Sizing drinking water reticulation systems 1

[Arniston]

I want to size a drinking water plant for a small community of 200 homes and an estimated 600 people.

Can someone please tell me what rules of thumb are used to relate peak design flows to the average daily flow for a network of this size?

 

[semo]

You are a little short on info.

1. Will the system provide only potable water or also fireflow, industrial, commercial, institutional, etc.?

2. Is this in a rural area? Is it upper or lower income area.

3. How much storage will be provided?

All of this has an affect on the water use.

If this is something to be approved by a state agency, you will need to follow their guidelines.

 

[bimr]

Semo has some good points. But you can make some assumptions.

First, assume that you are talking about residences instead of commercial water uses. Secondly, assume that fire flows are not part of this exercise.

A rule of thumb in the USA is about 105 gallons per day per person for average daily water consumption or 44 gallons per minute averaged over the 24 hour day.

The peak hourly flow depends on the factors Semo stated, but you can estimate the peak hourly flow of between 300 gallons per minute (for a development density of 4 units per acre) to 750 gallons per minute (for a development density of 1 unit per acre).

You can buy Water Supply and Pollution Control by Viessman and Hammer on bookbinder.com

 

[Arniston]

Thanks guys for your help. Yeah, I should have been more specific.

The development is residential about 200 houses (6 units per acre) in a middle income area.

The reticulation would be for internal potable uses only. External and firefighting is provided via a separate non-potable supply sourced from reclaimed water (a combination of membrane treated secondary effluent and site stormwater). The latter system would be sized conventionally.

The potable supply is from the municipal authority so storage is not an issue. The authority charges a development contibution based on average and maximum potential demand. It is the potential internal maximum demand for the whole 200 unit development that I am having trouble figuring out.

 

[stanier]

Suggest you go to

http://www.haested.com.[URL unfurl="true"]http://www.haestad.com/library/books/awdm/agree.asp

[/URL]

Go into the library and online you will see their book Advanced Water Distribution Modelling.

Enter Civilquiz and get as a prize a discount on the book. Normally USD195 but you can get up to 50% off. It comes with free Watercad but only 25 nodes.

You can use Epanet, which is free, for extended period simulation. This enables you to size pumps, pipes and reservoirs. particularly if you need to cover variations in demand for diurnal and seasonal flow rates. It will also cover fire situations etc.

 

[bimr]

If you go to the referenced textbook, on Figures 4-3 and 4-4 you will find:

Maximum day value = 700 gpd per dwelling unit or 140,000 gpd or 97 gpm. The peak hourly is 225 gpm.

 

[semo]

I think Bimr has similar numbers to what I would use.

We typically figure 100-105 gallons per capita per day for average daily flow with 3.5 persons per home (Census data can verify that number). Peak daily flows I've seen range from 1.5 to 1.7 times the average daily flow. This would give you 73,500 gallons per day average with 117,600 gpd peak daily.

For small developments I would figure the peak instantaneous flow as I do rural water systems with the equation 9*n^0.515 where n is the number of persons. This would equate to 263 gpm.

As I said, very similar to Bimr's numbers.

 

[ptmoss]

Am I correct in reading your projected water use as 100 gallons per day per person?

 

[semo]

Yes. Typically, the more industrial a community, the higher the average daily use per person. That is based on average daily flow divided by population served.

In rural areas where there are no industrial users or fire flow, I see this value closer to 60 gallons per capita day (gpcd). Similarily, when the industrial users and their flows are removed from the equation, I typically see the value drop well below 100 gpcd.

100 gpcd gives a safe number to use in a municipal setting.

 

[bimr]

ptmoss,

Just to be clear. You have to design your system for the peak flow of 225 gpm not the average daily flow of 97 gpm. These numbers are based on water studies and population densities. Since it is generally important to document where your project assumptions come from, I referenced the book:

"Water Supply and Pollution Control" by Viessman and Hammer. This is a standard water resource that is probably available on bookbinder.com

 

[JasonG]

I am working on a similar project, though much smaller, only 30 people in 24 units. Could Semo, or anyone else, tell me where the peak instantaneous flow equation (9*n^0.515) came from so I can reference it?

Interestingly, this peak factor coresponds well with the fixture unit method.

Thank you

 

[semo]

USDA in Missouri has been the driving force for the use of this equation here. It is called out as coming from the Missouri Division of Health (original DNR). It was presented in a Health Department circular circa 1966 as 12*n^0.515; but, data seen by USDA shows this to be a bit excessive.

I have a printing of an article (provided by USDA) from "Water and Sewage Works, July 1951" written by D.R. Taylor, General Manager that provided a table of water use based on the number of houses and class of houses. These values were obtained through metered users over several years of testing. This table, when plotted, graphs the Class 3 home (Avg good subdivision with 80% two to 3 bedroom houses with 1 bath and 20% two to 3 bedroom houses with 2 baths. Houses owned by occupant, with average amount of lawn and shrubbery requiring average amount of lawn sprinkling) at the 9*n^0.515.

I don't know if this information is printed elsewhere; but, from experience, it is working on our systems. I've had system owners come to us (after extending lines to pick up additional users) with pressure/volume problems and this equation, when used, shows the problem to be real similar to what is found in the field.

 

[JasonG]

Thank you for responding. I'm glad I asked - the Class 3 description does not fit this Site. My project consists of 24 single bathroom, single bedroom elderly housing apartments in 3 bldgs. One of the buildings has a common laundry room with 2 washers. There is 1 hose on each bldg.

If it isn't too much trouble, would you mind providing the appropriate info for this type of development?

Thank you.

 

[semo]

Looking at the Class 2 houses ("Small houses with one bath, on small lot. Small rental houses and small rental duplex houses. Very little lawn sprinkling. Either side of a duplex is considered as one house."), the coefficient would be 7.3.

Class 1 houses ("Sub-Standard houses with barest minimum of plumbing where little or no lawn sprinking is expected") will have a coefficient of 5.75.

Remember that this design method does not take into account any fire flow. It is domestic use only and give the change in times, I don't think I would use the Class 1 houses for any design. I would rather err on the side of caution. Doesn't cost much to install oversized lines; but, it certainly costs to replace undersized ones.

Even comparing the 7.5 coefficient to the 9 coefficient with 24 units, you are only talking about 3 gpm difference. I'll almost bet, the same size pipe will be used for either and the coefficient won't make a hill of beans in this instance. Butttttt, I don't have all the data you have to support my hunch.

 

[JasonG]

Great thanks a million!