Gravity recirc loop strange problem

Discussion in 'Plumbing Forum, Professional & DIY Advice, Tips & ' started by benze, Oct 29, 2010.

  1. benze

    benze New Member

    Messages:
    63
    Location:
    Montreal, Canada
    Hi,

    I just tried putting in a single branch gravity recirculation loop and am running into a really strange problem. I've got a bottom feed hot water tank (ie: the cold water enters at the bottom), and the hot water exits at the top.

    My incoming cold water supply is 1" PEX and my outgoing hot water is 1" PEX. The 1" PEX is branched off to 3/4" PEX which is connected to 1/2" copper to my kitchen sink. I tee'ed into the 1/2" copper and ran a 1/2" pex back to the HW tank. At the tank, I removed the original drain faucet and put in a Tee and connected the return line to that point.

    So here is the weird problem. When I first put this config together, everything "seemed" to work better. I was getting hot water at the kitchen sink within 3-5 seconds vs. the original 40+ secs. However, I soon noticed that the hot water coming out of the faucet wasn't as hot as it should have been. I figured that I was having a problem with reverse flow; that I was getting cool water going up the return line and making the hot water at the faucet less hot. So I added in a check valve on the return line, figuring that would prohibit any cool/cold water from travelliing back up that path.

    However, now the recirc loop doesn't seem to work at all. I'm back to waiting 40secs to get hot water from the faucet.

    I'm really confused at this point as to why it seemed to work without the check valve and fails with it.

    I know that a recirc pump would resolve the situation, but am trying to figure out why it would seem to work better without the check valve than with it.

    I've used a SharkBite check valve, if it makes any difference.

    Thanks for any insights!

    Eric
  2. johnjh2o1

    johnjh2o1 Plumbing Contractor for 49 years

    Messages:
    1,143
    Location:
    South*East
    What type of check valve did you use a swing check or a spring check? A gravity system will not open a spring check.

    John
  3. hj

    hj Moderator & Master Plumber Staff Member

    Messages:
    26,677
    Location:
    Cave Creek, Arizona
    Gravity will often not open a swing check either. We used to install the check valve an angle so that the "flap" hung slightly open to allow the circulation, but back flow would push it closed. A small hole in the flap will do almost the same thing.
  4. benze

    benze New Member

    Messages:
    63
    Location:
    Montreal, Canada
    I am using a SharkBite check valve, and I'm not 100% sure, but I believe that a SharkBite checkvalve is a spring valve. That's what you get for being lazy. I had both in my hands, and figured for an extra 2$ for the SharkBite, I would just be lazy and save the sweating of a swing valve.

    I guess I'll go grab myself a swing check valve Monday and test it out. Ideally I want to install it on a vertical pipe; will a swing check valve work on a vertical pipe? Will the backflow be sufficient to close the flap? Do I still need a hole in the flapper? It should, in theory, hang open if there is no back flow, correct?

    Thanks!

    Eric
  5. arfeller

    arfeller New Member

    Messages:
    77
    Location:
    Port Angeles, WA
    You do not want to put the check valve vertical. The valve should be placed on a horizontal. I have a gravity system that works very well and I used a swing type check valve on the horizontal with a 1/8" hole drilled through the center of the check valve flapper.

    You may be able to get away with a slightly smaller hole. Good luck!
  6. benze

    benze New Member

    Messages:
    63
    Location:
    Montreal, Canada
    HJ mentioned that he has installed them at an angle; any idea what the max angle would be? Is 45deg too much? Putting it at horizontal would require a bunch of extra effort to make it a nice looking / neat job. If I really need to put it horizontal, can I put 90 elbows right before and right after? Or is the sudden change of direction too difficult for the gravity circ to handle?

    Does your 1/8" hole not cause you backflow issues? What size valve is it? (3/4"? 1/2"?) I realize the flow wouldn't be enormous, but I assume you would still have a little backflow through there? Or is the recirc loop pressure strong enough to counter any water trying to backflow through the hole?

    Tx,

    Eric
  7. Hairyhosebib

    Hairyhosebib New Member

    Messages:
    173
    Location:
    Arizona
    They do make a spring check for solar hot water heater systems that take very little pressure to open. Maybe only a 1/2 LB PSI.
  8. johnjh2o1

    johnjh2o1 Plumbing Contractor for 49 years

    Messages:
    1,143
    Location:
    South*East
    Solar systems have pumps they don't run on gravity.

    John
  9. hj

    hj Moderator & Master Plumber Staff Member

    Messages:
    26,677
    Location:
    Cave Creek, Arizona
    1. the return line to the heate should be flowing DOWNWARD, if you expect it to work, and a swing check does NOT work when it is pointing down.
    2. The angle depends on the angle of the "flap" in the check valve. It is installed at slightly more than that angle so the flap is slightly open to allow flow, but still closed enough so that any backflow will close it entirely.
    3. That is why I said the small hole will "do almost the same thing".
    4. A gravity system has basically ZERO pressure so any resistance will stop the circulation. In fact any up, then down, (or vice versa), offset will also stop it from working, which is why few gravity systems will work unless the system was installed with it in mind.
  10. arfeller

    arfeller New Member

    Messages:
    77
    Location:
    Port Angeles, WA
    Hopefully I attached the photo correctly. I'm not a plumber and installed this gravity loop myself. So there may be errors...

    Sometime in the future I will be pulling out that plastic nut between my brass fittings and hot water heater. Also adding a T with a quarter turn valve to empty the tank is also something I should have.

    The pex is 3/4" and so are the fittings. I do not have any issues with back flow.

    Attached Files:

  11. benze

    benze New Member

    Messages:
    63
    Location:
    Montreal, Canada
    Thanks for the tips HJ. Have a few questions based on your response though:

    Yes - return line is flowing downward. If the Swing check is installed vertically, the flap would be essentially be normally open (ie: just hanging down loose). I was thinking that if the backflow was strong enough, part of the stream would catch the edge of the flap and push it upwards to a horizontal position thereby shutting the backflow. But would it stay closed, or would the backflow pressure drop enough for the flap to open up again?

    As in point #1, if the backflow closes the flap, would the pressure then not drop thereby allow the flap the reopen? And once reopen, then reclose due to the backflow? Is the flap basically opening/closing/opening/closing cyclically while the tap is on?


    If I find I need to place it on the horizontal with a hole drilled thru the flap, can I put 90s right before and right after so I end up with vertical->90->swing check->90->vertical? Or do I need some run along the horizontal plane to keep the circulation/momentum of the water moving.

    Thanks again for the tips. I'm learning much.

    Eric
  12. arfeller

    arfeller New Member

    Messages:
    77
    Location:
    Port Angeles, WA
    Your walking a fine line with hj, I think this is about to get interesting : )
  13. hj

    hj Moderator & Master Plumber Staff Member

    Messages:
    26,677
    Location:
    Cave Creek, Arizona
    The backflow would NEVER close an upside down check valve, period. The backflow will close the check valve ANY TIME it occurs, and keep it closed as long as there is a reverse pressure differential, which is what you want to happen. As long as the valve is horizontal, it makes no difference what you do on either side of it.

    Artfeller;
    I would NOT, repeat not, have connected to that water heater drain valve. They are not designed to hold pressure when opened. You should have opened the valve all the way, then while pulling on it rotated it the opposite direction, to remove the plastic piece. Then you would have had a 3/4' brass pipe to screw whatever fittings you need to make the connection and provide a drain valve. Using 3/4" pipe was overkill. you could have used 3/8" or 1/2" and got the same, or better, performance.
  14. benze

    benze New Member

    Messages:
    63
    Location:
    Montreal, Canada
    Good point; had forgotten that the pressure on the faucet side of the checkvalve would drop since the tap would be open thereby allowing the checkvalve to remain closed.


    Not quite sure if your definition of an "upside down check valve" is one that is rotated vertical at 90deg, or one that is installed in the wrong direction.

    Thanks again for all your tips. Will try a couple of things out today now that the plumbing store is open and can swap my sharkbite for a swing check.

    Eric
  15. arfeller

    arfeller New Member

    Messages:
    77
    Location:
    Port Angeles, WA
    Yeah, I goofed on that assemble. My buddy told me about removing that drain plug after I had it all together. It is on my list... I think I'll move it back up to the top now that you have reminded me : )
  16. benze

    benze New Member

    Messages:
    63
    Location:
    Montreal, Canada
    Ok - so now I'm even more confused. I picked up a 3/4" Watts tee swing check valve and a couple of 45s. Before I installed it, I did a quick test with a garden hose to see how much pressure it would take to close the flap. Well, even at a minimal flow (faucet no more than a 1/4 open), the valve would close at 45 deg no problem. Just for fun, I tried it at full vertical (ie: upside down vertical), and at a faucet 3/5 open, the check valve would close. So I figured 45 deg should work fine.

    I pulled out the Sharkbite that wasn't working and sweated in the 45 elbows the check valve. And then ran tests with my faucets. With the circ loop closed, my hot water temp is about 125F. When I open the recirc loop the temp drops to about 105. Which means to me that I am still getting cold water return thru the check valve, which means that the flap isn't closing.

    So I am utterly confused.

    Tomorrow I'm going to try and get 22.5 instead of 45 and see if that makes a difference, but I would still like to understand the logic and physics as to why it isn't closing. Unless the backflow is very slow and minimal, I would expect the flap to close. And given that there is a 20deg drop in the water temp, I am assuming (wrongly maybe) that the backflow is still pretty big.

    Any ideas?

    Tx,

    Eric
  17. hj

    hj Moderator & Master Plumber Staff Member

    Messages:
    26,677
    Location:
    Cave Creek, Arizona
    quote; think I'll move it back up to the top now that you have reminded me :

    Do not do that, unless you intend to add a pump to the system. Just connect your existing piping, with a tee and drain valve directly to the water heater's pipe.
  18. hj

    hj Moderator & Master Plumber Staff Member

    Messages:
    26,677
    Location:
    Cave Creek, Arizona
    You are making this a lot more complicated than necessary. Hooking a hose to the outlet of an upside check valve WOULD close it because the rushing water would "grab" the flap and push/pull it closed, but that is NOTHING like what happens when the valve is actually installed in a plumbing system. You are testing the valve wrong. BREATH through the valve and you will get about the same pressure a gravity system exerts on it. How about a picture of what you have because it is starting to become illogical.
  19. benze

    benze New Member

    Messages:
    63
    Location:
    Montreal, Canada
    Thanks for all the help HJ. I'll get a picture posted later today. I guess I just don't understand the physics of a gravity system. After trying the Sharkbite and seeing that fail, your help has made me understand that the pressure from a gravity flow is minimal. I assumed incorrectly it seems that the pressure from then tank backflow would be significant and be enough to "grab" the flap.

    The other thing I just cant get my head wrapped around is the idea of a 1/8" hole through the flap. I understand that hole will allow the gravity loop to circulate even if the flap is in the closed position. And will prevent all backflow, save and except anything that runs through the 1/8" hole. My question then becomes is this any better than a ball valve 1/8 open?

    Thanks,

    Eric

    Attached Files:

    Last edited: Nov 2, 2010
  20. hj

    hj Moderator & Master Plumber Staff Member

    Messages:
    26,677
    Location:
    Cave Creek, Arizona
    A ball valve is NOT a precise method, since you do NOT know when it is "1/8 open" and even then it would be more than the 1/8" hole. The hole is a "fall back method" which I never used. I just mentioned it as an alternative way to maintain flow. The small hole will create more than enough "resistance" so that the water has an easier path going the "right way" to the faucet, thus backflow would be negligible at best, or worst.
    Last edited: Nov 2, 2010
Similar Threads: Gravity recirc
Forum Title Date
Plumbing Forum, Professional & DIY Advice, Tips & Any way to have a gravity recirc loop with a tankless water heater? Jul 4, 2012
Plumbing Forum, Professional & DIY Advice, Tips & Gravity vs Pump recirculating loops - which is more energy efficient? Oct 23, 2010
Plumbing Forum, Professional & DIY Advice, Tips & Converting a pumped recirculation hot water system to GRAVITY Apr 23, 2007
Plumbing Forum, Professional & DIY Advice, Tips & Gravity Hot Water Recirculating Loop Questions Jan 17, 2007
Plumbing Forum, Professional & DIY Advice, Tips & Gravity fed recirc loop question. Feb 3, 2006

Share This Page