Res. Re-Circ Sizing

Hilly

I can't seem to quickly find the information I am looking for. I am curious about sizing a return lines for a domestic re-circ line. It is about 75-100' of 3/4 pex and then likely 75' on the 'return' portion. I am just curious about using 1/2 vs 3/4. I know I used to have charts and tables but I am out and about today and don't want to download a hundred pdf's looking for the right charts.

hot_rod

1/2 or in some cases 3/8 works.

Of course the correct way is to run the numbers.

The size required is related to the gpm needed.

And the gpm needed is related to how much heatloss in the loop.

Insulated loops require much less gpm, and hence pipe size.

Without insulation a DHW record loop is basically a hydronic heat loop

Hilly

Does sizing take different consideration when taking into account that new 'res' circs have limited runtime? With those it seems to me that you really just need to overcome the headloss. I've always used the small res grundfos, always with 1/2 pex returns and have yet to have an issue but something has me wanting to do the numbers. I'm probably just overthinking it here today.

JUGHNE

I have gravity recir with no pump at all. Maybe 60' 3/4" out and 60' 1/2" return. All horizontal runs with minimal slope down on return only. Minimal insulation on both....3/8-1/2 foam sleeves.

The hot water is there right now......no pump.....no KW.

It seems that if the water moves at all you will have pretty quick hot as you want.

Not recommending this, but for example, just a good drip on a bathroom faucet will keep your line & water hot. Try that experiment and you can see how little flow is needed. More insulation helps of course.

hot_rod

Hilly said:

Does sizing take different consideration when taking into account that new 'res' circs have limited runtime? With those it seems to me that you really just need to overcome the headloss. I've always used the small res grundfos, always with 1/2 pex returns and have yet to have an issue but something has me wanting to do the numbers. I'm probably just overthinking it here today.

By limited run time do you mean a timer or thermostat control? If so, same sizing. Some of the new smart systems "learn" usage patters, but I suspect you still need to move enough gpm to keep that delta t around 10° between leaving tank and return.

Sure a gravity thermosiphon works, I think the tank needs to be below then piping for that to work?

JUGHNE

Yes, I have a basement. Recir is piped into the bottom of tank with swing check valve (an 1/8" hole may be drilled into the flapper if flow is hard to maintain....not in this case) and ball/flow control. Hard pipe for the return helps maintain the pitch of pipe returning. I used up some 3/8 ID type L for almost 1/2 the run, the remaining is 1/2" ID L.

I saved $100's on a pump, another chunk of cash for aquastat, and some on KW. No future pump replacement.
Yes it is a perpetual heating loop but I feel the ROI on the hardware/KW savings will be a wash.....hard to say.
Indirect tank with Mod Con.

hot_rod

Our friend PAH, Dave Yates has written a few articles on pummels DHW recirc. maybe for Fine Home Building.

I wonder if insulating the entire circuit well would stop or slow the thermosiphon. Once the delta T is gone, in theory the flow would stop?

JUGHNE

That was my thought also. I purposely left about 6' of the vertical drop un-insulated at the WH. I think of it as a cooling leg. It is in the cooler basement boiler room. In the summer it is of course a heat gain. Other times it is added heat within the conditioned space.

Hilly

Piping like the diagram dave shows then you could quite simply cut in a circulator without having any extra costs. I could essentially pipe it as so and call it 'circ ready'

hot_rod

Gosh the new, tiny ECM recircs probably only draw 20- 30W. I wonder that electric meters can even read that low of a draw?

The pump doesn't need to run if the loop is warm.

I think California now requires 1" insulation! Maybe the pump only runs a few hours a day when well insulated in a conditioned space?

The pump gives you better control and quick response, IF a gravity cannot or does not work.

JUGHNE

Yes, that is a selling point for the system....don't work....add the pump later.
"Selling point"....not much to sell here is there? Just pipe.
I think I would spring for 1/2" ID copper to avoid dips which could stop the thermo siphon action.
Here is the article from some years ago.

hot_rod

I see you have hoarded old, informative articles also. Remembering when the office was wall to wall file cabinets

Zman

Keep in mind you only need to recirc enough to make up for the heatloss of the pipe. Usually you can do 2 ft/sec or less. Pipe insulation is an absolute must. Oversizing DHW circs causes more copper pipe erosion than any other system.
Siggys software will run the math for smaller pipes. I recently looked at a commercial kitchen that was wasting $0.12 an hour with insulated pipes.

I have a need to calc a pipe heatloss for a building with 3",4" and 5" piping. Does anyone have a program to calc that? The formula is not that ugly, I could always plug it into a spread sheet.

JUGHNE

To gain head height for the gravity the recir line is bought up as high as possible under the master bath lav (farthest fixture).
Added a ball valve there for service.
This point is still below the HW faucet which lets the air bleed out.

Paper and clippings are still my method, but then I'm old.

hot_rod

Remember also recirculating through the tank will lessen any stratification and reduce useable output. So run the pump as little as possible.

On larger commercial tanks you may see a record port mid tank or up higher to prevent breaking up the stratification layering.

JUGHNE

I was thinking that constant recir flow would even out the temp in the tank????

In my case I have 120gal "Rheem Solar Air Head Tank" with electric element (disconnected) in upper spud. The lower spud has pump inlet line from the HXC, then the hot connection from HXC goes into a top tank inlet with about 24" of dip tube. DHW comes right off top of tank. Cold has a longer dip tube to bottom of tank.
Recirc line connected to lower tank drain.

The electric upper tstat controls the boiler and pump.

Hilly

If you have top inlet/outlet water heater and you connect the re-circ to the drain port would you still utilize both check valves? I imagine the answer is still yes.

hot_rod

I was thinking that constant recir flow would even out the temp in the tank????

If you are storing water for DHW or buffering a wood boiler maybe, then you want to stratify the tank as much as possible.

I've seen solar tanks set at 180° pop the T&P relief valves at 210. The control sensor is at the bottom of the tank, so it stacks higher temperatures at the top. Tall tanks will stack 30 maybe 40 degree differences.

Really when you add a constant DHW recirculation, you really have an open hydronic heating loop, in addition to DHW

The tank as you mentioned will blend if you circulate from top to bottom.


Commercial tanks and some of the HTP combo tanks have ports mid point for circulation or accessing the hottest section of the tank.

JUGHNE

Hilly, I assume the check valves you refer to are the ones installed in the factory nipples.
Is the cold side really be a check valve?? or the pop off could eventually open.
The hot side must be somewhat of a flow check to prevent migration of hot.

My tank is too old (1994) to have either.
I have a true check on the cold with expansion tank.

If the flow check? was left in the hot side then maybe it would be too much for the gravity recir to overcome??

FWIW, my summer gas usage averages 22.5 CF NG at about $1.47 per unit. DHW and cook top burners.
Maybe that is too much for 2 people.....but then I do not have my own Lear Jet

Larry Weingarten

Hello, Just to throw in a wrench, if a "demand system" is used for recirc (it has by far the lowest heat loss) then you'd want the return line to be 3/4" to allow good flow. Demand pumps mostly run less than five minutes a day.

Yours, Larry