This discussion has been locked.
You can no longer post new replies to this discussion. If you have a question you can start a new discussion

Hammer: Turbine Curve in Relation to Quadrant Curve

Hi
I wish to clarify how the Turbine Curve should be sourced for inputting into Hammer in relation to which quadrant curve is being used.

I understand the Quadrant curve is defined by the selection of the specific speed. In my case, I am picking a specific speed as close as possible to the actual turbine I am trying to model. For example; My turbine is 200, I am picking 170 in Hammer.

When selecting my turbine curve (H vs. Q), should I try to match this to the specific speed of the quadrant curve in Hammer (Ns=200) or to the specific speed of the actual turbine (using turbine hill chart) for the actual turbine (Ns=200) even though the hill chart wll not match the qadrant curve being used in hammer? Which will produce the greater accuracy?

Thank you for your assistance.

Parents
  • Hello Mattyjb,

    When you select a specific speed for your turbine, this tells HAMMER to use a predefined quadrant curve (actually a family of quadrant curves, one for each position of the wicket gate) during the transient simulation. The points on the quadrant curve are relative to the initial operating point of the turbine. The initial operating point of the turbine is based on the Turbine Curve that you're asking about.

    So, technically it does not matter which you choose since the values are relative. However, as far as which is most "accurate", that's an interesting question, which has two parts. First, in order to be comparing "apples to apples", logically it would make sense that if you select the 170 specific speed, you should base the turbine curve on the 170 speed. On the other hand, to match the actual performance of the ns=200 turbine, it would be more accurate to use a specific speed of 200 in the model. However, that would require entering your own quadrant curve for each position of the wicket gate. You can find documentation in the Help on how to do this, but its often difficult to find this data, and estimating based on the closest available speed (170) is often a good enough approximation.

    So, it's hard to say which turbine curve you should use in the model since you're using the 170 speed. Technically neither will exactly match the exact ns=200 turbine in the field. It's very possible though that there may only be a slight difference in the initial operating point (initial flow and head drop) between either option, so I would recommend a sensitivity analysis. Meaning, try the curve corresponding to the 200 speed, run the model, then try the curve corresponding to the 170 speed, run the model and compare the results. If they isn't a significant difference, then you may not need to worry about it. if there is, you might consider picking the one that gives the most conservative results, or perhaps pursue entry of the actual ns=200 curve.

    Hope it helps,


    Regards,

    Jesse Dringoli
    Technical Support Manager, OpenFlows
    Bentley Communities Site Administrator
    Bentley Systems, Inc.

Reply
  • Hello Mattyjb,

    When you select a specific speed for your turbine, this tells HAMMER to use a predefined quadrant curve (actually a family of quadrant curves, one for each position of the wicket gate) during the transient simulation. The points on the quadrant curve are relative to the initial operating point of the turbine. The initial operating point of the turbine is based on the Turbine Curve that you're asking about.

    So, technically it does not matter which you choose since the values are relative. However, as far as which is most "accurate", that's an interesting question, which has two parts. First, in order to be comparing "apples to apples", logically it would make sense that if you select the 170 specific speed, you should base the turbine curve on the 170 speed. On the other hand, to match the actual performance of the ns=200 turbine, it would be more accurate to use a specific speed of 200 in the model. However, that would require entering your own quadrant curve for each position of the wicket gate. You can find documentation in the Help on how to do this, but its often difficult to find this data, and estimating based on the closest available speed (170) is often a good enough approximation.

    So, it's hard to say which turbine curve you should use in the model since you're using the 170 speed. Technically neither will exactly match the exact ns=200 turbine in the field. It's very possible though that there may only be a slight difference in the initial operating point (initial flow and head drop) between either option, so I would recommend a sensitivity analysis. Meaning, try the curve corresponding to the 200 speed, run the model, then try the curve corresponding to the 170 speed, run the model and compare the results. If they isn't a significant difference, then you may not need to worry about it. if there is, you might consider picking the one that gives the most conservative results, or perhaps pursue entry of the actual ns=200 curve.

    Hope it helps,


    Regards,

    Jesse Dringoli
    Technical Support Manager, OpenFlows
    Bentley Communities Site Administrator
    Bentley Systems, Inc.

Children
No Data