To pump or to use gravity head- which is best and most energy efficient? 6

[ahika]

Hi All,

I am working with a local council in an area of New Zealand that is very flat, so everything needs to pumped for reticulated water to houses.

In this system, water is transferred (using pumps) from a bore and stored in a main reservoir (3,500cu m)where it is treated. From the main reservoir, the water is pumped to a small town reservoir which is located in a tower with approx 38m head. The town reservoir holds 385cu m. We are only talking about a small town in rural NZ with a population less than 2,300 people.

The pump station is due for an upgrade and one option on the table is to pump directly into the reticulation loop and remove the storage reservoirs. Effectively the pumps would provide all the pressure (approx 420kPa) for the town.

My questions are these:

1. Will energy use decrease/increase with a direct feed into the reticulation loop and why?
2. One disadvantage of a direct feed relates to the lack of capacity if there was a power outage,earthquake, disaster etc. What are the advantages?
3. Any other consideration greatly appreciated.

 

[BigInch]

You must study your demand usage to understand how energy use will be affected. Off hand I would say you could do either for nearly about the same energy usage, except for the lift to 38 meters.

Firewater pressure and reserve supply should not depend on pumps, unless you have appropriate reliability such as would be provided by 2 diesel driven pumps and one electric.





"People will work for you with blood and sweat and tears if they work for what they believe in......" - Simon Sinek

 

[Artisi]

Without getting too specific or academic, if the daily demand is xxxx litres this has to be pumped to 38m head over a day / night into the reservoir with a reasonably fixed flow / head duty, also if you have off-peak electricity rates available pumping could be done during these periods. If pumping directly into the mains and maintaining a pressure of around 420kPa whether you have max or min demand at the time looks like a problem - how will you control pump flow / pressure for min. demand and flow / pressure and for max. demand. It can be done but not easily and I wouldn't think cheaply.

To me, seems pumping into the reservoir looks like the most cost effective and easiest.

It is a capital mistake to theorise before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts. (Sherlock Holmes - A Scandal in Bohemia.)

 

[bimr]

1. Energy usage would be approximately the same for either a gravity system or a pressure system. The energy used will be proportional to the water pumped at the pumped water pressure. At night the energy used will be minimal for either system. Most gravity systems pump during the day when maintenance workers are available.

2. For either system, you should have a standby generator available in case of power failure.

The disadvantage of the pumped system is that you will lose about 10% of the storage capacity. However, with the limited elevated storage capacity that looks to be approximately 3-4 hours, there is limited value in having the elevated tank. One would expect that the working volume in the elevated tank is only 1-2 hours.

An advantage of the pumped system is that there is no maintenance on the storage tank. If major maintenance is needed on the elevated tank, it would make sense to eliminate the elevated tank.

A disadvantage of the pumped system is that the control system may be a little more complex but the control system cost should not be prohibitive. Because the elevated tank has such a small volue, the control system may be the same for either pumped or gravity flow.

The booster pumping system would be similar for either the pumped or gravity systems. The pumps should definitely be supplied with VFD's for the pumped system.

3. It would not be a good idea to eliminate both storage tanks if that is what you are considering.

The ground storage tank has adequate capacity for fire water storage if that is an issue.

If the water well fails, it will take at least a week to repair the well. The ground storage tank has about a weeks volume and the elevated tank has about a days volume.

A system without water storage will operate fine as long as the well is running. It the well pump breaks down, the water system will shut down immediately.

It does not seem practical to pump from the well without some storage between the well pump and booster pumps. With a small water supply system, you may be looking at peak flows of 7-10 times the average flow. Your well system may not be able to produce the peak flow capacity required without storage.

 

[BigInch]

Alternatively, and considering today's energy costs, most gravity systems pump during the night when electric costs are minimum.

"People will work for you with blood and sweat and tears if they work for what they believe in......" - Simon Sinek

 

[77JQX]

I work for a water district that has about 100 pump stations, and uses both approaches (1) pumping to elevated storage, and (2) constant pumping in a closed loop system. While in theory both ways of pumping do the same job (same quantity of water delived at the same pressure), in practice, the pumping to a tank is much more efficient.

With a tank, you can operate a pump in its sweet spot, and as demands vary throught the season, you adapt by changing how many hours per day the pump operates. Pumping to a tank typically has an efficiency of 75% or greater. Plus, since the fire storage is in the tank, we only need a 2 pump station to have redundancy.

With constant pumping, you need to operare a pump 24/7, regardless of the demand (the 2,300 household demands are too large for an effective hydropneumatic system). So you'll need to specify pumps that meet the full range of demands. For our system, the ratio of summertime peak demand to wintertime minimum demand is about 20:1. Add in fire flow, and the pumping demand ratio gets to 200:1 (or larger). So how many pumps are you planning to use? We use 4 at most - including two redundant fire pumps and a backup genset. With that range of flows, and that few pumps, there's no way to keep a pump station operating efficiently (VFD or not). Our typical constant pumping zone efficiencies are less than 20% on an annnual average.

There are a few advantages to constant pumping. Lower capital costs. Storing water can give rise to quality concerns - disinfectant residual and disinfection byproduct problems are smaller without storage tanks. Sometimes, there just isn't a good place to put a tank.

Without the tank, your new pump station just got more expensive due to fire flow, and your operations just got more expensive due to poor efficiency. I'll leave it to you to do the economic comparison between tank maintenance and pump station costs. But unless you have serious water quality concern with your tank, I bet you're better off keeping it.

 

[cvg]

agree with biginch, in my experience with several small and large water districts, the goal was for replenishment of tanks at night to save on energy costs and pumping during the day only if necessary. These were gravity fed with the tanks "floating" on the system. It was only necessary to pump in the day time if there was insufficient storage in tanks to last all day or insufficient pumping capacity to fully replenish storage tanks during the night. No VFD's were required and pumps only needed to be sized for replenishment flow rates, not for peaks.

 

[ahika]

Hi All,

I thank you for your thoughts.

77JQX- I really get your points regarding the different flow requirements meaning the potential for oversized pumps.

Elevated storage is effectively a battery providing a 'grace period' should issues occur esp as power outages are common.


Thanks again

 

[BigInch]

77JQX Excellent advice.

"People will work for you with blood and sweat and tears if they work for what they believe in......" - Simon Sinek

 

[Artisi]

Yep, got a star from me.

It is a capital mistake to theorise before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts. (Sherlock Holmes - A Scandal in Bohemia.)

 

[BigInch]

Me 2 "summertime peak demand to wintertime minimum demand is about 20:1. Add in fire flow, and the pumping demand ratio gets to 200:1 (or larger). So how many pumps are you planning to use? We use 4 at most - including two redundant fire pumps and a backup genset. With that range of flows, and that few pumps, there's no way to keep a pump station operating efficiently (VFD or not)"

Without any storage you'd need a whole lot of different sized pumps. With 3 or more pumps, you can make almost any flowrate you need to, do it relatively efficiently and without a VFD.

"People will work for you with blood and sweat and tears if they work for what they believe in......" - Simon Sinek

 

[bimr]

Sorry, but I have to disagree with 77JQX comments as they are general and not appropriate for this particular project. Appropriate for a large system with numerous pumps station, yes.

77JQX comments are appropriate if there is a substantial volume of water in the elevated tank. In this particular case, there is minimal water in the elevated tank. 90% of the water storage is in the ground water storage tank. There will be a limited working volume in the elevated storage tank because the elevated tank storage volume is small. In addition, there is inadequate fire water storage (US volumes) in the elevated tank (unlike the systems that 77JQX is familiar with). The fire water storage has to be in the ground storage tank.

There is minimal water in the elevated tank and this is a small system. With a small system, the peak flow/average flow ratio is much higher. Because of these factors, in this particular case, the pump system for the elevated tank is probably going to have the same pump capacity as that for a pressure pumping system.

Another point to consider is that the water mains for this particular small installation are probably small diameter and will not allow adquate water flow with just gravity pushing the water.

A hydropneumatic tank is also not needed for a pressure system. This is commonly accomplished with 3 pumps working on VFD's. At night, there is minimal flow, so there is almost zero power usage.

I know some of you do not like VFD's. And I am not promoting VFD's. However, the VFD is the simple method to maintain pressure in a pumped system. It is much more complicated to try to do this with multiple pumps with different capacities while attempting to match variable discharge flows.

I have installed pressure systems like this that are meeting US fire flow requirements. However, for this particular case, I believe it will be a much easier design because I would not expect to be required to supply the larger fire flows typical in the US.

 

[zdas04]

bimr,
I would use every one of your arguments against 77JQX as points in support of his basic theses. It is a smallish system? Good reason to reduce complexity. I live in a community of 40,000 people that supplies water from a reservoir using pump discharge and every time the fire department hooks up to the system I don't have water flow to the second floor bathroom. This is a larger system than the OP is talking about, with considerably more resources available and they can't successfully deal with that 200:1 turndown ratio. Fewer resources demands less complexity.

Even if the tank is just a buffer with less than a day's capacity, the tank is far better at regulating a variable demand than you could do with pumps and VFD's. You can size your pumps to be in the sweet spot of the operating curve all the time and simply turn them on and off to meet outflow requirements.

As I read through this thread, all of the points I wanted to make were in 77JQX's post so I gave him a star too instead of repeating all of his points.

David Simpson, PE
MuleShoe Engineering

"Belief" is the acceptance of an hypotheses in the absence of data.
"Prejudice" is having an opinion not supported by the preponderance of the data.
"Knowledge" is only found through the accumulation and analysis of data.

 

[bimr]

zdas04

You are emphasing one of my points. You comment:

"I live in a community of 40,000 people that supplies water from a reservoir using pump discharge and every time the fire department hooks up to the system I don't have water flow to the second floor bathroom"

It would appear that your commmunity water system can not supply the peak flow because the pumps have inadequate capacity at peak flow. Municipal systems should be able to maintain a minimum pressure of at least 20 psig at all points in the system during fire flow.

Making something simple is no solution if the solution does not work. Suggest you make a complaint to the public works department as that is an unsafe situation.

 

[ahika]

Hi All,

To expand further, this project has about 200 pump stations throughout the region so 77JQX's comments are relevant.

Another option already mentioned is to remove the small town reservoir and pump into the reticulation from the main reservoir. But you still face the same issues already discussed.

But discussions here have identified some good points:

Constant pumping system
- The flow rate will never be constant making it very difficult to specify pumps. The system would probably need to be multistaged with speed drives but there will be times when the pump selection is not ideal.
- Pump efficiencies can be as low especially if the pumps are over-sized
- Summer peak flow, winter peak flow and fire fighting flows need to be carefully considered when specifying the correct pumps.
- Lower capital cost especially is the reservoir is due for a serious upgrade
- System controls can be difficult
- Requires a backup generator system for fire fighting during power failures
- No maintenance of a reservoir

Elevated storage system
- Fire fighting flow is not an issue as the tank is usually sized accordingly
- A pump can be specified for constant supply and achieve good pump efficiencies
- As demand for water increases the pump operates for longer hours during the day to meet demand but the pump size remains the same.
- Two pumps required for redundancy- may require four pumps for redundancy in a constant pumped system
- Elevated storage is effectively a battery and can be used during power outages, natural disasters or pump/equipment failure.
- Expensive to maintain
- Lowest operating costs

 

[bimr]

Your first post describes this as a standalone small village with a single well. Now you say that there are 200 pump stations. Are all these systems interconnected?

You should note that the water system is a system consisting of raw water, piping, pumps, tanks, etc. It is not possible to come to any firm conclusions without knowing details of the entire system. For example, how would one know the flow capacity that you can get out of the system without knowing the piping size. Same for regulatory requirements such as fire flow capacity. Same for age of the system components.

Regarding fire flow, the typical minimum fire flow requirement for NZ is 150 l/sec for 3 hours. That is 1,620 cubic meters. So your elevated tank does not have adequate capacity. Elevated tanks usually hold fire flow + daily working volume + reserve.

http://www.fire.org.nz/...Fire.../da516e706c1bc49d4440cc1e83f09964.pdf

Therefore, the statement "Fire fighting flow is not an issue as the tank is usually sized accordingly" seems to be questionable.

It also makes the statement "A pump can be specified for constant supply and achieve good pump efficiencies " questionable. To meet the fire flow, your system inlet pumps would have to be sized for 150 l/s + the peak summer hourly flow, not the average daily demand.

One final comment is that you should note that municipal customers get preferential treatment from the power companies because they are large power uses and franchisers. As a result, the municipalities pay less than industrial users for power. So efficiency is important, but not an overiding concern. It is more important that the system is designed correctly.


If your elevated tank has adequate volume, you still have to have installed pipes with enough capacity to transfer the water.

 

[ahika]

Hey Bimr,

I wish you statement regarding power pricing was true in my country.

Here, industry (which i define as users with a load greater than 5MW continuous) pay the least for their energy. We have a system for industrial users called time-of-use (TOU) pricing where the price changes every 30minutes. Over the long term TOU pricing is the best option for insudrial users but the price can fluctuate. In NZ, 70% of our electricity is produced from renewable sources (mostly hydro and geothermal w some wind in the mix). During a "dry-year" the price per kWh can go up to $0.50/kWh- which equates to millions of $ per month. Our last dry year was in 2008.

However, in a wet year it is not uncommon to get pricing of $0.005/kWh- thats half a cent per kWh!

From my perspective,Im in the energy efficiency industry, I want the price to be high! When the price is high I'm busy.


Regarding your other statements- none of the systems are connected but I'm guessing 77JQX is the same.
I didnt want to go into the detail and I think I provided enough information as everyone here has "hit the nail on the head", including yourself!

I just hope I can give back to this forum with some useful feedback-


Cheers

 

[bimr]

Regarding: To pump or to use gravity head- which is best and most energy efficient?

The point that I was trying to make was also identified in the attached paper: HYDRAULIC DESIGN OF WATER DISTRIBUTION STORAGE TANKS

"The costs, however, are highly site-specific and the comparison must be made on a case-by-case basis.'

http://xa.yimg.com/kq/groups/21948400/443004144/name/Water_Storage_Tank_Hydraulic_Design.pdf