Radiant heat floor puzzle.

Discussion in 'Plumbing Forum, Professional & DIY Advice' started by dberryco, Jan 22, 2014.

  1. dberryco

    dberryco New Member

    Messages:
    13
    Hi folks,

    Short story: my new radiant heat installation doesn't work well. I'm having pressure related problems and seem to have no end of air in the system despite it holding pressure with no leaks.

    Long story: I have a very simple (at the moment) radiant floor heating system consisting of 10 zones on a 12 zone manifold. I'm using a Grundfos UPS 32-80/2 VersaFlo Cast Iron Circulator Pump 115V, 1/2 HP. I'm filling the system with propylene glycol using a 1/2HP utility pump. It circulates thru a 200K BTU on demand water heater.

    The manifold is: http://www.pexuniverse.com/12-branch-stainless-steel-radiant-heat-manifold.

    House is small. 700 sqft. basement, 700 sqft. main level, 500 sqft. bedrooms upstairs. Zones are: basement closet, basement bedroom, basement bath, living room, kitchen, pantry, foyer/bath, bedroom upstairs #1, bedroom upstairs #2, bath upstairs.

    I'm supplying and returning from the manifold with 3/4" pex. Individual zones are 1/2" pex.

    I managed to get it to work for a few days, and for those days, it worked wonderfully. Took about 12 hours to get the cement slab warm, then it was great. Then I noticed my cheap chinese pump was leaking. Hence the nicer Grundfos multi speed model.

    Here's what is happening: I turn the system on, and I'm only focusing on the basement 3 zones at the moment. Thermostat turns pump on, fluid flows and I get 1.2 GPM each thru the 3 zones. (Bedroom, closet, bathroom). Each is about 150-200' of 1/2" pex in concrete. Slowly, the flow rate slows to 1GPM, then .8 then .6 and finally flow stops and the water heater shuts off.

    I started off with a 1/8HP taco pump and upgraded to the $700 pump thinking that more power was needed. I get the same results using the utility pump and putting fluid into the system. After I shut it off, I have air in the system. I flush the air out, and restart the procedure. It works for a short time, then pressure falls and then it stops flowing.

    I've tried narrowing it down to 1 zone. Same result. I've switched to an upstairs zone. Same result. I've concluded I have some sort of vacuum leak or the pump is cavitating, or something mysterious is going on involving differential pressure and hydraulics.

    I get the same result when filling the system from a 5 gallon bucket of coolant, or when cycling the system when it's a closed loop.

    I have a pressure expansion tank installed. The system holds pressure of 30PSI indefinately, but I have no way of telling if there is a vacuum leak.

    Every time I cycle the system, I get air in the lines. They are white pex, so I can see the bubbles move (the coolant is blue).

    It's become something of a mystery. It's repeatable, consistent and baffling. There must be something stupid I'm doing or something I've missed. I'm sure someone here has run across this before and maybe has the answer.

    My next step failing any better advice, is to replace the manifolds with something I build myself with copper tubing and valves for each zone. I would loose the flow gauges, but I could live with that. I can install automatic valves and individual pumps for each zone if needed, but it seems like an unneeded step if my current manifold works with individual zone thermostats.

    Any ideas? Vacuum leak? Too much pump for the job? Cheap Chinese manifold? Forgot to install widget x in location y?

    Thanks,
    Dan
  2. hj

    hj Moderator & Master Plumber Staff Member

    Messages:
    27,247
    Location:
    Cave Creek, Arizona
    AIr does NOT leak INTO a system. It has to be either sucked in or pumped in, so you have to find the source of the leak, usually a pump shaft seal, but not likely with the pump you are using. You might be getting cavitation, which will separate entrained air out of the water, if the pump is "sucking" less water than it needs, thus creating a "vacuum".
    Last edited: Jan 23, 2014
  3. mikeplummer

    mikeplummer Plumber

    Messages:
    190
    Location:
    Ontario, Canada
    you have not mentioned presence of an air eliminator in the main loop? you should have an air eliminator and you want to make sure your pressure tank connection is closer to the inlet side of circulator.... note: air eliminators work better on the hot (feed) loop ... I would generally install a spirovent about 12" "before" (upstream) the circulator inlet....and the pressure tank connects into the bottom of the spiro via FPT connection in spiro
  4. dberryco

    dberryco New Member

    Messages:
    13
    The manifold kit came with an air eliminator installed on the return manifold. I initially used it as per the instructions, but it seemed to add more air than it eliminated, so I closed off the valve stem cap and forgot about it. Could it be faulty and causing all my problems? I assumed with it capped off that it was no longer in the system.

    Thanks!
    Last edited: Jan 23, 2014
  5. dberryco

    dberryco New Member

    Messages:
    13
    This is helping. Air IS getting sucked in. I'm 99% sure of that. And the 2 replies have given me an idea it's the air eliminator or it's connection to the manifold. I'll be checking it today to confirm.

    Thanks!
  6. Reach4

    Reach4 Well-Known Member

    Messages:
    2,795
    Location:
    IL
    I think it would be interesting to have a pressure gauge on the input to the pump. It seems to me that the intake would have to go below zero PSIG to suck any air in.
  7. mikeplummer

    mikeplummer Plumber

    Messages:
    190
    Location:
    Ontario, Canada
    you are correct in your assumption... this happens when the pressure tank is connected to the system at a poor location....this is why it is important to have the pressure tank connect to the system close to the pump inlet....
  8. hj

    hj Moderator & Master Plumber Staff Member

    Messages:
    27,247
    Location:
    Cave Creek, Arizona
    quote; but it seemed to add more air than it eliminated,

    That could ONLY happen if there were a partial vacuum at the air eliminator, which, by definition, would mean that the pump was "sucking water" rather than pushing it. It could NEVER happen if your entire system were at design pressure.

    quote; .this is why it is important to have the pressure tank connect to the system close to the pump inlet....

    The pressure tank is supposed to be installed at the point of zero pressure change, which is at the boiler itself.
  9. mikeplummer

    mikeplummer Plumber

    Messages:
    190
    Location:
    Ontario, Canada
    the point of zero pressure change IS the pressure tank connection

    if you had a heat source with a low flow resistance you could locate the pressure tank near the inlet (HJ's implication perhaps?)
    since you are using an on demand water heater you will have considerable flow resistance through the heat source so you will want the pressure tank connected closer to the pump inlet
    Last edited: Jan 24, 2014
  10. Tom Sawyer

    Tom Sawyer In the Trades

    Messages:
    3,309
    Location:
    Maine
    post a picture of the boiler, circulator, valves and manifolds
  11. dberryco

    dberryco New Member

    Messages:
    13
    I ordered a new spirovent and will be here friday. I found that my old one was leaking. I'll post a follow-up in a few days either way. I'm pressing on with wiring at the moment since I've already passed my rough plumbing inspection.

    I was fortunate that my no-name (looks like a riefeng) uses standard 1" NPT pipe threads on the input/output of the manifold even if the pex connectors are some odd European standard.

    Relocating the expansion tank is probably half the puzzle, and of course the leaks on the suction side too.

    Thanks all for the help so far.
  12. hj

    hj Moderator & Master Plumber Staff Member

    Messages:
    27,247
    Location:
    Cave Creek, Arizona
    Air eliminators do NOT work good on the return line, because by then the air has already collected in the heating loops. And an air vent on the return side of the pump CAN suck air if the pump is creating a vacuum.
  13. mikeplummer

    mikeplummer Plumber

    Messages:
    190
    Location:
    Ontario, Canada
    I hear my original post echoing regarding placement of spiro....replacing the old one with a new one in the same location on the return line wont help
  14. hj

    hj Moderator & Master Plumber Staff Member

    Messages:
    27,247
    Location:
    Cave Creek, Arizona
    quote; the point of zero pressure change IS the pressure tank connection

    The tank connection does NOT "define" the point. If that were true you could put the tank ANY WHERE in the system and it would work the same, but that is NOT true. The BOILER is the point of zero pressure change and ideally the tank should be connected on the boiler.
  15. mikeplummer

    mikeplummer Plumber

    Messages:
    190
    Location:
    Ontario, Canada
    the pressure at the tank connection will remain and that is why it is the point of no change - the pressure at the boiler will vary with system demand. I don't know how to post pics or I would show you my source
  16. dberryco

    dberryco New Member

    Messages:
    13
    Still no joy.

    I replaced my air eliminator with a spirovent. I can't seem to eliminate the air in my system. I now think maybe PEX isn't suitable for suction applications. Maybe I'll just replace all the pex with copper. I may call a plumber that specializes in radiant floor heating. I've just about run out of ideas.
  17. houptee

    houptee Member

    Messages:
    186
    Location:
    Monmouth County, NJ
    What are you using a tankless water heater as the boiler?
    If so what make and model?
  18. dberryco

    dberryco New Member

    Messages:
    13
    A Renaii 200K btu tankless heater. This doesn't seem to be the problem as I have turned it off during most of my troubleshooting. IE. bypassed it.
  19. houptee

    houptee Member

    Messages:
    186
    Location:
    Monmouth County, NJ
    Im not an expert on Rinnai but tankless water heaters restrict the flow rate thru the unit as the incoming water gets colder, so maybe when you turn it off it closes the flow valve to its minimum position and hence you see flow getting restricted?
    Maybe try to totally bypass the water heater, disconnect the pipes and put a piece of pex from the in to the out pipes, then run the circulator and bleed system of air, see what the flow meters show without the water heater in the system at all.
    Just an idea before you decide to tear everything out!
  20. Dana

    Dana In the trades

    Messages:
    3,027
    Location:
    01609
    Replacing the PEX with copper will do nothing to resolve the issue. A spirovent or air scoop can't work if it's not located in an appropriate place relative to the pump, with an appropriately placed expansion tank as well.

    In this case, whether pumped direct or primary/secondary the pump on the tankless loop should be pumping TOWARD the tankless, with an air scoop & expansion tank on the suction side of the the pump. If it's plumbed primary/secondary this may or may not be adequate for purging/venting air on the radiation loop(s), depending on the exact topology.

    A Spirovent can't work if it's located a negative pressure (relative to atmospheric pressure in the room). In a well designed systems there won't be any such point, but on a high pumping-head/low flow situation the suction side of the pump could end up being below atmospheric pressure. Pumping only toward the Spirovent would allow it to work.
Similar Threads: Radiant heat
Forum Title Date
Plumbing Forum, Professional & DIY Advice Cutting concrete floor containing PEX radiant heat lines Apr 28, 2013
Plumbing Forum, Professional & DIY Advice Balancing valves for PEX manifold (NOT radiant heating!!) Mar 1, 2013
Plumbing Forum, Professional & DIY Advice Radiant Heat, Pex Supplier, Parts, Outdoor Boiler Insulated Line Aug 17, 2012
Plumbing Forum, Professional & DIY Advice Radiant Heating Jan 23, 2012
Plumbing Forum, Professional & DIY Advice Radiant heat -- replace mutiple circulating pumps with single variable speed pump Dec 18, 2011

Share This Page