Thermosyphon Line size

[cityboy172]

I'm Installing a thermosyphon line around my pump that will open in the event of a power loss. How big does that line need to be? The boiler is an EKO 60, 2000 gallons of storage. Figuring that I have enough storage to safely act as my "dump" zone.

It's piped in a way that It should roll into storage pretty good. Boiler to storage loop is 1 1/2", thinking of downsizing for the bypass to reduce the cost of valve, just not sure how far. Would 3/4 be enough? Looking at my fitting pile, I have a 1 1/2 x 1/2, but that seems a little small.

 

[ewdudley]

There's a lot of variables, makes it tough to throw out meaningful 'internet advice'. However the two main variables you need to worry about are how high is the top of storage above the top of the boiler; and what is the pipe diameter.

Best way to approach it is to go ahead and compute the head generated by the weight of the return water column vs. the weight of the supply water column, and then compute the gpm according to the equivalent feet of pipe of some particular diameter, same as any hydronic circuit.

Failing that, just go ahead and make your thermo-siphon 'bypass' circuit a 1.5" straight-shot, and put the pump in a parallel circuit with all the elbows. Even so you'd probably want 1.0" for the pumped section at the very least.

I never could find a suitable off-the-shelf low-resistance check-valve for the thermo-siphon back-flow preventer. All the loading units use a nice flapper-valve that offers very little resistance to forward flow, but I don't know where you can get a comparable standalone check-valve.

I ended up using a conventional swing-check, but I removed the bronze swing plate and made a replacement out of HDPE. Had to insert a couple pieces of copper wire into the plastic swing plate so it would sink into place.

(How does return temperature protection fit into you scheme?)

Bottom line: those loading units sound pretty expensive, but when you realize all the components they provide, and all the problems they make go away, and in such a compact solution, it's hard to come up with a better alternative (in the specific case of using thermo-siphon to storage for power-fail heat dump).

 

[Floydian]

https://www.hearth.com/talk/threads/thermo-siphon.100396/

Here's some great info ewd provided if you really want to know what your system can thermosiphon.

Brace yourself.

And +1 on the loading unit. It's a very nice all in one solution with temp gauges, isolation ball valves, mixing
Valve, 3 speed pump, etc. At some point the ECM pump versions should start showing up here.

Not sure LK Acaso, like I have, makes a LU sized for a 60Kw boiler but I'm pretty sure Laddomat(sp?) does.

Good luck,


Noah

 

[ewdudley]

Here's a photo of my setup when I was putting it together, which doesn't use a loading unit. It uses a boiler recirculation loop with return water injection controlled by a PID controller. The pumps were all salvage parts from friend and the PID controller was inexpensive, but even so with all the valves and fittings it all added up, plus it was a lot of fussy work to get everything within 1/16" or so in order to get it all to fit up. And the check-valve on the bottom right is the modified swing check described above, which took a lot of R&D.

As you can see it's a long way to go to save some cash, but it can be done and it does perform nicely.

To do it again I would not bother with the 'Boost Return Inject' pump. I just set the return temperature high enough so that the supply temperature is high enough on the first time through storage, which means the return temperature from storage is never high enough to require the second pump.

The boiler is elevated about 20", which is something I would definitely do again. Although it was pretty much necessary to make room for the plumbing, I really like having the boiler up high for loading, tending, and cleaning.

 

[cityboy172]

Do I want to bypass my mixing valve also? Or does that concern go out the window in this scenario? (assuming I do, not sure)

 

[ewdudley]

Do I want to bypass my mixing valve also? Or does that concern go out the window in this scenario? (assuming I do, not sure)

I think you would have to since the mixing valve would close off the port coming from storage if the return temperature from storage was below the setpoint of the mixing valve.

You need to pretty much replicate the circuitry of a loading unit. Here's a drawing of a Laddomat. The thermo-siphon circuit, "stage 4" below, is a straight-shot that bypasses the mixing valve and pump:

 

[cityboy172]

Just looked over a 1 1/2" actuated valve I had. I had over written it off because it was norrmally closed, but looked again and converted it over to normally open (spring return). So, I'll bypass my mixing valve and pump ( pump has check valve in it to prevent flow through boiler when boilers off).

Thanks for the help guys.

 

[ewdudley]

Just looked over a 1 1/2" actuated valve I had. I had over written it off because it was norrmally closed, but looked again and converted it over to normally open (spring return).

Nice find. It's great to have a situation where you can take advantage of gravity flow to meet fail-safe design goals. I'm sure the folks at Fukushima wish they had.