Feeding a tee through both runs in parallel circuits

dave123

If two piping circuits are in parallel, and pumped independently (only one circuit running at time), is there a reason not to join the circuits through the runs of a tee, with flow going out the branch? In other words, flow from one circuit at a time would enter each separate run of the tee, and exit through the branch. Would this function any differently from an elbow for the circuit that was running? This is assuming a check valve is present in each run to prevent flow through the unused run of the tee.

The unused run of the tee would just have static pressure, and would function like the back wall of an elbow. Is this correct?

Thanks.

Jamie Hall

More or less. There is slightly more headloss that way than there would be with a smooth elbow, due to the turbulence at the turn.

dave123

But it's not considered poor practice for some reason? I've been looking at lots of pics of installed systems, and have never seen it done this way. It seems like the second circuit always leaves and enters via the branches of the tees.

mattmia2

Do you mean primary secondary piping where one of the takeoffs from the primary is at a "corner"?

Jamie Hall

From the standpoint of the pure hydraulics of the thing, in single phase systems (such as hot water heating), the loss through a T between the two runs and the branch varies depending on the flow directions. Straight through the run has, as you would expect, the least loss. In one run and out the other and out the branch, the branch has considerably greater loss; in fact, greater than an elbow in that location would have. In through the branch and out both runs, about the same as the corresponding elbows would have. In both runs and out the branch, somewhat more loss than the corresponding elbows.

In certain applications you want to come either in the branch and out both runs, or in both runs and out the branch: specifically where you have two circuits fed from, or feeding, one, and you want to have the same or at least very similar head loss in both circuits,

Steve Minnich

In what type of system would both pumps never be running at the same time?

dave123

Steve Minnich said:

In what type of system would both pumps never be running at the same time?

I'm referring to a boiler piped primarily to a buffer tank, with a parallel run to an indirect tank. Each circuit with it's own pump. and only one pump runs at a time for domestic hot water priority.

hot_rod

That is certainly workable. I have done that where there is either an indirect call, or a heating call. Never two at the same time.
Basically that is how a 3 way zone valve flows.
Here is an example of how I piped a few.

Steve Minnich

Interesting. I thought of DHW priority which is time limited, but not that. 

hot_rod

They are not so easy to source but Nibco still makes double ells in a few sizes. That would be a nice fitting for the return in that schematic.
They sent me a 3/4 x 3/4 x 1"sample.

Jamie Hall

Nice. Those would have a loss very close to a simple elbow going either way (but ot through the run!!!)

hot_rod

Jamie Hall said:

Nice. Those would have a loss very close to a simple elbow going either way (but ot through the run!!!)

Exactly, these are ells, not a substitute for a tee

dave123

hot_rod said:

That is certainly workable. I have done that where there is either an indirect call, or a heating call. Never two at the same time.
Basically that is how a 3 way zone valve flows.
Here is an example of how I piped a few.

But wouldn't a plain tee with check valves blocking the unwanted flow work as well? Or would you worry that the check valves wouldn't last as long, and then flow would end up in the wrong circuit?

hot_rod

dave123 said:

hot_rod said:

That is certainly workable. I have done that where there is either an indirect call, or a heating call. Never two at the same time.
Basically that is how a 3 way zone valve flows.
Here is an example of how I piped a few.

But wouldn't a plain tee with check valves blocking the unwanted flow work as well? Or would you worry that the check valves wouldn't last as long, and then flow would end up in the wrong circuit?

Yes a tee will work fine, they are used in plumbing and heating all the time. The flow path is not as friendly as @Jamie Hall mentioned in the "turn" direction.
The EL charts show this.
Whereas a double ell would be the same EL as an ell in either flow direction.

What you do want to avoid, and I believe the code forbids is a bullhead where flow comes or goes from both run direction into or out of the branch. You may, will, get odd flows from that type of use. Sometimes they are equal flow from both directions, the next time one is flowing at a different rate, very unpredictable.

STEVEusaPA

hot_rod said:

That is certainly workable. I have done that where there is either an indirect call, or a heating call. Never two at the same time.
Basically that is how a 3 way zone valve flows.
Here is an example of how I piped a few.

Just curious. Why would you need the 3 way valve, especially if only one zone could run at a time?

EBEBRATT-Ed

If only one pump runs at a time it will be fine

hot_rod

STEVEusaPA said:

hot_rod said:

That is certainly workable. I have done that where there is either an indirect call, or a heating call. Never two at the same time.
Basically that is how a 3 way zone valve flows.
Here is an example of how I piped a few.

Just curious. Why would you need the 3 way valve, especially if only one zone could run at a time?

You're right, it could be done with a single circ.
I feel ZVs do give a better shutoff compared to trusting integral checks.