How Do You Determine Pumping Power for a Homemade Water Chiller?

  • Thread starter Thread starter tectactoe
  • Start date Start date
  • Tags Tags
    Power
Join the discussion
Registration is free. Start your own thread to ask a follow-up.
3 replies · 3K views
tectactoe
Messages
38
Reaction score
0
I have a heat pump water heater that cycles the refrigerant in reverse, thus using the cooling portion of the evaporater to chill the water rather than to heat it (basically, a 50-gal storage homemade chiller).

The problem is, I need to circulate the water throughout the tank, or else all of the chilled water will sit near the bottom.

There's an opening/valve near the bottom of the tank (normally used for space heating I believe, not sure) and the tank inlet isn't being used - so there's my recirc path. But the path will contain some bends and twists since the valve is at the bottom pointed one way and the inlet is on the top pointed another way. Also, the pipe will probably need to reroute a little to accomidate the place where I'll screw in the pump to the outside of the tank.

So that's my question - I'm going to use a pump to recirc the water. But how powerful of a pump do I need? I know it has to overcome the head in the pipe - but how exactly do you figure that out?

I'll be using copper pipe with a given diameter and I'll be able to measure the length/height of the set up... I know bends and vertical lifts contribute to loss, but how, exactly? I don't want to get a pump that won't do the duty, but I don't want overkill either.

Thanks, guys and gals.
 
Physics news on Phys.org
I wouldn't think you'd need much at all, maybe a little aquarium pump or something. Calculating the head is as easy as P=(rho)*(g)*(h), so that gives you your minimum theoretical head needed not taking into account piping losses.
 
Yeah, I know the pump will be a very small one... I'm actually more worried about getting a pump that's way too big, (cost issues, etc). so if I can calc pressure loss from (rho*g*h), how do piping losses factor in? Looking at the tank, I'm probably going to need about four 90 degree elbows in there. Just as well, does the copper pipe friction supply any kind of reasonable loss?

I know these elbows and such may not provide losses significant enough in my specific case, but I'm to the point now where I'm just downright curious and would like to know how to account for these other losses.

Thanks for your help
 
The loss depends on the velocity through the pipe, so if you have a big enough pipe, you won't need to worry about it. You can look up a pipe friction calculator online to size the pipe.