Water supply pump and pipe sizing

[Mech110]

Hi,

I have a pump sizing task at hand and want some expert opinion. Pumps have to supply domestic water from an underground water tank (ugwt) of 237,266 gal. to 2 overhead water tanks (ohwt) each of capacity 39,688 gal. meaning in total approx. 79,000 gal. According to UPC, total water supply demand is 650 gpm. Adding roughly irrigation and factor of safety I take it to be 700 gpm. According to my understanding of the water supply demand calculated from UPC, it already includes diversity factor. Therefore, we can say that the ohwt will be emptied in 79,000/700 = 113 mins. Therefore, I have to supply water in these tanks at a greater rate so that the feeding pump (from ugwt to ohwt) does not run continuously.

1. If I select 3 pumps of 1250 gpm each to run in sequencing (through electronic controls), each pump will run for approx. 1 hour in every 6 hours so the duty cycle will be 1/6 = 16.67% (I am not sure on this duty cycle terminology..correct me if I am wrong). Is this approach fine? or I need to change something?

2. Building is approx. 150' high. A 74 psi head pump will be fine? Secondly, I plan on supplying water to ohwt using 8in dia. PPR pipe. Is it an optimum choice? It gives the velocity as 10ft/sec and friction loss approx. 1.5 psi/100 ft from "Fairly Smooth" chart of UPC (chart A 105.1(2)). I wish to use 6in pipe but that would give me a higher velocity and friction loss as well or i have to reduce the flow from 1250 gpm.

3. Lastly, due to the smaller capacity of the ugwt as well as ohwt (architect is not allowing them to be bigger), I plan on using 60,000 gallons in ugwt for firefighting purposes using fire pumps as the building is partially sprinklered and dedicate ohwt to domestic water use only. I will have to use a booster pump for the top most floor to make pressure at fixtures greater than 8psi and use pressure reducing valves at lower floors to keep it lower than 80 psi. What do you say? Am I missing anything?

Thanks to all of you for the help you are going to give!

 

[itsmoked]

1) I'd rather my pumps run more continuously as starting up is probably the most stressing operation to everything including seals, bearings, motor windings, motor starters, cable connections, etc.

I'd try to work things (slow the emptying/filling process) so one pump almost runs all the time with others running occasionally to 'keep up'. At night with minimum usage the all-day pump would shut down and your sequencing would start a different pump for the next day's festivities.
N+1 meaning you have just enough pumps to run the system plus one so one can be out for service.

2) 150 feet? I'd want a heck of a lot more than 74psi. Perhaps 100psi. What's needed for the fire sprinklers on the 15th floor?

3) OK you get it. Have you considered splitting the system? Run one pump for, say, the first 80Ft (74PSI) and a different one for the remaining floors.





Keith Cress
kcress -

http://www.flaminsystems.com

 

[bimr]

For the domestic water supply, your building will probably require 2 pressure zones. Plumbing codes usually limit the high water pressure to 80 psi. Codes often limit the low water pressure to 20 psi, unless there are fixtures such as flush valves that require greater pressures.

I would suggest you review the typical design requirements for high-rise buildings before you begin design:

High-rise Water Distribution

High-rise Plumbing

 

[Mech110]

Thanks a lot for both replies. However, I would like to point out that at the moment I am just talking about the connection between ugwt and ohwt; the pumps supplying water to ohwt and the pipe carrying it. So in light of this,

@itsmoked: 1. how can i slow down the things? As i told you that 700 gpm is the water supply demand of the building calculated according to water fixture units (Uniform Plumbing Code). The only way I could think is to increase size of the ohwt which is not possible. So what do you suggest in this case?

2. This pump is not for supplying sprinklers. Sprinklers and standpipes (combined system) would be supplied by separate fire pumps. Therefore 150/2.31 gives me 65 psi head while adding friction losses in 8in pipe, I think 74 psi or max. 80 psi could be enough? Still disagree?

3. I would split the system if i were using booster pumps. I am supplying water to fixtures through gravity fro over head tank.

@bimr: Thanks for your input. However, please note that Im not talking about booster pumps to supply water. The scheme is to use a normal pump to feed overhead tank and then use gravity fed pipes to supply water to the fixtures.

 

[LittleInch]

Mech 110,

I think looking at this that you've not got or not telling us all the information.

Domestic water consumption is a highly irregular flow with large flows at certain times of the day ( morning, early evening) during the week and a completely different set of flows at the weekend or hot weather and very low flows at night.

Hence why you have a decent sized tank to even out the flow between one and the other.

It is far from clear to me if you've thought about this or what the 700 GPM represents - peak flow allowing for diversity or an average flow. Your numbers appear to think it's an average 700gpm over a 24 hour period - hence about 1 million Gallons a day. At an average of 100 gallons/day that's 10,000 people?? That's a big apartment building.

Hence many system use a variable flow system, either by a control valve or by variable sized pumps, e.g. run one at say 300gpm first and only if you get to a low level - say 30% full, would a second pump take over at say 600 gpm until you got back to say 80% full then revert to your smaller pump.

As many others say here, you're much better running a pump continuously for life improvement purposes and smaller cables / switchgear etc so better to have a smaller unit running for longer periods than starting and stopping a pump which is too big ( yr 1250 gpm unit)and in reality wouldn't be running every 6 hours but more like every 12 to 14.

74 psi looks a little low, but depends on your frictional losses. Static lift alone is 66 psig...

Fire pumps will need to be designed to NFPA / UL requirements and may need diesel back up pumps. Ditto pressure regulating valves are not your simple units, but need to be high reliability units and only certain ones can be used.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[Mech110]

@LittleInch

Thanks for your detailed reply brother.

Domestic water supply, ok tell me then what does this 700 GPM mean according to Uniform Plumbing Code? I followed this approach:

I calculated number & type of private and public fixtures and then using the chart given in UPC CHART A 103.1(1), it gives me 650 gpm. I added 50 gpm approx. for lawn sprinklers (will later calculate its exact value though). Building is residential+commercial. If we just talk about residential, my total fixture units are approx. 5500 equivalent to 600 gpm approx. (864,000 gal. per day) However, when I calculate number of people assuming 2 persons per bedroom, I get 926 people. Assuming 100 gal. per person per day, we get 92,600 gal. So why the difference? What I am missing?

About pumps, I got your point. Pumps should be running for long hours. So I am considering to user booster pumps instead of using overhead tank.

 

[LittleInch]

I think your replies to the associated post (

https://www.eng-tips.com/viewthread.cfm?qid=441842

) if anyone's interested, answer your queries much better than me.

Pumped water supply issues for domestic supplies are not easy to get right which is why tanks to buffer between outflow variations and pump rates are commonly used.

The trick is to balance the smallest tank you can get away with without running out of water or loosing too much pressure - both of them big complaints when it happens.

I don't normally do this sort of stuff, but I would go for a tank if you can as it makes life so much easier.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[Mech110]

Well the associated post has turned into another discussion "the wide variation between results obtained using different approaches of calculating water supply demand; UPC, IPC, Average Daily Consumption, ASHRAE Modified Hunter's Curves, Water Demand Calculator by IAPMO published in UPC 2018. Check out that and suggest anything!

 

[LittleInch]

Well that's why you need to design in flexibility and allow for a large flow variance.

Ultimately the consumers just want water on demand at a certain pressure and if they don't get they will bitch and moan.

This isn't the first high rise building so maybe you need to do a bit of poking around in your company as well as on this website to be able to get to a design you're comfortable is the right balance between user requirement, cost and complexity.

No one said it was easy.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.