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

Water modeling - flows passing through a closed pipe

I am attempting to model a connection an existing distribution system.

The goal is to supply two houses with drinking water and fill tanks for fire flows.  What I have done is brought the water to the housing complex, then one line proceeds to the houses where it splits and services the houses.  another segment diverges and heads up hill approx. 50 feet to a tank.  I have controls set that close the line to the houses and open the line to the tank when the line from the tank to a fire hydrant is opened, I accomplished this by setting simple controls that tell them to open or close based on the level within the tank.  The tank should be stagnant and only drained when the hyrdant line is opened.  This works fine when I set the simple controls to trigger when the level in the tank goes below 4 feet (max height is 6 feet).  There are also simple controls set to open the fire hydrant line at 7 AM and close at 8 AM with it's initial status as closed.  Problems arise when I change the simple control height to anything above 4 feet.  The line from the hydrant will open at various times throughout a 24 hour period even with the controls telling it to only be opened between 7 and 8 AM. 

So, my question is; how is the link from the tank to the hydrant overriding the controls and opening at other times?

My guess is it has something to do with the in ability of the gravity system to to fill the tanks beyond 4 feet.  There is no pump on the line from the distribution line to the tanks, so I am relying on the head in the line to be able to refill the tanks.  But still why this would cause the line to the hydrant to open at random times I am not sure. 

Also, if anyone could help me figure out how to find out what the maximum pressure and flow I could achieve at the hydrant that would be great.  currently it tells me I have 640 or so gpm but zero pressure, which might be true considering I have less than 1 psi at the tanks and only a decrease in elevation of 47 feet from the tanks to the hydrant. 

One more, I want to know at what pressure if any the hydrant could achieve 500 gpm.  Do I set the base demand at 500?  How would my pattern look for this if I just wanted the hydrant to flow for 1 hour.  Currently I have the multiplier set to 1 between 7 and 8 AM and 0's for all other hours of the day.  the base demand is set at 500.  The patterns do not seem very intuitive to me.

I have never used a water modeling program prior to this and have taught myself from the manual.

Any help would be great, even "you are in way over you head!"

Thanks.

  • If the pipe from the tank to the hydrant is long, you may want to consider locating the tank next to the hydrant and placing a fire pump in between. That coudl be less expensive.

    For aesthetic reasons, you can place the tank and pump underground.

  • To be legal you should separate your potable and non-potable water. You can do this with an air gap or backflow preventer on the tank fill line. Once you do that, run a non-potable fire line from the tank to the hydrant. Yes, that may cost a bit more but you can make the tank fill line very small and lay it in the same trench (with adequate separation) as the larger fire pipe.

    You could even connect the fire line and potable line at some other location as long as you have a backflow preventer.

    Best of luck,

    Tom

  • I do realize the issue with the possible contamination of the water supply from the tanks, and that my current plan is not ideal or possibly legal.   It may be that once the tanks are full they can be manually disconnected.  I would just hate to have that water supply there without the possibility to use it if the tanks run dry and water is still needed.  As I become better with EPAnet I plan on exploring different options. Perhaps once I figure out modeling the water quality the turnover of the tanks will not be an issue.  According to NFPA I only need about 5000 gallons of storage.

  • To answer your questions.

     

    1) closing of the houses:  These two houses are at the end of a dirt road and the only two structures that would require fire fighting.  My thoughts were if there is a fire in either one of them, because of their close proximity that no one would want to use water if the other house was burning.  However, I did model the system as well by removing the control on the link to the house and it appears that it would be sufficient to still allow flow to the houses just at a reduced pressure of about 30 psi.

     

    2) Stagnant water:  I knew when I used that word someone would bring up that point.  The reason I would want the tank to be only for fire flow is because of water quality issues.  My thought was to use a double check valve and/or a valve that would only open when the hydrant was activated.  What makes this an interesting situation is originally this house was supplied water from well that was pumped into a series of 2500 gallon storage tanks that then supplied the houses via gravity.  That system does not work any longer.  Therefore, we have been granted permission to tap into an existing small community distribution system about 1500 feet away.  We could meet the water quality issues by supplying the houses directly from the distribution system, because where we would tap is directly outside the chlorination shed.  However, meeting fire flows has been a headache, while trying to maintain a gravity system only.  This is why I decided to look at using the existing tanks to store the necessary volume of water per NFPA 1142. 

     

    I thought that might be the case that the model would satisfy the demands of the hydrant even if the pipe should be closed.  However, the hydrant should not be demanding any water except between the hours of 7 and 8 AM.   The pattern I have applied to the hydrant only demands water between 7-8 AM.

     

    Regarding the hydrant, i currently have the emitter set at 50,000 because the manual said to set it at 100 times the desired demand when trying to determine the maximum flow available  at particular pressure. (page 132 manual).  If I get what you are saying; I should not have an initial demand set for the hydrant node when attempting to determine the maximum flow available? Just an emitter coefficient and elevation + desired pressure head.  After I determine the available flow I can remove the emitter coefficient and just add the desired flow?

     

    I will attempt to attach my model, It is fairly elementary, as it is very rough and based off of "eyeing" the layout of the pipe versus actually surveying it.   My goal is to just get an idea of what is possible and what the options could be.  The existing system is modeled as a large reservoir with an abundandt supply of water. I set the length of pipe between this reservoir and the point of connection to the actual distance from the point of connection to the nearest tank in the system which would be supplying the flow to my system.

     Sorry, I am not seeing how I would attach my file.

    Thanks for your help I really appreciate it.

     

  • Sounds like you are going through some of the usual teething troubles people have when they start to do hydraulic modeling. We should be able to help you out there, but first a couple of questions about your overall design:

    1. If you close the pipeline that goes to the houses once the tank elevation drops, how do the houses get water?

    2. Are you sure you want a tank full of 'stagnant' water connected to the same network as you use to provide drinking water? You would need to be extremely careful not to allow backflow from the tank to the houses, otherwise you could have a disaster on your hands.

    Now regarding your modeling questions, it is a bit hard to tell without seeing it. You could attach the model here, or (if you have a SELECT subscription...which you may not if you are using EPANET) submit it to tech support.

    One thing to note is that in many hydraulic models the demands are always satisfied, even if that means water has to travel through a closed pipe. Yes, this is unintuitive and it gives some strange results...but my first guess is that this is what you are seeing controls overridden.

    Your pattern for the hydrant seems ok. If you have set an 'emitter coefficient' to model the hydrant you will need to set that to zero when you add the 500 gpm demand.

    Hope that helps a bit.

    Regards,

    Mal