DHW FPHX inline with injection mixing?

[varadhammo]

Here's an idea I've been thinking about. I'm planning on doing DHW with a FPHX + small SS circ "sidearm" to an electric tank water heater. The building is all radiant floor, planning on outdoor reset injection mixing with Tekmar 356. Injection legs pumped by UPS 15-58 off the GARN primary loop, through 100' 1-1/4" PEX to building. My original plan was to leave stubs on the supply and return risers for attaching a FPHX for domestic hot water, with it's own circ on the system side to pull from the GARN loop independently.

Recently though I've been toying with the idea of putting the FPHX directly in the injection supply line. It would eliminate a pump and simplify the controls. I'd size the FPHX to ensure a low pressure drop at the design injection flow rate (probably use a 40 plate). The theory is that when any zone (space heating or DHW) calls, the 356 powers up. DHW call would be managed by a tekmar 156 (on/off differential setpoint) with sensors in the WH and a thermowell in the GARN. If DHW calls, the water heater loop will run and the HX will pull heat off the supply injection leg causing the temperature in the system loop to drop. The 356 will try to compensate and ramp up the injection circ to meet its target (right up to full speed, if the HX is pulling out a lot of BTUs). If DHW calls and space heating isn't calling, the 356 will power up and will still see a low system temperature (because the piping in the system loop will be cool) and so will run the injection circ up to full speed through the HX and the P/S tees. No heat should migrate into the system loop (system circ is grundfos alpha w/ spring check).

A sketch is attached. Could you guys point out any problems/pitfalls with this idea? From what I've read, I haven't heard of anyone doing this... maybe for good reason!

Thanks,
Jake

 

[ewdudley]

(Are the pump symbols pointing the opposite direction of that intended?)

 

[varadhammo]

(Are the pump symbols pointing the opposite direction of that intended?)

I don't think so... are you referring to the injection pump pointing toward the GARN? I thought I'd seen it drawn this way, pumping the return leg rather than the supply. Is that wrong? Or are you referring to something else?

 

[ewdudley]

I don't think so... are you referring to the injection pump pointing toward the GARN? I thought I'd seen it drawn this way, pumping the return leg rather than the supply. Is that wrong? Or are you referring to something else?

Never mind, I was confused by the diagram and didn't even see that the garn was on the left.

Looks to me like it ought to work just as you describe. Your independent pump idea might be preferable if you're planning on supplying DHW in the summer. This could be done simply by placing a smaller FPHX in parallel with the injection loop line instead of in series as shown; with its own tiny pump and with a check valve in the parallel leg of the main line.

 

[Karl_northwind]

I would put the 1558 pumping from the garn to the FPHE, and not have a primary circulator on the Garn. up the size of the 1558 to maybe a taco 0015, and off you go. don't count on the 356 responding to the temp drop from the DHW FPHE rapidly, as it will be mixing with the heating circuit temps. I do love me a 356 though. I use them a lot on GARN installs. if you need to force the injection mixing circ to full speed, just use a RIBU1C relay (google) to disconnect the output from the 356 and connect the hot circ lead to line voltage. Presto: full speed for high load applications.

karl

 

[Karl_northwind]

you could also eliminate the circ on the DHW side, and use it and a flow switch as instantaneous heating, preheat for the dhw tank.

karl

 

[varadhammo]

Karl,

I thought about overriding the 356 with a relay like you suggested... Then how would I protect the system side from being hit with too high a temperature when the 15-58 is on full speed? Seems like I'd have to run dhw as priority only and disable the system Circ whenever the override was active.

I also thought about the preheat with flow switch arrangement. Aside from eliminating a circ, what advantage does this have? One disadvantage seems like if the trench wasn't warm already, a short dhw demand would be over before hot water arrived from the garn.