Results 1 to 4 of 4

Thread: Fire Hydrant head losses

  1. #1
    Member
    Join Date
    Dec 2016
    Posts
    57

    Fire Hydrant head losses

    I am performing some hydrant curve simulations. However, I got a doubt.
    Fire hydrants also experience a local head loss.
    Is there an option to include a local head loss in a fire hydrant node? Is it ok to increase the node elevation to create the additional head loss?

  2. #2
    Forum Moderator

    Innovyze Employee



    Innovyze Employee



    Join Date
    May 2015
    Posts
    419
    Vladimir,

    I hope all is well. You are correct that actual flows from a hydrant may differ slightly, but this is not generally what the fireflow tool or the hydrant curve tool is being asked to do.

    For both a hydrant Curve and the fireflow tool, the question being asked is how much water can I put at a junction and what will impact will that have on the system pressure within the model.

    A hydrant curve essentially runs a series of demands using the interval specified until the pressure at the junction is at or just above 0 psi. This produces a curve that identifies how the system responds with the increase in demand at that location. This represents a curve of how the system pressure at the hydrant would change as the flow increases and helps the user understand mathematically within the model how increased demand (from a fire) will impact the system pressures. This is a great way to assess the system pipe capacity which is generally the major limitation restricting the flow available.

    Example Hydrant Curve
    (click if need larger image)
    Hydrant Curve.png

    In the Fireflow tool, the model represents results which indicate how much total demand can mathematically be placed on a junction to get exactly 20 psi (or whatever minimum residual pressure you use) at the junction tested as the Available Flow number, or what Design Flow can be placed on a junction and still satisfy the pressure constraints identified in the search range (Generally at or less than the available flow). These values are all inclusive of demand at a junction, because they represent how much actual water could be pulled at that location to meet the criteria specified. This means if manually checking results you need to subtract the actual demand out or you might get pressures just under your pressure limit. Since if a Fire is occurring one would expect water use to stop at that location ( with people leaving for safety) and or the fact that system demands are generally much much lower than the fireflow that the differences are minimal, but again it comes down to what the tool is "asking" how much water can be placed here (in total) and meet the criteria specified.

    If one is doing "flushing" like in a unidirectional flushing analysis one wants to predict with high accuracy what the best estimate of the flow that would come out of a hydrant flowing to atmosphere. In these cases one generally wants to know as accurately as possible the flow that would come out of the hydrant. in our UDF software we assign an emitter coefficient to the hydrant that is calculated based on the Nozzle diameter and the Orifice Flow Coefficient (a value usually ranging form 0.6 to 0.9 that is usually developed by field experiments using pitot readings on the flow.). This allows the UDF tool to predict the actual expected flows from the hydrant with greater accuracy.

    If you care to see how these are performed you can review the UDF User guide which is installed with the software even if you do not have a copy of the UDF software. The file is called "InfoWater UDF User Guide.PDF" and is located in the C:\Program Files (x86)\InfoWater\Help directory of your computer. Please note that while the UDF calculator can easily handle US and SI units, the formula shown in the UDF User Guide is for US units only and is highly unit specific. But it is essentially what can be found in textbooks or via google as for governing flow out of a nozzle to atmosphere, but you will want to find one specific to SI units if using those.

    Here is a brief excerpt :
    (click if need a larger image)
    Hydrant Emiiter Calculations - UDF.jpg


    For your situation you are concerned about the losses in the hydrant nozzle which is an understandable concern. Since there are an incredible number of hydrant models and flow nozzles, it is not really feasible to include this within the model software, but you need to ask yourself two questions

    1) How much of a difference would it really make?
    2) Will it even matter if most hydrants would be connected to a fire truck with a pumper?

    Generally as a modeler/engineering designer you are concerned with the actual system piping itself and will the system piping and facilities itself have enough capacity to provide the needed fireflow to the hydrant itself. As long as the system can provide what is necessary at the extraction point, this is generally what concerns the "engineering side" as most of the flow restrictions are caused by the pipelines rather than the nozzle itself. There will most assuredly be losses in the long hose lines and in the hydrant itself and this is why most fire fighters make use of trucks with pumping systems after the hydrant to make up for losses that may occur.

    In addition, most hydrants (at least in the US are built with specifications limiting the hydrant to have a maximum of 5 psi headloss at 1000 gpm, with many manufacturers often greatly under those limits, such that while there may be losses, those are often small. Thus this contributes why this is not generally an engineering concern as much as the firefighter concern and this is accounted for by using a pumper truck.

    If you are really concerned about this, I would simply work out adding an equivalent minor loss to one of the hydrant lines and make sure you are running the analysis on a hydrant connected by the smaller hydrant connector pipe rather than running it on a junction right on the main itself. They "hydrant stub" itself will induce a significant headloss and if not included your results would likely be highly overestimated compared to the results with the hydrant stub included. This should be able to account for any additional losses if you are really concerned about them. Minor Loss headloss = HL =K*( V^2/(2G)) where V is the velocity and G is the Gravity constant and K is the Minor Loss Coefficient.

    If you raise the node elevation it will simply reduce the pressure at the junction as Pressure = HGL - Elevation (with unit conversion as needed to get your pressure units). This would in essence shift the hydrant curve down. But you may wish to use a hydrant elevation that is more in line with where the nozzle center-line is rather than the typical ground elevation if you want to get nitpicky. Any field pressure comparisons would need to account for the correct elevation to get accurate pressure comparisons.

    Its really up to you , the model does not directly account for this, but as I noted, you can find ways to account for it. But there are pretty good reasons for it not including this as it's not really the question the model is asking, and it often is not really the concern of the engineer who is focused on is there simply enough pipe capacity to get the water to the hydrant which is usually the primary goal of the modeling analysis. In addition the hydrant and nozzle loss if generally minimal compared to the system losses and are often accounted for in the pumping truck the fire fighters use.

    Please feel free to respond if you have follow up questions on this issue.

    Patrick Moore
    Last edited by Patrick Moore; June 19, 2018 at 11:35 AM.

  3. #3
    Member
    Join Date
    Dec 2016
    Posts
    57
    Thank you very much
    I got the doubt because I got good results in the calibration; about 2 psi difference and identifying critical bad condition pipes with very low C (some already verified a leak or scaling). Thus, I wanted to get more detail
    I got interested into this detail because 4 psi difference in hydrant curve may be equivalent to more than 400 gpm (as shown in the figure). I was trying to get closer in replicating an in situ hydrant curve

    Thanks for the comments

  4. #4
    Forum Moderator

    Innovyze Employee



    Innovyze Employee



    Join Date
    May 2015
    Posts
    419
    Vladimir,

    If you are attempting to calibrate hydrant flows using the model, I would generally not use the hydrant curve or fireflow tool either as they are often not the best way to compare model and field results. The best way I know to do that is to set up scenarios where the boundary conditions match (both field and model) and to compare the pressures and pressure drop between the model and the field. A common calibration criteria used in the US is to match pressures and pressure drop to within 5 psi.

    Hereare some threads on the User forums that may be useful to you where we have already discussed calibration suggestions:



    Very slight differences in the available head can easily change the predicted results at 20 psi (or whatever you are using for your fire minimum pressure) and it is generally best when calibrating to compare actual flows rather than "predicted" values at some pressure limit. This is also why when doing hydrant tests you want to really stress the system and have between 10-25 psi pressure drop when flowing to best have enough pressure drop to calibrate your model. This may mean you have to flow more than one hydrant.

    Patrick Moore

Tags for this Thread

Posting Permissions

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