System Curve

[Cordovilla]

Dear fellow EPANET users,

Do you have a specific way of working out the system curve (flow - pressure relationship) in EPANET in pumping systems?

Just to clarify: I am not talking about the pump curve, but the SYSTEM resistance curve, i.e. the relationship between pressure required at the inlet node for each flow forced through the pipe.

I currently do it by inserting a reservoir node at the inlet, then manually varying the head of the reservoir to find the resulting flow through the pipe, but I find it rather tedious. I was wondering if someone has a more "automated" way.

Thank you in advance. Jose Cordovilla

[Daniel Antonio Colocho Arévalo]

Good morning José. I´m sorry to say that I do it just like you do.

2012/3/16 Cordovilla <jose.co...@solvereglobal.com>

--
Books on Epanet and development: http://www.arnalich.com/en/books.html
 
---
www.arnalich.com
 
To send a message to the group, email: epa...@googlegroups.com
 
To unsubscribe, send a message to: epanetD+u...@googlegroups.com

--

Atte.

    Ing. Daniel Colocho
danco...@gmail.com

Paz y bendiciones a usted y su familia.

 

La única vida que vale la pena vivir es aquella en la que Dios es primero, luego y por último. Todo lo demás será añadido con abundancia en misericordia, amor, prosperidad y benignidad.

[Laurent Wismer]

Dear Jose,

You can try to put a wide & high tank instead of a reservoir, then run it through the time. As it will empty you will see the flow versus head for the plot of your tank.
Hope it helps you.

Dear fellow EPANET users,

Do you have a specific way of working out the system curve (flow - pressure relationship) in EPANET in pumping systems?
Just to clarify: I am not talking about the pump curve, but the SYSTEM resistance curve, i.e. the relationship between pressure required at the inlet node for each flow forced through the pipe.

I currently do it by inserting a reservoir node at the inlet, then manually varying the head of the reservoir to find the resulting flow through the pipe, but I find it rather tedious. I was wondering if someone has a more "automated" way.

Thank you in advance. Jose Cordovilla

---

The ICRC - working to protect and assist people affected by armed conflict and other situations of violence. Find out more: www.icrc.org

This e-mail is intended for the named recipient(s) only.
Its contents are confidential and may only be retained by the named recipient(s) and may only be copied or disclosed with the consent of the International Committee of the Red Cross (ICRC). If you are not an intended recipient please delete this e-mail and notify the sender.

---

[Roy Flynn]

Good Afternoon.
Wish I had a better answer but I use a process similar to yours only I use a node with fixed elevation and change flows to determine head. I typically generate two system curves to give an operating range by keeping node at lowest elevation and change flow rates. then change node elevation to highest level and change flow rates to generate pressures.

Just a slightly different twist to the scenario you described.

Roy Flynn, PE
MSD
Metro Division
Flood Protection Manager
Office: 1731 Brownsboro Road
Mail: 700 West Liberty Street
Louisville, Kentucky 40203
Cell Phone: (502)648-4731
Office Phone: (502)540-6514
Fax : (502)896-6365
E-Mail: Fl...@MSDLOUKY.org

>>> Cordovilla <jose.co...@solvereglobal.com> 3/16/2012 5:26 AM >>>
Dear fellow EPANET users,

--

[Cordovilla]

Laurent, your suggestion is so good and simple I almost feel embarrassed.

Thank you! Jose

[Patrick Moore]

The best way I have found to create a system curve at a pump station is to do the following:

1) Shut off the pumps using initial controls

2) disable the pump controls from turning on

3) Place a negative demand on the discharge of the pump station and an equal demand on the suction side.

4) Run the model.

5) the difference in head between the suction and discharge node is the head required to push that flow

6) If one progressively run a series of flows at 500 to 1000 gpm intervals to the max flow desired, you can generate the specific head curve for that pump station

7) be aware that you need to account for varying water levels in the tanks controlling the suction and discharge  head in each zone.  I have in the past set the suction tank to its lowest operational level and the discharge tank to just 0.01 ft below overflow as one condition ( Highest static lift condition) and ran a second series with the suction tank 0.01 ft from overflow and the receiving tank at its lowest  (lowest static lift curve) These two curves will represent the range of operating conditions for the system curve.  Lastly if you run the highest static lift under zero demands (forcing all water to the tank you have the highest system curve that will be experienced.  If you run the lowest static lift condition with the highest system demand you will get the lowest system curve condition.  Any operation of the pumps will fall in-between the two curves with the pumps generally operating more often near the lower portion of the two curves as the pumps are more often running when the receiving tank is low.  You can then overlay pump curves on the system curve and use this to guide you in pump selection.

Hope this helps,

Pat Moore

[Lucas Vasconcelos]

Just use a node with a negative demand. Epanet will calculate what pressure is needed at that node to insert that flow in the system. By using a time pattern and a demand multiplier, you can make the flow increase.

Be aware that epanet WILL have that flow in your system. If there is no outlet (a tank, reservoir or noozle) Epanet will either give you a system unstable warning or, if you allow tries enough, it will occur some weird numerical error without any warnings. I also suggest you not to try to match a negative demand with the sum of the positive ones (without any other outlets for the flow) since it will most likely cause some hard to detect numerical error. If this is the case, insert a bogus tank or something like that and check if it is receiving any flow.