Windmilling propellers
 

I cannot convince a non-aircrew colleague of mine that on large supercharged piston engines with constant-speed-propellers, if you lost an engine during cruise due to fuel starvation, it would continue to windmill at the selected rpm with accompanying boost (manifold pressure) and oil pressure. The only indication of the engine failure -if you missed the initial hiccup of parameters - would be the falling cylinder-head or coolant temperature and yaw and decay of airspeed.
Can anyone point me in the direction of getting written confirmation of this ???

hi
this stuff is complicated, you cant simply find it on wikepedia and so on, I think you should send an email to an aviation authority and get some confirmation
I agree with you, I know its true through my training

Have you considered the control range of the pitch mechanism ?

Unless you have an autofeather, autocoarsen, NTS, etc system operating in cruise flight, you are more likely to end up with a significant drag load on the failed side rather than, as you appear to be suggesting, a magical low drag system ? Typical end result is yaw, roll, and a really good view of the ground below - all of which probably might alert the pilot to an impending problem.

You wont find a written confirmation because the statement is complete wrong.The prop has to go to feather position otherwise the drag will
be too high.The engine will not turn anymore-means no more indications.

Like so many ground instructors have advised: RTFQ, people.

The poster wants info to support his statement that this type of engine would continue to rotate at the selected RPM with a supercharger providing Manifold pressure, etc.

His point: the engine will spin at the same RPM, but you'll get yaw and drag so watch out for the other indications.

(I'm not a piston guy, so I'm not one to judge the veracity of his statements.)

Ok I read the question :)and sorry no I can not point you to source where you can find information on this. But I think you are getting things muddled up.
If an engine would fail due to fuel starvation, the engine`s power output would start to get less there for the prop governor will decrease pith from coarse pitch to fine pitch to maintain RPM.(reducing the force it takes to turn the propeller) just like when you reduce power during descent/approach. You can even hear the the propellers running to fine pitch (change in sound and vibration). Secondly no piston engine will keep on windmilling due to the internal forces of turning the engine (try turning a big six cylinder engine on the ground...) It might be different for turbo prop engine's but I have no experience with these.

And when the engine is not working no supercharger will work or turbocharger.
And Boost Indication will show 0 Boost on ground if at sea level and ISA(When engine is stopped). So inflight you would find that it would indicate the actual air pressure(when engine is stopped) there for being less then 0 Boost. Normal Cruise power setting on my Airplane being around -0,5 to +1 Boost. So yes you would notice it.
And as the engine is not working Oil pressure would be 0psi, but Oil temp would only slowly decrease as well as Cylinder heat temperature. But they would initially stay the same.
And last but not least as John has said you would notice the failure as the props being in fine pitch would mean your really creating a lot of drag. Therefore creating yaw. Hopes this helps:OK:

The key is to follow the prop response to torque from the motor (be it turbine or piston).

Constant speed propeller systems regulate a positive torque output from the motor and convert that torque into a regulated speed. Any increase in torque results in an increase in thrust by virtue or regulating a specified rpm limit with the prop control.

In your scenerio an engine is now attempting to regulate a negative torque (not a reduction in torque). In the case of a negative torque the engines prop control responds by attempting to increase prop rpm. HOWEVER the attempt to increase rpm is in relation to a positive torque applicaiton, since you now have a negative torque deviation the actual result is a reduction in rpm since there is now a negative torque being applied and the blades are commanded to a flatter pitch rather than a higher pitch which would be required to increase rpms...

Since your scenerio has resulted in a negative torque deviation, the prop (by virute of its design to regulate positive torque) will attempt to increase rpm by commanding a reduction in blade angle in relation to the relative wind. This reduction in blade angle will therefore reduce engine rpm with a significant increase in drag. In a negative torque situation the blade angle will always be the inverse of what is required to actually increase the rotational speed again because the system is simply designed to respond to torque from the engine.

In cruise, if you were to reduce the torque or in this case the manifold pressure prop rpms will rise despite the response to a reduction in torque, however it is a REDUCTION in positive torque being applied. As the MP is reduced the blade pitch will gradually return to a flatter pitch to compensate for the reduction in available torque. Eventually and as somebody has already mentioned the blades mechanical limits will be reached and the props ability to compensate for rotational rpm will result in a gradual reduction in engine rpm. If airspeed is slow enough a windmilling prop will cease to rotate, however the extreme rise in drag as a result of the flat blade pitch will more than likely cause a loss of directional control.

Because of the requirement to regulate positive torque, multi engine prop systems have to be fitted with a feather position that permits an expanded range of movement relative to blade angle past a flat pitch detent. This allows the prop to move from the flat blade pitch detent to a pitch that follows a rotational direction that reflects a negative torque angle until the blade is positioned in a low drag position relative to the airplanes relative wind.

A feathering mechanism requires a spring that forces the blades into the negative torque position, again there exists no ability for the hydraulic forces to act negatively upon the props internal mechanisms. The springs will force the prop beyond the limits of the hydraulic system thus feathering the prop and greatly reducing the resulting drag.

It has obviously been a long time since anyone here flew pistons! :8

virgo isn't suggesting this at all, john - merely pointing out that a constant speed prop is just that. The control range of the pitch is relevant however.

The prop is feathered by the pilot. The question is about what happens before the prop is feathered (and the engine is thus still turning.)

Sorry,Micky - the air we fly through can provide a lot of force (enough to lift the entire aircraft into the air ;) ). All piston engines keep windmilling after a failure (discounting mechanical lock-up) until the pilot feathers the prop.
Superchargers are directly geared to the drive train. If the engine is windmilling, the supercharger is turning, and thus doing it's job to compress the intake air. Turbochargers are powered by the extra energy in the exhaust, if the fire isn't burning, a turbocharger will stop working - but that isn't the question here.
After the engine has stopped (after being feathered by the pilot) yes, this is the case. While the engine is windmilling, the oil pump is still being driven, so oil pressure is still normal.

Sorry, absolutely incorrect. You are confusing torque sensing (for auto-feather systems in turbines) with prop governors. Prop governors sense prop RPM, and adjust the pitch to compensate. If the RPM drops, the governor commands a finer pitch - if the engine is running, this reduces prop angle of attack, and thus prop drag and the RPM increases back to the governed level. If the engine is not running, the RPM drops, the governor commands finer pitch, this increases angle of attack (relative wind is striking the front face of the prop in a windmilling situation) and the relative wind thus drives the prop faster.

So, back to the original question:

Supercharged engine, with a non-mechanical engine failure. Prop governor maintains prop RPM. Supercharger & oil pump is mechanically linked to the crankshaft so they still operate normally.

Indicated RPM, oil pressure & manifold pressure all indicate the same after the failure as they did before, and this will continue until the propeller gets to the fine pitch stop, the point of which depends on the airspeed of the aircraft.

The fire has gone out, so Cylinder Head Temp. reduces rather quickly, oil temperature reduces less quickly and (as the prop is now being driven by the airspeed) the aircraft yaws into the failed engine and the airspeed reduces. It gets a bit less noisy as well.

[anecdote: check captain checking new pilot on a Heron arrives in the cruise. Check Captain reaches over and pulls all four mixtures out to the same point - as a demonstration of "efficient" engine management. Pilot under check thinks something is wrong - so pushes mixtures back in - and with the surge of power realises that the Check Captain had shut down two engines without realising it!

[True story, that.]

When any engine (turbine or piston) are running and by definition of "running" (generating positive torque, regardless of rpm) said engine transmits that torque to the motors output shaft or in this case the prop.

Depending on the amount of torque being generated (regardless of rpm) the engine will continue to accelerate (as a result of torque being produced) until something limits that maximum rotational speed, in this case a a prop governor increases blade pitch thus providing an increase in thrust output and the resistance to rotational torque hence limiting rotational rpm.

A prop governor is ONLY designed to respond to POSITIVE TORQUE from the motor (turbine or piston) a governor RESPONDS to torque not rpm (You can't have rpm without torque). RPM is simply a byproduct of said torque and rpms are only used to determine where a governor is operating at in respect to applied torque.

If you want to debate how a governor functions, then yes internal pressures that regulate the hydraulic forces sent to a prop hub are a result of the governors intenal pumps rpm and therfore the resulting pressures generated by that hydraulic pump.

In a NEGATIVE TORQUE SITUATION the prop governor will respond counter to POSITIVE TORQUE and the prop and windmilling engine will reduce it's rotational rpm relative to the speed of the flow of air accross the props blades.

This is a critical aspect to the design of a prop because if the prop were designed to increase the blades angle in a negative torque situation, the engine and prop would overspeed and result in a catastrpphic runaway situation.

There are inertial devices that prevent a prop from going into a negative blade pitch in flight (Feather). These arm when an over-ride is activated in the prop control system thus allowing the prop to continue past a flat pitch angle to a feathered blade position and are moved their by way of a piston and spring that act against the blades internal positioning arms.

Prop governors work as a result of torque output. Torque output determines the rpm and the limits of said rpm are a result of a torque limit set by the pilot thru the governor.

RPM's are a result of torque, torque is not the result of rpms...its of them laws of physics kinda things ya know.

If the engine is not running (producing positive torque), the RPM drops due to the governor commanding finer pitch as a result of the negative torque condition, this decreases the angle of attack and the prop rpms are reduced.

Google article:
Google Image Result for http://www.hariguchi.org/flying/info/figs/csp3.jpg

Perhaps you should read your own link!
(emphasis mine)

Not correct, I'm afraid. The governor is there to govern the RPM - and that is what it does. reduce power on a fixed pitch prop, and the RPM drops. Reduce power on a constant-speed prop, and the RPM remains the same (the hint is in the name!).

Reduce power on a constant-speed prop, and the RPM remains the same (the hint is in the name!).

I am still concerned if folks don't include the "within the control range" caveat.

A BMEP gauge is the answer!

G'day ;)

It is not surprising that you cannot convince a non-aircrew colleague, because, your information is incorrect....in so far as the following engines are concerned:
Pratt&Whitney R2000, R2800CB16, R4360...CurtisWright R3350 turbocompound series.

Yes,the RPM remains the same (in your selected scenario), the oil pressure (usually) remains OK, however, the manifold pressure and BMEP are most definitely reduced when the fuel flow ceases due to fuel exhaustion.
And yes,I've flown all of the above types on a variety of aircraft, IE: DC4, DC6, DC7, 1649 Constellation, Stratocruiser (B377).

Thanks everyone for your inputs - even those who have answered questions that weren't asked !!!!!
(Checkboard obviously understands)
411A........Are those engines fitted with superchargers or turbo chargers ? (I did specify a "Supercharged" engine). Obviously if there's no exhaust the manifold pressure will drop to atmospheric pressure.
Not many British engines had BMEP indication.

RTFQ....

But that is not what he was asking....

And to refer to my link...


His question is "if you lost an engine during cruise due to fuel starvation, it would continue to windmill at the selected rpm"

The answer is:

No, the engine is producing a negative torque (windmilling) by virtue of the air passing accross the blades. A Governor is only designed to respond to POSITIVE TORQUE to achieve a selected rpm, since the torque is now a negative the result is a reduced blade angle to the flattest blade pitch position (Mechanical stop).

The question is not how a governor works, but how a prop responds in respect to the available torque with an engine failure.

And again, the answer is that the rpms will decrease as a result of negative torque and the governors response to that negative torque is reducing blade pitch resulting in lower rotational rpm from the flatter pitch and reduced blade angle. And again this is why the prop will need to be feathered since the drap rise with a flatter pitch will exceed the residual thrust available from the other engine and you WILL lose altitude ( loss of airspeed then lift, remember that equilibrium thing from your private pilot days?)

All have superchargers, some two speed.
One engine I specified has both a supercharger and a General Electric turbosupercharger (R4360) and another (the CurtisWright R3350 turbocompound engine) has both a two speed supercharger and three power recovery turbines (PRT's).
IF you want to learn more about these large complicated engines, visit AEHS Home and become a member.

The governor doesn't care where the torque comes from, it cares only about the RPM. In the situation of a windmilling prop, the net torque remains "positive" in the sense that you mean it, with a negative torque from the engine offset by a positive torque from the air forces on the prop.

The mechanism of the governor is the same: the blade pitch adjusts until the net torque manages to drive the prop at the set-point RPM. If the engine starts ingesting pure air, the torque from it will reduce, and (in principle) the pitch will become finer and finer until it is being driven at the set-point RPM by the air forces, or it hits the fine stop, in which case the prop has to slow down.

I find it hard to believe that the other engine on a twin will continue to produce enough power that sufficient airspeed will be maintained so as to windmill the prop on the dead one at a cruise RPM. But if you had 6 or 8 engines, I can quite believe that one could fail and the aircraft would settle at a lower speed with one prop, in effect, in reverse.

What I don't get is the significance of the supercharger in the original Q. Doesn't the same effect occur on a normally aspirated engine with a wide open throttle?

Largest supercharged piston I've flown is the Hercules 14 cyl. sleeve valve radial.
IIRC, to re-start in the air: RPM lever out of feather and press feathering button had it windmilling merrily away. Lot of push in that Q formula.
(There were some other things to actually get it to run - fuel and ignition would probably help :) )

Thanks bookworm. You're absolutely right about the operation of the propeller control unit (CSU)The significance of the supercharger is that if the engine rpm is maintained and the supercharger is engine driven (as compared to a being turbo-charged), the boost (manifold pressure) and charge temperature will be maintained.
So you've got a dead engine - causing drag - that maintains the same RPM, Boost, Charge temperature and oil pressure as the functioning engines.

If you were totally distracted at the time of the failure - setting up an attack on a submarine or dealing with another emergency (an electrical fire in the fuselage), the failure could be missed until the aircraft has slowed down.......which I agree on a twin would be pretty quickly but with another three engines it takes a bit longer.

411A..............my information is NOT incorrect, I've had it demonstrated to me, seen it and done it. Have you any experience of multi-engined British piston engined aircraft ?

Sorry, it is incorrect, in so far as the American piston engines that I described.
I've flown 'em all...not just seen 'demonstrated'.:rolleyes:

Negative...the Brits were very good with turbopropellor designs.
Pistons?
Second fiddle, by far, compared to the American designs.

Pistons?
Second fiddle, by far, compared to the American designs.

Hmmm... I guess that's why the P51 was eventually fitted with licence-built Merlins, then... :}

Ok, let's use the KISS method

1. The RPM will drop noticeably
2. every other engine parameter will drop with the RPM

Why? no fuel=no combustion=no turn

The propeller will now be driving the engine which will be acting like a brake due to compression in the cylinders.

The RPM will vary depending on airspeed and blade angle and engine type.

What won't vary is that you'll be trying to feather the prop/restart engine and hold the a/c level(twin), might not be as noticeable on a fourbanger but you'll know pretty darn quick.

Hope this helps.

FM

Funny, when I was doing multi-engine training, in order, it was:

Identify
Verify
Feather

So many of these posts are something along the lines of, "You shouldn't see these confusing indications because you should feather the engine!"

See the list above. :ugh:

and that is:

Identify - with the leg.
confirm - with the throttle
feather - with the prop lever.:rolleyes:


the instruments CAN be confusing under pressure!:=

"dead" leg - "dead" side. . . Check.

Now which throttle do you pull on a 4-engine prop? Or do you guess on the first pull?

No guesswork involved with the large American radial piston engines...you look at the torquemeter (BMEP gauge) and it will tell you straight away, if the engine is developing BHP...or not.
Not on the abovementioned types.
A propellor feathering pump is used to drive the prop blades to the feathered position.
Just....push the feather button.

Exactly my point. http://images.ibsrv.net/ibsrv/res/sr...ies/thumbs.gif (i.e. It's not as simple as dead foot = dead engine.)

So much theory. Scenario originally given was running a tank dry, not initially changing throttle. With a DC3 THIS is what happens. At 150 knots TAS, there is no change in RPM or MAP. At 90 knots, there is some decrease in RPM and because the supercharger RPM drops with it, some drop in MAP. All of course accompanied by (expletive deleted) as throttle is hastily reduced and fuel tanks changed over.
John Tullamarine is usually right on these matters. It's all about the point at which the prop hits the fine pitch stops when there is no power coming out of the engine. At low airspeeds it must go full fine then it behaves as a fixed pitch prop if speed is further reduced. Also less RPM = less MAP. At higher airspeeds, because the engine is turning and producing oil pressure, the governor/prop combo MAY be capable of constant speed control, depending on TAS.
As an aside, I once had a prop run away to just over 2700 RPM on approach but we got slowed down and found that at 90 knots the RPM was quite controllable with throttle, so we kept it running rather than feather it.
Can't speak for other supercharged types.
Heron with the old Gypsy Queen engines, which were not supercharged (as someone recently tried to tell me - he was confused with the DH Dove): Similar reaction if you switched off the ignition in the cruise. Nothing to see except a gentle yaw. When airspeed reduced the RPM decayed, but now (because it was not supercharged), there was no visible change on the manifold gauge. From memory manifold indication was meaningless on those engines anyway as they had an ingenious single lever control of throttle and RPM - for once the Poms built something that was simple for pilots, if not the engineers. Basically if you could hear noise, the engines were working, and if one failed you knew because it got noisier as they went out of synch.

simple answer
 

simple thing is the engine will windmill a little until feathered

any reciprocating internal combustion engine is only around 30% efficient, therefore the other 70% goes in heat...... and the requirement to turn the engine and keep it going.

so if fuel runs out..... no bang no power, just drag......
and a supercharger takes 50% of the power it creates to run itself... its the law of conservation of energy, you don't get something for nothing!!!! - a turbocharger uses wasted energy (heat out of exhaust) to drive a compressor, so it is grabbing some of that 70% waste and using it to increase volumetric efficiency, that is the ability to induct the same air volume as the capacity of the engine

If you want to increase volumetric efficiency over 1, you need forced induction...

anyhow.... a windmilling piston engine that is supercharged may give some MAP but not much, as turbocharged engine wont give any MAP as there is no heat coming to run it..

Cheers

Dash

Virgo and Checkboard have got it right, as applied to a supercharged (not turbocharged) engine, and provided the fine pitch limit is not reached. Virgo's reference to submarines suggests that he is an ex Shackleton man, as I am.
If the fuel supply to an engine was cut, there would be a momentary dip in the RPM as the CSU compensated for the power loss, but it would return to its original RPM. The MAP would also dip and return, since the supercharger speed was related to engine RPM. The oil pump being engine driven, the oil pressure would be maintained.
If you missed seeing this brief dip, you were then faced with yaw and reducing airspeed. It would be some time before coolant temperature would show, as the radiator flaps would normally be in auto, and would close to maintain the temperature.
Having decided which side the failed engine was (i.e. dead leg), and after increasing power all round, the procedure was to close the throttle of the
inboard engine on that side, while watching the RPM gauge.
If the RPM dipped, that was the good engine, going from thrusting to windmilling. Restore the power on it promptly, as you are now double assymetric.
If the RPM did not dip, it was because that engine was already windmilling, i.e. it was the failed engine, so go ahead and feather it.

In the cruise or higher speed range that I describe with the DC3 and Heron ( and I expect most other types) of course, when the engine quits due to fuel starvation or because some mug has hit the magneto master, there will be a MOMENTARY fluctuation downwards in RPM while the CSU makes a correction, but then the indications will be as I describe. Also, although the theory may say otherwise, the throttle needs to be retarded before re-introducing fuel/ignition. Because the pitch has fined off during the power loss, a gob-full of power would cause a momentary overspeed until the CSU gets ahold of it and coarsens the pitch to restore RPM to where it was at the time of the power loss. No theory - simple fact.

Thank you SO much Oxenos and Mach E Avelli.

I knew that eventually someone who knew what they were talking about would confirm my statement.

Just to add $0.02 to the prop governor discussion:

The governor is designed to this paradigm: To increase RPM, reduce blade pitch angle (i.e. reduce AOA); to decrease RPM, increase pitch (incr. AOA).

As long as the engine is producing positive torque, this paradigm applies nicely.

But if the engine ISN'T producing torque but is instead absorbing torque from the windmilling prop, the above paradigm creates bad results. As the RPM initially drops, the governor moves blades to a lower pitch angle, which now means a NEGATIVE AOA.

The prop now speeds up, and probably passes into a mild overspeed. The governor senses this, and reduces pitch angle some more. This of course means a MORE NEGATIVE AOA, windmilling the prop to an increased overspeed. This cycle repeats until the blade pitch reaches some mechanical stop. It's supposed to be a pitch-lock mechanism, which hopefully contains the problem before something breaks.

Absolutely incorrect, barit1.

Repeating my post above:

Finer pitch when the engine is producing power increases RPM.

Finer pitch when the prop is windmilling also increases RPM.

In the first case the engine is driving the prop, the airflow is striking the prop on the "back" of the blade (the side the pilot sees) so a finer pitch reduces the blade angle of attack, decreasing the drag and so allowing the engine to spin faster.

in the second case the airflow is striking the prop on the "front" of the blade, so a finer pitch increases the angle of attack from the relative wind, which increases the force that wind is applying to spin the prop, which increases the RPM.

The prop governor thus controls the pitch in the correct sense in both cases.

More modern systems do have all sorts of exotic pitch locks, auto-coarsening, beta backups etc. Not applicable to the original question.

A bit off the topic but to reply to an earlier statement.

stevefCorrect to a point.

It was the supercharger technology that made the Merlin superior to the Allison orginally fitted to the P51. It took the Yanks to refine the Merlin for mass production.

As much as Rolls Royce has everyone believe that they produced the finest engines, they were effectively hand made. Each piston individually fitted to each bore and each crankshaft individually fitted to each journal. You couldn't take a piston out of one cylinder and fit it to another. The Roll Royce produced ones are/were a nightmare to own.

The "Parkard" Merlins were a much better engine. True, a British design at the end of the day.

Back to the original question.

I would expect a supercharged engine to act in a similar manner to a normally aspirated one in the situation described in the question.

That is RPM and Manifold Pressure would not change or if so only momentarily, oil pressure would stay the same also. Temps would drop slowly and of course there would be some yaw and speed would decay.

Lack of fuel flow, assuming a flow gauge is fitted, would be the major indicator.

What is the basis for this assertion?

Wikipedia (fwiw) gives the following:

" By the end of its production run in 1950, almost 150,000 Merlin engines had been built; over 112,000 in Britain and more than 37,000 under license in the U.S"

Wikipedia breaks down the numbers:

" Factory production numbers:

• Rolls-Royce: Derby = 32,377
• Rolls-Royce: Crewe = 26,065
• Rolls-Royce: Glasgow =23,675
• Ford Manchester= 30,428
• Packard Motor Corp = 55,523 (37,143 Merlins, 18,380 V-1650s)
• Overall: 168,068"

Ref Gunson 1995

My Apologies. I was clearly not thinking clearly! Checkboard's right.

Yamagata Ken, check the wikipedia (fwiw) for the Packard V-1650 entry.

I'm no historian in this regard but I'm guessing it went something like this:

Great design+improved production technique=155k engines.

UK/US synergy at its best.

One dot right, you are on centreline.

In long range cruise on R2800 there were 2 leaning techniques, "2 bmep drop" and "11 bmep drop". Couple of times a dozey F/E would go past 11 bmep on the lean side, leading to a sudden drop in bmep to almost zero. Prop kept turning same RPM, MAP stayed same unless throttle moved. Capt blood pressure rose. Engineer bought many beers.

I suspect if airspeed low enough, prop would not have enough "fine pitch" available, and RPM would start to decay with decaying airspeed. But that area of flight is academic and of interest only to the very brave.


Use of feathering pump became very apparent when I had a runaway prop on DC-6B. Prop governor(which normally controls RPM ) mechanically broke (governor control spring failure) resulting in loud howl from prop and Capt simultaneously.
Luckily, feathering pump pressurises governor control piston to port oil to "increase pitch" side of prop servo piston. It seemed like an age, but prop eventually slowed a bit, and then feathered normally. Ten more grey hairs.