Dynamic pressure changes

[ENGINEERRMECH]

Hello Forum-
I have a question concerning pressure changes. We are trying to make small pressure changes to a static pressure of 5psi by varying the volume in a cylinder. We have had no problem making these pressure changes, but an interesting phenomenon is occurring. At the leading and trailing edges of the pressure pulses there are very fast pulses that are approximately 2-3x the amplitude of the pressure pulse. I do not know what is causing these extra pulses, or even what to call them, but they are dependent on how fast the velocity changes. Does anyone know what I would call these extra pulses, or does anyone know how to attenuate, or eliminate these extra pulses?

Thanks,
ENGINEERRMECH

 

[quark]

May be your pressure transmitter is too sensitive and it may be close to entry and exit points. Are you continuously discharging the air from cylinder during the experimental set up?

 

[hydromech]

It looks a lot like under shoot and overshoot, how are you varying the volume of the cylinder?

 

[BigInch]

better yet a diagram of your setup
and what fluid

**********************
"Pumping systems account for nearly 20% of the world’s energy used by electric motors and 25% to 50% of the total electrical energy usage in certain industrial facilities." - DOE statistic (Note: Make that 99.99% for pipeline companies)

http://virtualpipeline.spaces.live.com/

 

[ENGINEERRMECH]

Hello BigInch-
I've attached a pdf to show my prototype setup and the fluid is air.

Thank you,
Engineerrmech

 

[ENGINEERRMECH]

Hello quark-
May be your pressure transmitter is too sensitive and it may be close to entry and exit points.
Engineerrmech: We do have a sensitive pressure sensor due to it's other uses in the future product. What do you mean "too close to entry and exit points"?
Are you continuously discharging the air from cylinder during the experimental set up?
Engineerrmech: The system is closed so we do not discharge air during the dynamic pulses.

Thanks,
Engineerrmech

 

[ENGINEERRMECH]

Hello hydromech-
It looks a lot like under shoot and overshoot, how are you varying the volume of the cylinder?
Engineerrmech: The volume is changed by moving a simple pneumatic cylinder by means of a stepper motor.

Thanks,
Engineerrmech

 

[BigInch]

Diaphram snap action at start and finish of travel, check valve opening and closing?

**********************
"Pumping systems account for nearly 20% of the world’s energy used by electric motors and 25% to 50% of the total electrical energy usage in certain industrial facilities." - DOE statistic (Note: Make that 99.99% for pipeline companies)

http://virtualpipeline.spaces.live.com/

 

[btrueblood]

There is a restriction in the line from the cylinder to the test volume, such that it takes a finite time (about 1 second in your case) for the pressure/partial vacuum in the cylinder to leak down/up until the two volumes are in equilibrium. The p'ducer is a much smaller volume, so it responds more quickly to the cylinder pressure.

Calculate the leak down time consant

 

[gerhardl]

1. I believe what is occuring is comparable to what is happening in water by sudden flow changes, examplified by suddenly closing a valve: pressure peaks / shock waves when the kinetic energy is transferred to pressure energy.

2. By water you try to avoid this by slower flow, using wider pipelines and slower movements and longer closing (opening) times, sometimes also water expansion tanks, pressurized with constant overpresuure air to expand /damp the shockwaves.

3. How to cope with your particular problem would be to try to balance your total system, depending on how exact and how fast you have to have your application to react.

Some suggestions:

a. Slow down all movements to maximum allowed time.
b. Check if any components (vessels, pipelines, all) or restrictions should be larger or smaller to constrict or even out the flow.
c. Look at check valves. Select spring loaded or soft closing types and 'correct' sizes (perhaps smaller?).-
d. Closing time and sizes for solenoid valves, throtteling of any ports in and out?
e. Set in needle valves some places to experiment with slowing /constricting the flow?

 

[BigInch]

gerhardl, I think so too, but I think its coming from the action of the diaphram pump, so it'll be hard to control unless he can slow down the pump.

Egrmech, you might have some success with adding a volume accumulator on the pump discharge to smooth things out.

**********************
"Pumping systems account for nearly 20% of the world’s energy used by electric motors and 25% to 50% of the total electrical energy usage in certain industrial facilities." - DOE statistic (Note: Make that 99.99% for pipeline companies)

http://virtualpipeline.spaces.live.com/

 

[hydtools]

Does it matter what the pressure does where the sensor is placed? Would not the pressure near the 'pressure engine' be of more interest? Then you would see the effect of the 1/16 tubing on the system pressure of interest and probably not see the spikes.

Ted

 

[BigInch]

The spikes should be present anywhere you put the pressure gage, unless its so far downstream from the source that they have essentially dissipated. Is the object to not see them or to eliminate them?

**********************
"Pumping systems account for nearly 20% of the world’s energy used by electric motors and 25% to 50% of the total electrical energy usage in certain industrial facilities." - DOE statistic (Note: Make that 99.99% for pipeline companies)

http://virtualpipeline.spaces.live.com/

 

[btrueblood]

Biginch, this is air, not water! The "spikes" are caused by the 1/8-to-1/16th inch reducers between the cylinder and the 400 mL volume. He should either open up the tubing to a consistent and larger diameter, or tap the 400 mL bottle for a pressure reading. If tapped in the bottle, the spikes would magically disappear, and instead you'd see a classic exponential decay/rise "square" (some EE's call it a sharkfin) wave.

 

[ENGINEERRMECH]

Hello Btrueblood-
Thank you for the tip, when I moved the sensor to the 400mL volume the extra pressure pulses did magically disappear. How would you explain this? Is this because the large volume isn't affected as the easily as 1/16th tubing's volume? I have a couple other questions, when is air considered compressible vs noncompressible? Depending on how fast I move the pneumatic cylinder the extra pulse would grow substantially, why is this?

Thanks,
Engineerrmech

 

[BigInch]

When the increase in air mass arrives at the 400 ml container it is divided by a much larger volume resulting in a much lower pressure.

Air is considered compressible, whenever the compressibility factor does not affect your calculations,

http://en.wikipedia.org/wiki/Compressibility_chart

**********************
"Pumping systems account for nearly 20% of the world’s energy used by electric motors and 25% to 50% of the total electrical energy usage in certain industrial facilities." - DOE statistic (Note: Make that 99.99% for pipeline companies)

http://virtualpipeline.spaces.live.com/

 

[btrueblood]

Emech, you are welcome.

"How would you explain this? Is this because the large volume isn't affected as the easily as 1/16th tubing's volume?"

Yes. It takes a finite amount of time to move air into or out of the 400 ml volume, and a much smaller finite amount of time to fill/empty the tubing and components connected to it. FWIW, the same thing happens with any fluid...if you have instruments that can respond fast enough to see the pressure transients.

" I have a couple other questions, when is air considered compressible vs noncompressible? "

Mmm. Air is compressible. All (real) fluids are compressible. Incompressibility is a fictitious condition used to simplify analysis. You can sometimes ignore the compressibility, sometimes not. Aeronautics and fluid flow generally says to ignore compressibility when you are below Mach 0.3, or pressure ratios that will drive those speeds.

But, acoustics treats air as "linearly compressible", if you will, at zero speed (pressure waves travelling thru still air). That's not quite correct, but you get my meaning.

In your case, you are deliberately compressing the air to create a pressure change, so you should model the air as being compressible (and that multiply recursive statement oughta make the English teachers shudder). If you tried your experiment with water, you know what would happen - your stepper motor would stall, because the water is much less compressible than air, and the motor wouldn't have enough torque to move the piston.

"Depending on how fast I move the pneumatic cylinder the extra pulse would grow substantially, why is this?"

You are compressing the cylinder side volume, causing its pressure to rise. The pressure decays thru an outlet (to the 400ml bottle) that can pass only a fixed rate of air per sqrt(diff pressure). Smash down the small volume faster, and the pressure goes up.

A little spreadsheet model of the volumes of the cylinder and bottle, tracking the mass of air in each, and connected by the flow thru the tube (just model it as an orifice) will teach you a great deal. And, once you do this, you will have taken the first steps on the path to Computational Fluid Dynamics (CFD), using Finite Difference Methods (well, really finite volume methods, but hey).

Next thing you know, you'll be coloring in pretty pictures and drawing streamlines...and that way lies madness.

 

[btrueblood]

Oh, and apologies to all my my earlier hasty replies, but both times I was late for a cribbage match in the shop and had to rush off before writing a complete answer. No biggie, Engrmech caught what I was saying, and the problem (if it was one) is fixed.