if these valves are mid-configured? (everything on a PT6 is analog like an old-school steam locomotive and tweaked and adjusted to spec).

The answer is:

If the valves are adjusted too LOOSE, to let the air OUT too much, then we never get that maximum pressure differential across the compressor because the air is dumped overboard and we just don’t make enough POWER!

If the valves are adjusted too TIGHT, to keep the air IN too much, then we get a high pressure differential across the compressor at low RPM (a differential that can only be handled at high RPM) and the compressor STALLS!

So BOTH of those outflow valves must be adjusted to spec.

NOW that you understand how these things work, it’s time for the war story:

My bird, N844X, was NOT making much power at altitude.

Redline temperature, but not too much torque.


Suspicion: An outflow valve was not closing, so air was leaking out and stopping me from making torque.


LO PRESSURE VALVE is SUPPOSED to close at 80% Ng, but a little rubber hose duct-taped to the engine on the ground for ground runs, and going into a bottle of water to show when air was coming out of it, showed it was ACTUALLY closing at 70% Ng! Hmm.. it’s closing too EARLY! We should expect a COMPRESSOR STALL! But we observed NO compressor stall, sooooooo…. it must be fine, right? It’s fine. Everything’s fine.

Pasted Graphic.tiff

Now, the HI PRESSURE VALVE is SUPPOSED to close at 90% Ng, but a bar-b-que temperature probe in the exhaust duct beaming temperature to my phone by bluetooth in flight showed the hi-pressure valve was still OPEN at 94% Ng… we know it was still super hot in the duct!

Ahhhh-HA!!!!! PROBLEM FOUND! The high-pressure valve is too loose, letting air out of the engine, and I am not making torque! Let’s FIX that!

And the lo-pressure valve being too TIGHT and closing too SOON? Let’s ignore that. It’s fine. Everything’s fine. No compressor stall, so no reason to tweak it, right?

Pasted Graphic.tiff

SOOO the hi-pressure valve was sent in to a shop, adjusted, and sent back.

NOW we got a nice TIGHT hi-pressure valve!

OOOO! WAIT! That means we have a lo-pressure valve that is way TOO tight, and hi-pressure valve that WAS too loose and is now just right!

That means the engine overall is now too tight, right??? …. umm…. but we never has a problem before with the engine being too tight… so it must be OK…..

Pasted Graphic.tiff

Cowl on, onto the runway, add power, BANG!!! BANG!!! Compressor stall!!! Compressor stall!!!

Abort take-off.

Can you guess what happened?

Initially, he LOW pressure valve was too TIGHT, and the HI pressure valve was too LOOSE, so they cancelled out at low speed and we got not compressor stall… we just could not make full power.

BUT, once that HI pressure valve came back with a proper adjustment…. BOOM: The too-tight lo-pressure valve was suddenly a problem, since there was no too-loose hi-pressure valve making up for it.

So that too-tight LO-pressure valve is actually a no-go, now that the HI-pressure valve is properly set!

So here’s the thing: When we saw the too-tight lo-pressure valve, and said “Well if you aren’t getting any compressor stall I guess it’s OK….”, we did not consider that the hi-pressure valve being too LOOSE was allowing us to get away with a lo-pressure valve that was too TIGHT!

And once we tightened the hi-pressure vale, the too-tight lo-pressure valve would make itself known with compressor stalls.

So here’s the lesson: Don’t just check your pressure on ONE valve, do them BOTH, and don’t allow EITHER of them to get out of spec, because an error on one might offset a different error the other… until you fix just ONE of them and find yourself with a weak or stalling engine.