Booster Pump Sizing based on Peak Hour Demand 1

[jobarr]

Question: Is it legitimate to use peak hour demand for the sizing of domestic water booster pumps?

I am working on a project where we need to size a booster pump skid to provide domestic water for a subdivision with 224 homes. There is already a separate fire pump which will handle fire flows, so our booster pump only needs to supply domestic demand.

Based on a year's worth of water meter data in the area, the average daily demand was determined to be 350gpd/house. This seems to be in-line with general guidelines for water usage in our region.

In order to determine the peak hour demand, I did the following: (350gpd/house)(224homes)(1.1 to account for losses in the system)(peaking factor of 4). This gives me a peak hour demand of 240gpm (or 14,373gph). For my scenario, is 240gpm a reasonable number to use as the basis for designing a booster pump skid?

Based on what I have been able to find by searching around on-line, this seems like a pretty realistic approach. However, when I step back and think about it, I get a little uneasy. It seems like it would be pretty easy for a 224 home subdivision to have an instantaneous demand that exceeds 240gpm. Is there an additional peaking factor that I should apply to the 240gpm in order to account for instantaneous demand?

This booster pump will supply all of the domestic water to the subdivision, and it provides the necessary pressure for the system. There is a storage tank next to the pump house that will provide adequate supply for the demand, and we will likely be installing a pressure tank to prevent the pumps from having to cycle on and off constantly.

Thanks in advance for any help you can provide. If there are good references that provide guidelines on this, that would be appreciated. Also, if anyone has a good guideline on how to choose a peaking factor based on the size of the population being served, that would also be much appreciated.

 

[bimr]

There is not much information available. The peaking factor is based on the population. You are correct for a small population, that the hourly peaking factor for a neighborhood will be greater than "4" , not just "4X" for all applications

Fair and Geyers textbook, "Water Supply and Wastewater Disposal" has a discussion on this. I looked at this some time ago and would recommend an hourly peaking factor of 7 for your project.

Other factors are the wealth of the community, size of property, and lawn irrigation.

http://www.eng-tips.com/viewthread.cfm?qid=80061

 

[ashtree]

Its most likely that the biggest peak flows will occur related to lawn irrigation. You may need to check if local ordinances, water tarrif or demand management initiatives influence the time or the way people use irrigation systems and water in general.
Many demand management initiatives are aimed at reducing peak flow rates even if they do not have a massive effect on overall water consumption.



Regards
Ashtree
"Any water can be made potable if you filter it through enough money"

 

[QualityTime]

For this type of flow variance you will need more than one pump instead of one big pump. The pumps will have to be staged to come on based on pressure

 

[jobarr]

Thank you all for your help.

With all of the domestic water booster pump stations in the US, it is surprising to me that there isn't a more common method to dealing with this situation.

Bimr, thanks for the reference for the peaking factor. In the link you provided in your post, there is reference that says the peak flow should be determined with the following equation: 9*n^0.515 where n is the number of people. Apparently this equation is from an old USDA document. Any thoughts on this?

I will need to write up a summary of my design approach for this project, so providing good references will be important.

Thanks again.

 

[77JQX]

You really need to understand the local variables that affect your peak flow. As Ashtree says, irrigation demands can drive the peak flow. If there is no outdoor irrigation demand, then you can use sewage flow peaking factors to get indoor usage peaks. But outdoor irrigation use will vary depending on location: Las Vegas is very different from Oslo. We can't help you with outdoor peaks without knowing your climate etc.

 

[jobarr]

77JQX,
That is a good point. Irrigation could be a driving force.

The project is located near the coast in Northern California (almost Oregon). The climate here is very mild and we get plenty of rain and fog. Many of the homes in this subdivision have landscaped yards, but I would expect irrigation usage to be quite a bit lower than in places like Las Vegas or Southern California.

Are you aware of any guidelines that would be good for determining irrigation demand for subdivisions? I am somewhat familiar with the AWWA M22, but this approach seems more valid for determining irrigation demand for individual commercial developments, not subdivisions with lots of homes.

 

[bimr]

That equation is for rural development and the results will be in the ballpark. The thread also has a discussion on lawn irrigation.

Fair and Geyers textbook also has a discussion on lawn irrigation.

 

[cvg]

suggest you find another water system, nearby, that serves a development similar to yours and then see if you can get some data from their operations. pick up the phone and ask around. you should be able to get some good information that way and way better than some emperical formula in a text book that will get you in the ballpark.

 

[jobarr]

Thank you all for your input.

After extensive on-line searching, I came across a few different references that provided some insight, but nothing that was very enlightening. Based on the information I found, typical domestic water peak hour factors range between 3 to 10 times the average daily demand, with smaller facilities having peaking factors on the higher end of the scale.

Most municipalities seem to use peaking factors of 2.5 to 4, but these are usually for larger systems than what I am dealing with.

Fortunately, we found an old document that described how the existing booster pump skid was originally designed to operate (the operators weren't sure about the proper pumping operation for the skid), and this document told us the maximum capacity of the existing booster pump that we are replacing. We then analyzed the monthly meter data for the existing portion of the subdivision, and were able to determine that the existing booster pump was sized for a peaking factor of 6. So, that is what we are using for our design, which will provide domestic water to the existing portion of the subdivision, and the planned expansion.

So, I guess we have solved the problem for now, but it sure would be nice to have a more reliable/repeatable approach to solving this problem in the future. It sounds like the best way to deal with this problem next time will be to track down reliable pumping and usage data from nearby areas. That is easier said than done, but at least we have a place to start the next time we do this.

Thanks again for your help, and if any of you come across any good referenced on domestic water peaking factors, please let me know.

 

[agua84]

jobarr-

I've been in similar situations with respect to estimating design flows, so I appreciate the thread. Here's an example of a peak to design flow relationship, for well-based small community water systems in New Hampshire - see table 405-5 on pdf page 13 in:

http://des.nh.gov/organization/commissioner/legal/rules/documents/env-dw405.pdf

Section 405.10 (pdf page 8 & 9) has some guidance on determining the "design flow" as used in the standard, either based on (1). a generic table (i.e. 150 gpd/bedroom, not including "exterior" use), (2). 2 to 3 times the historical average daily flow over 12 months of water meter data, or (3). the historical highest daily flow recorded over 12 months of daily meter readings.

Obviously this is written for a specific climate and jurisdiction, local info is always best.