Gravity vs Pump recirculating loops - which is more energy efficient?

Hi,

I've been reading up on recirc loops and am trying to understand which is more efficient for a residential use. Given a setup in which there is a dedicated hot-water return line, I realize that a gravity loop could potentially be used. My question is the following. From what I understand, a gravity loop is constantly working. This would imply that the hot water is constantly being pushed to the fixture, and the less hot water is being pushed back to the tank. Eventally the flow slows as the return line becomes saturated with hotter water, but once it starts to cool, the flow picks up again.

This circulation would imply that new water is constantly being pumped into the HW tank to be reheated; an obvious loss of energy.

So the question becomes, is it more efficent to have a gravity loop that is always running, or better to have a pump on a timer that only circulates hot water during "important" times (ex: before getting up in the morning, etc). To be sure that hot water is never far away, potentially run the pump for a min every 15 o 30 mins during daytime hours, etc.

Or is it more efficent just putting in a solenoid valve on the hot water return line that is closed at night, etc?

On a second note, is it possible to have a multi-branch gravity loop? ie: 2 or 3 different return lines from different branches in the house? Is there anything special (ex: check valves, etc) that needs to be done to ensure that one return loop doesn't interefere with the other?

Thanks,

Eric

IMO a gravity system must use less energy.

The amount of energy used to heat your water depends on your hot water demand, independent of whether you continually raise cool water a small amount or cold water a large amount.

It is more a question of which is easier to install. NOT ALL installations are candidates for a gravity system, because it has very stringent requirements. A pump will work with any system. A gravity system will work continually, but at a slower pace than a pumped one. For multiple returns you need a balancing cock and check valve on each one before going to the pump.

There is naturally an energy trade off for the convenience. You do save on water by not having to run water down the drain waiting to get hot. We pay almost $9 per hcf, so the water savings is measurable.

If you have a timer which accurately measures your demand, that would be ideal. Most likely, you will find yourself inconvenienced at times.

All in all, it is not a topic to lose a lot of sleep over, one way or the other!

After making a boatload of assumptions the convective water flow in a gravity system would be about 14 GPH compared with 3 or 4 GPM when using a pump. So, the water wouldn't get as hot as quick. This may not be a problem, though.

http://www.lawlervalve.com/content/Ron George Recirc March 2000.pdf

I recently added a gravity recirculation loop to our water system (Single loop) and it works great.

Some things that come to mind about gravity versus pumps is that if you have a gravity system you have to leave a section of the pipe uninsulated (about 10-15 feet). With a pump system you should be able to insulate the entire loop. The gravity system needs to have this cooling section in order for the water to circulate.

Our hot water heater is located in our unfinished basement so we have a good elevation difference for the gravity system. There is a check valve that I drilled ~1/8" hole in the flapper to allow minimal flow.

hj said:

For multiple returns you need a balancing cock and check valve on each one before going to the pump.

Click to expand...

Thanks for the tip, but can you please elaborate, or point me to docs that explain this better? Why the need for a balancing/check valve on each one? If I were to place a recirc pump on the primary return line into the tank, "pulling" water from the individual return lines, would that not work effectively? I realize that one line may end up circulating more water than another, but apart from wasted energy, does it really matter or affect the functionality of the system?

Or might the pump "pull" water only from the line with the least resistance, thereby not circulating water in the other at all?

Tx,

Eric