hydraulic power recovery turbines

[Vic123]

How do we conduct performance tests on hydraulic power recovery turbines? Is there a standard or a document for conduction the same. My main concern is how to measure the KW Output. Thanks

 

[JJPellin]

I have never seen a document that describes a methodology for performance testing a hydraulic recovery turbine. If I was asked to do so for any of our turbines, I would base it on simple measurements of the amp draw of the motor driver. But this possibility only exists if you have systems that are comparable to ours. All of our recovery turbines are on trains with a primary driver that is an induction motor. All of our recovery turbines are separated from the motor driver with an overriding clutch or clutch coupling. And, all of our recovery turbines have a bypass control valve than can be used to take the pressure drop and control the flow in the event that the recovery turbine is taken off line. With this configuration, I would do the test as follows:

It would be preferable to do the test with the unit at maximum charge rate since the efficiency of a recovery turbine drops off drastically it is pinched back to reduced rate. At max charge rate, I would record the amp draw, voltage and speed of the motor driver. I would take the recovery turbine off-line, allowing the clutch to disengage and the control valve to assume control of the flow and pressure drop. I would try to achieve the same total charge rate. Once the process conditions were stabilized, I would again record the amp draw, voltage and speed of the motor. I would perform the appropriate power calculations to determine the motor power under both conditions. The difference between the two power values would represent the power contribution of the hydraulic recovery turbine.


Johnny Pellin

 

[MartinSr00]

Do it just like a pump. Calibrated flowmeter on the inlet. Pressure guages (bled and corrected for height to a common datum -- see the Hydraulic Institute Standards) at the inlet and outlet. The shaft would need to attach to a generator (rather than a motor), dynamometer, or similar device. For horsepower, either calibrate it to obtain the motor efficiency (so you can back it out from the total efficiency)and use a wattmeter, or use a torquemeter (strain gauge instremented coupling) and a keyphasor (for rpm) to discern power.

Vary the flow from zero to some high value, and calculate the head and horsepower at each flow point.

 

[Vic123]

Martin
I think your approach is straightforward. A question regarding the horsepower measurement. I am not sure what you mean by " to obtain the motor efficiency (so you can back it out from the total efficiency)and use a wattmeter"...which motor is this?
I am planning on using a dyno for horsepower measurement which should directly give torque and rpm as the output. Is this correct?

Thanks

 

[MartinSr00]

Sorry for the confusion -- calibrate the generator to determine it's efficiency. If the total water-to-wires efficiency of the turbine/generator is 81%, and the generator 90%, the turbine efficiench could be reckoned at 90% (=.81/.90). This is always a bit of a fudge, because the generator may not have constant efficiency at various loads.

Also, you could have a problem on the electrical side. If the generator is large, you may not be able to hook it up to the utility grid in order to load it. You may have to find some other means to sink the electrical power.

A dyno circumvents the whole problem. If the dyno is calibrated, then you have a solid horsepower measurement that you can take to the bank.

I incorrectly used the word "motor" instead of "generator" because most of my pump test experience centers around pumps, not turbines.

Also, you may want to monitor NPEH (net positive exit head) to make sure that your runner isn't full of water vapor (cavitation).

 

[bingopin]

Don't forget about the potential for overspeeding the turbine(pump). Seen this happen a couple of times when the clutch failed.

 

[Vic123]

Is there anything I can do to ensure there is no overspeeding?...Maybe use a clutch??

 

[MartinSr00]

I'm not experienced with dynamometers and how all that works. If you used an induction generator tied to the electrical utility grid, it can't overspeed unless the breakers trip -- it's locked on to the grid. If it's a two pole generator, it's speed should stay pretty close to 3600 rpm (in the US) or 3000 rpm (in Europe).

Also, you should make sure that the unit has the capacity for overspeed unless the final installed system has the ability to absolutely protect against it (I doubt that it does). You don't want your power recovery turbine self destructing in front of your eyes because it goes to a no-load condition. Even if you have provisions to stop inlet flows after a trip of some sort, that will take a bit of time.

 

[JJPellin]

All of our hydraulic recovery turbines have overspeed trip protection built in as part of the unit. There are two systems that I am familiar with. One is a purely mechanical system that operates on the same principle as a typical system on a steam turbine. Other machines in our facility use a purely electronic system that uses non-contacting speed probes over a toothed wheel to sense speed. Both systems would close a valve to block inlet flow to the turbine in the event of an overspeed condition.

If you are testing this turbine in a lab situation on water, the risk of overspeed is very, very low. On low temperature water, I would consider it so low that I would consider performing the test with no overspeed protection. Running on a product with a vapor pressure above atmospheric pressure, the risk would be much higher and definitely need to be mitigated.


Johnny Pellin