Results 1 to 6 of 6

Thread: Fire Flow Simulation w/ Max. Velocity Constraint in InfoWater

  1. #1

    Fire Flow Simulation w/ Max. Velocity Constraint in InfoWater

    I have been utilizing a max. velocity constraint (15 ft/s) when running fire flow simulations in InfoWater. The idea is that velocities in excess of 15 ft/s would cause a main failure in the "real world".

    The use of a max. velocity constraint has left the office divided. On occasion we are seeing drastic differences in fire flow simulation results between using the constraint and not. This is especially evident in older areas of our distribution system where we have smaller diameter mains with high modeled roughnesses.

    I'm wondering what others are doing when running a fire flow simulation? Are you using a max. velocity constraint, or not?

    Chris Hathaway
    Service Provider for Citizens Energy Group
    Indianapolis, IN

  2. #2

    Join Date
    Apr 2013
    I have had this same question myself and have never truly come to a conclusion. I currently do not use a maximum velocity constraint but a large majority of fireflow analysis I conduct are for future development or expansion to newly developed areas where pipe age (roughness) is usually not a concern.

    Michael Quamme
    Fargo, ND

  3. #3
    The variance between using and not using this option will depend on the configuration of the model, meaning if hydrant laterals are modeled. If you are modeling laterals then the difference will be less. If you are applying this to a tee or cross intersection then it will be more.

    Without the velocity constraint, InfoWater tells you the residual pressure under the specified fireflow rate. Then it tells you how much flow can, in theory, be taken out of the hydrant while maintaining the minimum residual you specified. When modeling the hydrant lateral, you are closer to modeling a real hydrant (still missing nozzle losses, etc) so the numbers for this available flow reported will be closer to reality. In contrast, if you model fireflow at a cross joining a 12" and 8" pipe, the model will allow water to come to the node from 4 directions on large diameter pipes, so often the available fireflow is greatly exaggerated.

    When this gets applied to the velocity constraint, you may have a cross that said you can, in theory, get 8,000 gpm out of. However, this flow rate may cause a velocity of 25 fps, well over the 15 you entered. So when the velocity constraint gets applied, the available flow gets much lower. Likely in many of these cases, the available flow is higher than could be actually obtained.

    I would say the fireflow results are easier to decipher if the hydrant lateral is modeled, but there are downsides to this also. Some users deactivate laterals for normal runs, then activate them all for fireflow. As with anything, there are pros and cons to each and the key is in how the model results are viewed in light of the model schematic.

  4. #4

    Join Date
    May 2013
    As a subset of this topic: our hydrant laterals are generally 6-inch diameter lines. The hydrants, however, have 2.5-inch outlets. Would it make more sense to model a short piece of 2.5-inch line before the hydrant node?

  5. #5
    Forum Moderator

    Innovyze Employee

    Innovyze Employee

    Join Date
    May 2015

    Most people I have seen, do not include this additional pipe as they are often built directly from GIS and will only reflect the actual lateral. Given the large number of hydrants within a water system it could be very time consuming to add a short pipe as you are proposing.

    One thing to keep in mind though is that most hydrants (if not all) in the US are governed by the AWWA C502-94 standard. This standard states the following:
    1) Maximum headloss through the hydrant at 1000 gpm shall be less than 5 psi (usually on the 4 inch outlet)
    2) Maximum headloss through the hydrant at 500 gpm shall be less than 2 psi (usually on the 2.5 inch outlet)
    3) Maximum headloss through the hydrant at 250 gpm shall be less than 1 psi (usually on the 2.5 inch outlet)

    If this is the case, if you do feel inclined to add a short 2.5 inch pipe its length should be no longer than about 0.93 ft as the headloss though a 0.93 ft 2.5 inch diameter pipe with a HW C of 120 is right around 1 ft at 500 gpm. In other words any pipe added should not exceed the AWWA limits, and from what I recall most hydrants I am familiar with are usually under the maximums as well which might complicate matters.

    However, it may be easier to simply include this loss as a minor loss associated with they hydrant lateral. One could easily calculate what minor loss K would achieve the AWWA maximums.

    Fro instance for a 6 inch pipe at 500 gpm, the velocity is 5.67 ft/s. So for a max headloss of 2 and the equation for minor loss headloss ah HL_ml = k*V^2/2g or K = HL_ml/(V^2/2g) would be K = 2 ft/[ (5.67ft/s)^2/2*32.2 ft^2/s^2] or 4.00. Thus to simulate the AWWA max headloss in a typical 6 inch hydrant lead one would simply add a ML of 4 to the 6 inch hydrant lateral to add 2 ft of headloss at 500 gpm.

    While this could work for a single flow, it would be different at other flow rates. For instance at 1000 gpm and 5 ft of headloss the K would be roughly 2.5 for a 6 inch line. But this change would eliminate the challenges with adding a small leader line, especially given that there are two 2.5 and one 4 inch outlets on most hydrants.

    I think for many modelers, they are comfortable enough with the standard results that they do not account for these additional losses especially given the challenges associated with adding short pipes or minor loss coefficients to each hydrant lateral. Certainly most models I am familiar with and industry standard practice does not account for any specific loss at the hydrants. In fact many models don't even include the hydrant laterals which would more often have more loss than the outlets would.

    Anyway, hopefully this will provide some guidance for you so you can make the best decision for what will best reflect your modeling needs.

    Patrick Moore
    Innovyze Support.

  6. #6
    Forum Moderator

    Innovyze Employee

    Innovyze Employee

    Join Date
    May 2015

    One thing to keep in mind when using the max velocity constraint is that it is highly dependent on the pipe search range used.

    Just to the right of the Velocity constraint check box is a area that says "Pipe Search Range". This specifies which pipes are checked for velocity during the fireflow run.

    Image of fireflow tab in the Run Manager (click for a larger image)
    Fireflow Dialog - Pipe Search Range.jpg

    If you only select "connecting pipes" the model will only check velocities on the pipes directly connected to the node where the fire flow is running.

    If you have pipes in the domain there will be a second option to use "Domain Pipes" as the velocity check. This can allow the fire flow routine to run a complete check of velocity under fire flow for all pipes in the domain, not just those directly connected to the fire flow node. This is powerful as this is usually what the engineer/analyst would like to see as if using this tool, you want to verify the highest fire flow while keeping velocities below a specified value.

    In the many years I was an Engineering consultant running fire flow analyses, like your office, I often felt conflicted with using the velocity constraint. Obviously, using the pipe velocity tool will have the potential to lower the fireflow available at a location, but the pipe velocity at times was not as much of an issue for locations with lots of head to burn (i.e locations with a high static head). What I typically observed was that checking pipe velocities was a good way to find pipes that were likely specific areas where there were pipe constraints were limiting the available fireflow at a location and that knowing where these were could help guide the engineer in identifying where additional pipe capacity may be needed to meet the desired fireflows. In certain systems where there was enough head to burn such that nominally exceeding the pipe velocity did not impact the desired available fire flow, the velocity constraint seemed unnecessary. So for me, I usually ran the analysis without it first and looked at where the deficiencies occurred and then looked at pipe velocities in those specific cases. But its not a bad idea to use the constraint and back check areas with less than desired fireflow after the fact. As noted by Kedric below, the most useful factor in using the velocity constraint feature may be in capping the high available fireflows the model will calculate to more reasonable levels.

    Unfortunately I don't know of any specific guidance documents regarding modeling fire flow analyses that are available to provide guidance.

    I found that as an engineer, I was often more concerned about the potential variability in initial conditions being used to run fire flow analyses from different designers as different tank levels and different pump operation conditions can result in very different results as well. We would often encourage each water system to develop a specific set of operating conditions to use for each fire flow analysis, so that results were at least consistent. This meant deciding if pumps would be running or not, and what level the tanks would be set at (usually somewhere between 50-70 % full), but that is a separate issue!

    Perhaps once day a group like the AWWA Engineering Applications Committee, can work to develop fire flow analysis guidelines, but until then, it will come down to engineering judgment of the modeler to decide what is best for the analysis at hand.

    Innovyze Support
    Patrick Moore

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts