Click here to Find a Contractor in your area.
In fairness to all, we don't discuss pricing on the Wall. Thanks for your cooperation.
Need to contact us? Visit

Backup boiler piping for solar drainback system

SWEISWEI Member Posts: 7,356
Posting here because it's more of a boiler question than a solar question.  Existing drainback solar system with 800 gallon nonpressurized tank.  Tank has two copper HX coils, one for DHW and one for space heat.  Space heat is currently direct piped out of HX to floors with no ODR.  Barrier tubing embedded in concrete on ground level, suspended below (not touching subfloor) of 2nd level.  Backup heat comes from a W-M CGi on LP using manual changeover valves to redirect the boiler output from the floors to the tank.  Needless to say, it's been giving the owner fits for the entire 13 years since it was built.

We are tearing off the ground floor ceilings and installing Watts FlexPlates on the existing PEX (8" on centers.)  I plan to install separate motorized ODR-controlled mixing valves for each floor.  Still looking at control options but the question at hand is the best method of piping in the boiler.  We want long boiler run times, storing whatever is not needed for the floors in the tank.

Since the tank is an open system we have to work with the existing HX coils (no way to add a third since the tank is located on the other side of the house from the boiler room.)  I'm thinking about injecting the boiler water between the HX output and the hot side of the mixing valves, then allowing any excess to go back into the HX and heat up the tank, but not quite clear on the best way to do this.  I could add motorized three-way valves on either side of the HX coil but I suspect there's probably a simpler way.

Any suggestions?



  • heatpro02920heatpro02920 Member Posts: 991

    can you make a fast paint drawing of what you have and want to do? I think I did something similar about 2 years ago...
  • SWEISWEI Member Posts: 7,356

    owner has something the original installer did -- let me see if I can get a copy of that to start with.
  • Charlie from wmassCharlie from wmass Member Posts: 3,922
    pipe the boiler

    with closely spaced tees on the system supply to the mixing valve. could you change the three way mixing valve to a 4 way valve?
    Cost is what you spend , value is what you get.

    cell # 413-841-6726
  • SWEISWEI Member Posts: 7,356

    is certainly do-able.  Would save me a pump for the floor loops - how would two 4-way valves interact?
  • Charlie from wmassCharlie from wmass Member Posts: 3,922
    why a valve for each floor?

    are the required temps different for each floor? With variable speed pumps flow can be handled simply there.
    Cost is what you spend , value is what you get.

    cell # 413-841-6726
  • SWEISWEI Member Posts: 7,356
    two floors

    Ground floor is slab on grade (presumably insulated) has little to no solar gain, and very few windows.  It is not often occupied.

    Second floor has hardwood flooring over subfloor with staple-up and will shortly have graphite plates on those.  It has solar gain from the south plus large west- and north-facing windows and a fireplace.

    I could do a proportional mix, but there are four zone valves which the owner wants to keep.  I'll use them as high limits, but there will still be flow variations.

    I'm open to suggestions.  ∆T pumps should work for tank and boiler, or were you seeing a 4-way on the boiler to add return water protection (tank temp can be in the 80s on a cold winter morning.)
  • Charlie from wmassCharlie from wmass Member Posts: 3,922
    Keep it simple

    I would pipe the boiler in close tees on the supply side to the mixing valves. with a properly sized circ on the boiler flow rate should be covered with at most a manual mix through the boiler feed and return pipes. You do not want too much heat to get back to the tank as it is a big heat loss area.
    Cost is what you spend , value is what you get.

    cell # 413-841-6726
  • Charlie from wmassCharlie from wmass Member Posts: 3,922
    if it was a pressurized

    and super insulated tank you could pipe it for the boiler to be on the return pipe to charge the tank but it would be a waste of fuel. If you warm the tank then the sun comes out where do you store the solar heat if the tank is full of heat? you want only the system requirements to be drawn from the boiler with no excess go into storage as it is paid for heat. You want to leave the tank "space" for the "free" heat.
    Cost is what you spend , value is what you get.

    cell # 413-841-6726
  • SWEISWEI Member Posts: 7,356
    heat into tank

    Design day heat loss is probably around 50k and boiler output is 120k, so we want whatever is not used by the space to charge the tank.

    Later this year, I will be writing control code to model tank charge based on the previous day's use, weather patterns, remaining time between boiler call and sunrise, etc.  That code will be re-used on other systems once it proves out.
  • knotgrumpyknotgrumpy Member Posts: 207
    Could you

    Install an indirect to store excess heat from boiler and pump out of it to the larger tank?

    Just need to be big enough to grab the 'overflow' BTU's from the CGi.

    Would seem simple enough.
  • SWEISWEI Member Posts: 7,356

    may be in his future to replace the commodity tank heater, but that's a separate project.

    The solar tank is unpressurized, so the only way we have to move heat into or out of it is through the HX coil.
  • knotgrumpyknotgrumpy Member Posts: 207
    Drain back

    Do the solar panels just dump the water into the tank? If so, I was thinking about doing the same with the indirect. Plumb it backwards - boiler water into the storage side and use hx in indirect as a 'solar panel'. Pump through that. Pump from near bottom of and dump in top of solar storage tank.

    Maybe put indirect on boiler return? Just have to make sure return to boiler is above 130f.
  • icesailoricesailor Member Posts: 7,265
    Not enough Radiation:

    I think that the second floor is grossly under radiated. First of all, "Solar Gain" is over rated because there is absolutely no solar gain in at 2:00 AM when it is below zero and it is blowing the back end  out of a pig. That's my personal problem with now needing to use Manual "J" for heat loss calculations. Its fine to consider solar gain and North facing rooms for AC but not for heating.

    And a floor heating system like described is highly subjective. Smith's Environmental makes some really nice fan coil units with low temperature limit switches that work really well with low temperatures. I find that floors and rooms like this to be extremely difficult to accurately calculate. I'd do a really careful and accurate (read generous) heat loss calculation on the space. I'd add fan coil units to the space. I'll bet that the problems will go away. In my sailboat racing days, we always lived by the KISS rule. Keep It Simple but don't be Stupid. The more complicated and over engineered a system is, the greater the chances you will have a simple problem that is expensive to correct.

    Go over the figures of the previous designer/installer. You might find a lot of problems there.

    Those plates you are installing only change the heat from going down. They will not improve the flow of heat through the restrictive floor assembly. And there is a limit as to how hot you can make the floor.
  • Charlie from wmassCharlie from wmass Member Posts: 3,922
    If you put the boiler

    on the supply to the mixing valves. Then simple have the circ run after boiler shut down until boiler is below X temperature with the four way directing flow straight to tank it will store excess boiler heat post heat call. Keep it simple.
    Cost is what you spend , value is what you get.

    cell # 413-841-6726
  • SWEISWEI Member Posts: 7,356
    sounds simple enough

    but I'm having trouble seeing a picture in my head.

    I count six ports to connect

    Boiler supply

    Boiler return

    Tank HX supply

    Tank HX return

    4-way valve from boiler

    4-way valve to boiler

    but I'm not clear exactly what goes to what?
  • SWEISWEI Member Posts: 7,356
    second floor radiation

    is a classic example of suspended tube silliness.  PEX is fastened to plywood spacers to keep it away from the subfloor.  An inch or two below that is foil-faced polyiso with fiberglass batts below that.  FlexPlates will put the tubing in direct contact with the subfloor, followed by spray insulation.  This should allow us to drop the fluid temp from 180F to perhaps 120F, which will allow the solar to make a real contribution.  We're about to tear off the ceiling in an unused ground floor room and run a test.  The IR camera and some datalogging will tell us a lot.

    We average 350 days of sunshine here at an elevation 6,000 ft.  The house is up around 7,000 ft and if I had to guess it might get 330 days.  At 32 degrees latitude, that amounts to what is arguably the best solar climate in North America (on a year-round basis.)  We design and size for sunless days, but are aware of local realities.

    The house construction means fan coils would require at least as much interior demo as the ceiling re-work.  They'd also have to be sized for the solar water temps.  Stay tuned.
  • knotgrumpyknotgrumpy Member Posts: 207

    Did you come up with a plan?
  • Charlie from wmassCharlie from wmass Member Posts: 3,922
    too complicated think basic

    1. Pipe from tank to current mixing valves.

    2. Pipe boiler to supply line for mixing valves with closely spaced tees.

    3. Have current four ways divert back to tank if call for heat ends and boiler water is above a set temp.

    4. Make sure controls for each mixing valve do not divert to tank if there is a call for heat in either zone.

    5. Make sure burner stops and starts only on call for heat.

    Circulators for supply and boiler should run from feedback concerning system water temps.
    Cost is what you spend , value is what you get.

    cell # 413-841-6726
  • hot rodhot rod Member Posts: 7,059
    you want too much :)

    I think your plan to inject before the mix valve is sound. But heating the tank with the boiler, not so good.

    I see the dilemma on the run time but, heating the tank as a buffer for the boiler would not be ideal. Add a buffer just for the boiler if you have micro zones, or switch to a modulating boiler?

    Trying to get the best of all with multi input/ output system can get complicated. It comes down to how much $$ to spend to make it the "perfect" system.

    I've been chasing the best "simple" hydronic system my entire career. When I get close I stand back and see a lot of valves, pumps, controls and complication. Probably no one else would ever want to work on such a complicated system.

    When I'm ready to sell a HTP Phoenix with a solar coil would still be one of my first choices. It handles small loads well and has a simple means to add solar. DHW with an external HX.
    Bob "hot rod" Rohr
    trainer for Caleffi NA
    The magic is in hydronics, and hydronics is in me
  • Charlie from wmassCharlie from wmass Member Posts: 3,922
    Hot Rod the idea with dumping into the tank

    was simply to dump the excess heat to the tank for future use in theory. I would differ to your experience for sure.
    Cost is what you spend , value is what you get.

    cell # 413-841-6726
  • SWEISWEI Member Posts: 7,356

    there are no mixing valves at all - just a cast iron boiler feeding both the poured concrete ground floor slab and the 2nd floor suspended tube mess.

    I'm still trying to see where the boiler ports on the 4-way (let's start with just one) should connect.  Back to the tank coil, with the boiler tees on the supply line from the tank?

    99% of the mixing I've done has been with 3-way valves.  I'm still wrapping my head around the 4-way magic.
  • SWEISWEI Member Posts: 7,356
    spoke with Tekmar today

    and they suggested using 3-way mixing valves with closely spaced tees for both the boiler and floor loops.  While it requires an additional pump it does insure predictable flow in the main loop.  Given that the concrete ground level floor will require lower temps than the staple-up 2nd floor, sequencing them appropriately might allow more complete use of stored tank heat.
This discussion has been closed.


It looks like you're new here. If you want to get involved, click one of these buttons!