runnig without reservoir

[Maruf Muntasir]

How can I successfully run a model which has 3 pumps in EPAnet 2.0 without any reservoir or tank 

[Santiago Arnalich]

Not really, you need a reservoir to act as a water source or a sink to avoid epanet errors. Also, the other way to put water into a system is with a node with negative demands, but that usually ruins things with unreal results  as it forces the water in "no matter what". .

Why not pump from a reservoir? (As a remainder a reservoir is not a tank or a man-made "reservoir", but an infinite source of water which level won't change (a aquifer, a lake, a river....)

I hope it helps.

On Sat, May 23, 2015 at 10:17 AM, Maruf Muntasir <maru...@gmail.com> wrote:

How can I successfully run a model which has 3 pumps in EPAnet 2.0 without any reservoir or tank 

--
--
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

---
You received this message because you are subscribed to the Google Groups "Epanet and Development" group.
To unsubscribe from this group and stop receiving emails from it, send an email to epanetD+u...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

--

Santiago Arnalich
Consultant, lecturer and author
+34 671700686
skype: sarnalich
...
www.arnalich.com
Consultancy & training for Development

[Maruf Muntasir]

So reservoir/tank is mandatory for successful run?

 

You received this message because you are subscribed to a topic in the Google Groups "Epanet and Development" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/epanetD/MlkHaTWnvxU/unsubscribe.
To unsubscribe from this group and all its topics, send an email to epanetD+u...@googlegroups.com.

[Sam Hatchett]

this is correct. as per the manual, Error 223 is returned for invalid networks: "There are not enough nodes in the network to analyze. A valid network must contain at least one tank/reservoir and one junction node."

[Yessarian]

I know this is an old issue but in case someone else has this problem, or wants to circumvent the error that comes from not connecting a tank or reservoir to a network, there is a simple way to do this. You can just place a reservoir anywhere in the model, leave it unconnected and then the model will run without an error. To clarify though that I have used this primarily for the sizing of pumps for a pumping main. So, the output/let end of the line I use a node with a high emitter coefficient plus zero demand and on the pump end I place a node with a negative demand. The required head of the pump to meet my demand through the pipeline is then calculated at the starting node. It is a very nice approach. We have just completed a long pipeline in a refugee camp in Northern Uganda using this approach and the predicted head requirement is pretty much spot on and the difference between model and real world is probably more as a consequence of my choice in roughness coefficient and/or minor losses rather than anything else, which will always be the case. It is, at the end of the day, a model and will never be 100% accurate.

@Santiago - agreed that this is forcing water through the system but placing demand at outlets does the same thing and this is a very usual approach to network modelling. This causes worse errors, witnessed if you take the a network and model using emitters nodes and the same modeled with demand nodes, which is much more realistic.