Trying to stall an axial fan (cooling fan)Would it work?

[koolraj09]

Hi all,
I was thinking of doing an experiment with axial fan for determining it's stall condition. By stall, here, I mean airfoil stall. I want to do this experiment in order to gain a firm grasp on the concept of stall. I have read the theory but an experiment would definitely give me an intuitive feel about the concept.
So I wanted to make a fan stall and observe it's behavior.
The fan i am thinking of using is a CPU cooling fan and mounting it in a converging passage. I was thinking of varying its speed in order to make it stall.
Is it possible to do this (would it work)? Also what else can I do to make it work? All I want is to observe the behavior of the fan when it is stalled.
Also how would I come to know whether the fan has stalled or not? Is it the vibrations/sound of the fan?
Do I need some added instrumentation for determining this?

Any direction for making this work is greatly appreciated.

 

[NTW]

The stall happens when the airfoil reaches a given, critical angle. The term 'stall speed' is confusing, but applicable in airplanes, since in order to keep the plane flying, you have to get enough lift. If you reduce the airspeed, you have to keep the lift by increasing the angle of attack, but that works up to a point... The point is the stall. In a plane, it happens at a given airspeed, but the real culprit is the angle of attack...

In order to have the blades of the fan stalled, the angle of attack of those blades must reach the critical value, that varies with the airfoil, and I believe that with the Reynolds number too. You cannot, obviously, change the pitch of those blades, and it will never stall 'as built', however fast it may rotate, but you can change the angle of attack of the blades, and perhaps reach the critical stall value by placing the spinning fan within a mass of moving air moving against the stream produced by the fan.

If you make a drawing, with the vectors involved, you'll understand how the angle of attack varies with the relative velocity of the airstream, at a given fan rotational speed, compounding with it and resulting in the 'relative wind', as it is known in aviation...

 

[rcgldr]

I'm not sure if the pitch and diameter of the blades in a cpu fan are enough to create a stalled state even at very high rpm, short of reaching super-sonic speeds at the rotor tips. For high speed radio control models, the most likely models to stall the prop are the F5B contest modesl which only use motors in 2 to 5 second bursts at a time to achieve 120+mph in a vertical climb, then glide for most of the contest tasks. These have a relatively large amount of pitch (like 16 inches forward per revolution, on a 16 or 17 inch diameter prop). When stalled, the amount of thrust is greatly reduced, but a hard throw can get them started (usually once these models are setup, the throttle only has two states, off or max power).

 

[NTW]

I'm not sure if the pitch and diameter of the blades in a cpu fan are enough to create a stalled state even at very high rpm, short of reaching super-sonic speeds at the rotor tips. For high speed radio control models, the most likely models to stall the prop are the F5B contest modesl which only use motors in 2 to 5 second bursts at a time to achieve 120+mph in a vertical climb, then glide for most of the contest tasks. These have a relatively large amount of pitch (like 16 inches forward per revolution, on a 16 or 17 inch diameter prop). When stalled, the amount of thrust is greatly reduced, but a hard throw can get them started (usually once these models are setup, the throttle only has two states, off or max power).

Stalling does not depend on airspeed, but only upon angle of attack. When it reaches a critical value, the airfoil stalls. Airspeed has nothing to do with that...

 

[rcgldr]

Stalling does not depend on airspeed, but only upon angle of attack. When it reaches a critical value, the airfoil stalls. Airspeed has nothing to do with that...

I mentioned fan speed, not air speed. If the rpm is high enough, the induced flow being pulled into the fan can't keep up with the speed of the fan blades, and the effective angle of attack is increased.

For those F5B models, this can happen with a high pitch prop when in a static situation. The motors are producing about 3 1/2 horsepower or 2.6 kilowatts with a 16 or 17 inch prop with a 16 or 17 inch pitch. The model needs to be moving forwards at some non-zero speed in order for the propeller to produce maximum thrust, which quite often can be achieved with a good throw of the model (about 20 mph or 32 kph), or the model will quickly accelerate anyway since even the "stalled" thrust is still greater than the weight of the model.

 

[NTW]

I mentioned fan speed, not air speed. If the rpm is high enough, the induced flow being pulled into the fan can't keep up with the speed of the fan blades, and the effective angle of attack is increased.

For those F5B models, this can happen with a high pitch prop when in a static situation. The motors are producing about 3 1/2 horsepower or 2.6 kilowatts with a 16 or 17 inch prop with a 16 or 17 inch pitch. The model needs to be moving forwards at some non-zero speed in order for the propeller to produce maximum thrust, which quite often can be achieved with a good throw of the model (about 20 mph or 32 kph), or the model will quickly accelerate anyway since even the "stalled" thrust is still greater than the weight of the model.

A fan has, usually, a fixed pitch setting, and the angle of attack might be significantly increased only by a 'reverse inflow', a flow against the current generated by the fan. If the 'induced flow can't keep up', that would just result, I believe, in a lowered air pressure before the fan. I understand what you say; there may be a decrease in the inflow velocity, but I very much doubt that it may be important enough to vary the angle of attack by much, and stall won't take place... Perhaps shutting off completely the inflow could produce a kind of 'circulation' with partial stalling at some sections of the blades, but that's a complex phenomenon, not easy to observe and measure...

I know well all you say about models. I've even experimented it myself with a light airplane that I flew, when experimenting with (ground) pitch-adjustable props... For the same reason that you state that 'you needed a good throw', I have had to suffer extremely long takeoff runs when testing too coarse-pitched props... But thrust improved little by little, as my plane gained speed in the run. That was, of course, because the angle of attack, as 'seen' by the prop blades, was decreasing, leaving behind the stall zone...

In conclusion: IMHO, the only way to stall a fan is to put it in a wind tunnel, working against the flow. That way, a stall could be easily observed, probably...

 

[rcgldr]

For those F5B models, this can happen with a high pitch prop when in a static situation ... which quite often can be achieved with a good throw of the model.

I should have noted that this is by design. These models don't have any gear (wheels) so they can't take off from the ground, and have to be thrown. The maximum prop pitch is limited, since these models have to be able to be launched by throwing them, and many of them make sufficient static thrust that only requires that the models be guided mostly horizontal and slightly upwards to ensure they don't plunge into the ground upon launch.

A fan has, usually, a fixed pitch setting, and the angle of attack might be significantly increased only by a 'reverse inflow', a flow against the current generated by the fan. ... coarsed pitch props on model aircraft

As noted it is possible to induce stall on model aircraft props with a high fixed pitch factor, but I'm not sure if there are any computer / server fans with sufficient pitch for this to happen, even at very high rpms.

 

[NTW]

You quote me as saying: '...coarsed pitch props on model aircraft'. Well, I never wrote that. However, if what you mean is that those props will also be stalled by a external flow going against the propeller stream, well, yes: I believe that too...

Now, concerning the key point of the discussion, I hope that we all agree in that stall is an angle-of-attack dependent phenomenon, independent from fan- (or prop-) speed, and that can be induced by a reversal of the flow, as sometimes happens in jet compressors... Something koolraj09 might do to observe stall in that fan would be to find a wind tunnel somewhere (or to make a small one by himself).