Need for check valve in parallel circuits

dave123

I saw an article by Siegenthaler where he described his approach to piping an indirect tank in parallel with the main loop. The diagram is attached. I'm trying to understand the purpose of the check valve at the tee leading into the boiler return. The two flow checks higher up make sense, but what is the function of the check at the return's tee? In the article, he says it is there to "prevent heat migration into the return side of the primary loop."



I'm sure he's correct, but I don't understand that. The flow check higher up will prevent any mass flow, so does he mean to prevent "diffusion" of heat across non-flowing water? Or is this a case where water in one pipe might be flowing in two different directions?

If so though, why no need to put another check in the return of the indirect tank, leading into the same tee?

Thanks.

hot_rod

If the indirect is hotter than the boiler, the heat could migrate backwards. You can get two directional flow in a single pipe under the right conditions. Hot is always trying to get to cold, however it can.
Really no harm in checking S&R in any application, even primary secondary closely spaced tee connections.

Use spring type hydronic check for best seal.

dave123

But why no need for a second check, on the indirect's return to the tee? Presumably the primary loop is running most of the time, and the indirect is running much less frequently, so at least the pipe from the indirect to the tee will usually be colder than the primary loop. Wouldn't that potentially lose heat?

And what if the primary loop runs at 140F and the indirect is set at 120F? Won't that also lose heat?

I don't understand why both entrances to the return tee at the boiler don't need checks valves.

hot_rod

dave123 said:

But why no need for a second check, on the indirect's return to the tee? Presumably the primary loop is running most of the time, and the indirect is running much less frequently, so at least the pipe from the indirect to the tee will usually be colder than the primary loop. Wouldn't that potentially lose heat?

And what if the primary loop runs at 140F and the indirect is set at 120F? Won't that also lose heat?

I don't understand why both entrances to the return tee at the boiler don't need checks valves.

I think you are correct in wanting a check on the indirect also.

Must be an old drawing that it doesn't show circs with integral checks?

The Steam Whisperer

If the drops that connect the indirect return to the common return are deep enough ( 12 to 14 inches is about the minimum, IIRC) then thermal migration is stopped from the boiler loop to the tank, however, thermal migration can then happen from the hot tank to a cold boiler loop. The best solution is a loop of piping that comes out of the indirect, goes up 12 inches and then down 12 inches and then into the boiler loop. This nearly always stops migration in both directions. Sprung check valves alone often still do not stop thermal migration because water can circulate inside a single length of pipe, flowing down one side of a pipe and up the other side, sort of like one pipe gravity hot water systems or the air movement between interior windows and storm windows that are too far apart.

dave123

The Steam Whisperer said:

If the drops that connect the indirect return to the common return are deep enough ( 12 to 14 inches is about the minimum, IIRC) then thermal migration is stopped from the boiler loop to the tank, however, thermal migration can then happen from the hot tank to a cold boiler loop. The best solution is a loop of piping that comes out of the indirect, goes up 12 inches and then down 12 inches and then into the boiler loop. This nearly always stops migration in both directions. Sprung check valves alone often still do not stop thermal migration because water can circulate inside a single length of pipe, flowing down one side of a pipe and up the other side, sort of like one pipe gravity hot water systems or the air movement between interior windows and storm windows that are too far apart.

I understood all of that except the part about the swing check not necessarily stopping the ghost flow. Assuming the valve works correctly, how would anything get past it?

Zman

It does not take much to make a swing check open a bit. Thermal "buoyancy" can easily move a standard one. They do make weighted swing checks...
I like a design with just 1 check valve and heat traps as @The Steam Whisperer described.

The Steam Whisperer

dave123 said:

The Steam Whisperer said:

If the drops that connect the indirect return to the common return are deep enough ( 12 to 14 inches is about the minimum, IIRC) then thermal migration is stopped from the boiler loop to the tank, however, thermal migration can then happen from the hot tank to a cold boiler loop. The best solution is a loop of piping that comes out of the indirect, goes up 12 inches and then down 12 inches and then into the boiler loop. This nearly always stops migration in both directions. Sprung check valves alone often still do not stop thermal migration because water can circulate inside a single length of pipe, flowing down one side of a pipe and up the other side, sort of like one pipe gravity hot water systems or the air movement between interior windows and storm windows that are too far apart.

I understood all of that except the part about the swing check not necessarily stopping the ghost flow. Assuming the valve works correctly, how would anything get past it?

The water does not flow through the valve, but the water circulates around the inside of a single pipe... hot water on top and cold water on the bottom. It can drive even a good tech to the looney bin trying to figure out what is going on when they experience it for the first time.