Pressure through a curved hose - expected results?

[escape75]

TL;DR

Pressure through a bent exhaust produces more force than blowing it straight out.

Hello,

I am doing some quick experiments testing how pressurized air behaves when being ejected from a straight vs. curved object,
and I am wondering if this is the expected behavior and a possibly quick explanation of why this would be behaving this way...

I am using the model below and sending 100 PSI pressure via either the back connector (straight out) or top (bent pipe),
and I am getting about twice as much force applied by this object when using the bent pipe as opposed to straight out.

The object is applying pressure to a force meter which is reading the force on the opposite side of the exhaust.



Any insights would be appreciated!
Thanks!

 

[Lnewqban]

That reaction should be caused by the chage of direction of the fluid, rather than by the internal pressure on the flexible part of the curved pipe.

Please, see:
https://www.engineeringtoolbox.com/forces-pipe-bends-d_968.html

 

[escape75]

What's interesting was that when the bent pipe wasn't getting bigger towards the exit (like a funnel),
but instead was the same diameter throughout the force was actually smaller than the straight out exit.

Maybe there was air turbulence and not as much fluid was changing direction ...

The current one looks like this from another angle:

 

[russ_watters]

The object is applying pressure to a force meter which is reading the force on the opposite side of the exhaust.

Can you provide a drawing of that?

Are you sure the flow rates are equal? The curve can also lead to non-uniform velocity.

 

[escape75]

Yes, it looks similar to this, with the sensor (cyan), rails (blue) that allow the object to move up/down on the axis (red) and create force in the direction of the red arrow ...

 

[Arjan82]

That's not the expected behavior. The bend should only cause an extra vertical force. I presume you also attach a pressure hose of some kind? In the straight nozzle that is in the same direction as you are measuring the force, are you sure that is not in the way?

 

[escape75]

I am pretty sure it's not in the way, as the sensor basically touches the object, so there's virtually no movement required,- except just enough to read the force.

My pressure tank only sustains 100 PSI for a short time, so I'm essentially reading that burst force,
but I as well thought that the straight output should give me a higher force reading ...

I will re-check my setup and see if I can find any issues.

 

[256bits]

I as well thought that the straight output should give me a higher force reading ...

You must have had some reason to have come to that conclusion.

 

[escape75]

You must have had some reason to have come to that conclusion.

The fact that the straight output would be more efficient thus creating more force.
But maybe I'm thinking about this wrong :)

I was looking at this from the point of view that the ejected mass creates the force ...

 

[256bits]

What's interesting was that when the bent pipe wasn't getting bigger towards the exit (like a funnel),
but instead was the same diameter throughout the force was actually smaller than the straight out exit.

The force for the constant diameter bent pipe would be attributable to the change in direction of the fluid, as @Lnewqban mentioned. The fluid suffers a change in momentum in both the vertical and horizontal directions.

The funnel section is a divergent nozzle, used in all places as rocket engines to provide thrust.

 

[escape75]

So I think what you're implying is the force difference between the straight and bent funnels would mostly be related to length of the funnels (the in and out diameters are the same),- I'm still not sure why the bent funnel seems to be creating more force ...

 

[russ_watters]

So I think what you're implying is the force difference between the straight and bent funnels would mostly be related to length of the funnels (the in and out diameters are the same),- I'm still not sure why the bent funnel seems to be creating more force ...

Not exactly length, but shape. A converging-diverging nozzle requires a specific shape to produce maximum velocity in supersonic flow. It's possible that you're getting a higher velocity out of the curved one.

 

[escape75]

Sounds like I will need to either perform some calculations, or simply make different models that have different shapes but are all straight and see if I can get a better output out of a particular one ...

 

[escape75]

I'm still trying to figure out how the original object produces thrust, I'm assuming it's not the same way as a rocket engine which expels mass at speed outwards, but by fluid (air) producing pressure against the surface ?

Picture the following model, and by having pressurized air come out of the orange nozzle down and hitting the surface, and producing a force against the blue sensor, which of these would produce more thrust, seeing how one object is joined together with the nozzle,- both objects are unable to move as they're pressing against the sensor ... we will assume both are mounted to a rail and they can only move towards the sensor.

 

[russ_watters]

I don't see why they wouldn't produce the same thrust. But what are you thinking?

 

[escape75]

I just wasn't sure if the connection would change the behavior of the system ...