distribution plumbing...

  • Active since 1995, Hearth.com is THE place on the internet for free information and advice about wood stoves, pellet stoves and other energy saving equipment.

    We strive to provide opinions, articles, discussions and history related to Hearth Products and in a more general sense, energy issues.

    We promote the EFFICIENT, RESPONSIBLE, CLEAN and SAFE use of all fuels, whether renewable or fossil.
Status
Not open for further replies.

kuribo

Feeling the Heat
Dec 10, 2007
388
SW WI
Here's my tentative distribution plumbing plan. I have pex in the basement which I would like to split into two zones supplied by one pump, and I have it in floating slabs on the first floor, which I would like to also split into two zones supplied by a separate pump. P1 and P2 are Viridian variable speed, delta P pumps...Reason for two pumps and the crossover is the ability to run some heat in case a pump goes out. 4 zone valves as shown. Supply is from (2) 500 gal propane tanks in the basement. Supply temp set by outdoor reset with a motorized mixing valve. Calculated flows at design point of 80,000btu/hr load are 16 gpm total, 8 gpm upstairs, 8 gpm downstairs, then 4 gpm through each of the 4 zones, which yields about 0.5 gpm through all of the infloor 1/2" pex (all loops are the same length), with a 10F delta T.

Greatly appreciate any feedback....Thanks!

distribution.jpg
 
Ideally you don't want to "bullhead" a tee like that. Flow moves better and more evenly piping branches like this drawing.

It's not a big deal in water piping, but pumped flows prefer not to be bullheaded.
 

Attachments

  • Screen Shot 2016-01-18 at 8.53.23 PM.png
    Screen Shot 2016-01-18 at 8.53.23 PM.png
    30.5 KB · Views: 176
Thanks for the tip! I have read that somewhere, just didn't put it to paper!
 
Having all four ZV on the same manifold would simplify things and would achieve your goals.

A single VR1816 would push about 0.40 gpm through all 32 (250 ft?) loops if they were all calling at once, which would likely be more than adequate for normal conditions.

If you really needed more flow the second VR1816 could be placed in parallel with the first one and could be activated during extreme conditions. Then you would also have the redundancy you're looking for.
 
  • Like
Reactions: Karl_northwind
Yes, all loops are 250 feet....I thought about just installing one pump and have the extra one on the shelf but wasn't sure if one VR1816 would have enough grunt to get the flow required at the low supply temps and low delta T at my design point of 80,000 btu/hr. Putting them in parallel and just using one unless more is needed makes great sense, thanks! Would I need some swing checks to keep flow from messing with the 2nd pump when it was not on? Maybe something like this?

distribution.jpg
 
Last edited:
  • Like
Reactions: ewdudley
the flow checks in the circulators should do that, but I still use swing checks. belt and suspenders.
 
Pumps in parallel but feeding the manifold from opposite directions, nice.

Looks correct to me, and flo-check plus swing check, why not.

You may want to put your "crossover" valve back into the middle of the manifold so you have the option of isolating the two pumps from one another should you choose to run both at the same time. I can't say for sure how the two pumps would interact when both are controlling for constant deltaP or proportionalP.

Would they tend to balance at roughly the same fow rate, or would one pump dominate, or would they take turns dominating as the flow rates oscillated, increasing on one while decreasing on the other, then increasing on the other while decreasing on the one?
 
Last edited:
Good idea on the adding back the crossover!

.....One thing I am wondering about is the fact that the mixing valve will be throttled down a great deal when the supply water temp from storage entering it is say, over 140-150F, and the mix temp is only 90F or so....that's a lot of head loss for the circulator(s)....issue?

Looking at the bigger picture, if the mixing valve is set to output 90F water and my radiant is returning 80F at the design point max heat load of 80,000btu/hr at the design max flow of 16gpm, I'm failing to see what outdoor reset will do for me? Supply temps don't need to increase, and they can't really decrease much, the pumps won't decrease flow unless zone valves close, so wouldn't the system, with a constant supply temp of 90F, without outdoor reset, simply close/cycle zone valves if the temp was met on days when the heating load is less than the design temp? In other words, wouldn't this system just operate more or less at the constant 90F supply temp with flows varying as zone valves open and close-on the design day, the valves will be open most of the time perhaps, and on warmer days without the need for 80,000btu/hr, the valves will cycle......Do I have that right?
 
Last edited:
Good idea on the adding back the crossover!

.....One thing I am wondering about is the fact that the mixing valve will be throttled down a great deal when the supply water temp from storage entering it is say, over 140-150F, and the mix temp is only 90F or so....that's a lot of head loss for the circulator(s)....issue?
Do you have a three-way outdoor reset mixing valve in mind that has a high enough cV for this situation?

[edit:]
I think probably a three-way outdoor reset mixing valve that is big enough to meet your gpm needs would be many hundreds of dollars.

Your net flow from storage will be on the order of 1 gpm. You might consider something like a 15-58 on low speed or maybe one of the tiny Bell & Gosset ECM pumps for injection flow and just use a $150 Tekmar 256 in on-off mode for outdoor reset injection.
 
Last edited:
I have a motorized mixing valve I picked up with a high Cv already....don't recall the specifics but I think it's a 1 1/2" Danfoss with Cv in the high 20's.......I will have to dig it out and check. I did the ballpark calcs some time back and found a good deal (relatively speaking!) so I bought it...

As for the flows from storage, I get 1.6 gpm at 180F supply temp, 2.67gpm at 140F, 4 gpm at 120F, down to 16 gpm at 90F supply temp....

I suppose this highlights the difficulties of actually implementing a wide delta T supply...
 
Last edited:
I have a motorized mixing valve I picked up with a high Cv already....don't recall the specifics at the moment but I do remember it wasn't cheap....I will have to dig it out and check. I did the ballpark calcs some time back and found a good deal (relatively speaking!) so I bought it...

As for the flows from storage, I get 1.6 gpm at 180F supply temp, 2.67gpm at 140F, 4 gpm at 120F, down to 16 gpm at 90F supply temp....
(I was guestimating higher deltaT and lower btu / hour.)

If you've got the valve just double-check the cV and we can compute the flow for the whole system from storage, through mixing valve, to in-floor loops, and back.
 
I'll dig it out tomorrow and verify.....Thank you very much for your continued assistance!
 
I'll dig it out tomorrow and verify.....Thank you very much for your continued assistance!


Ideally the Cv of the control or mixing valve should be close to the actual load. When you over-size the valve it loses it's ability to accurately track and control the output, called "valve authority"

All valves should be sized by flow rate, Cv number, definitely not by pipe size.
 
Pumps in parallel but feeding the manifold from opposite directions, nice.

Looks correct to me, and flo-check plus swing check, why not.

You may want to put your "crossover" valve back into the middle of the manifold so you have the option of isolating the two pumps from one another should you choose to run both at the same time. I can't say for sure how the two pumps would interact when both are controlling for constant deltaP or proportionalP.

Would they tend to balance at roughly the same fow rate, or would one pump dominate, or would they take turns dominating as the flow rates oscillated, increasing on one while decreasing on the other, then increasing on the other while decreasing on the one?


Just be careful with swing checks, they are not ideal for hydronics, Swing checks need some pressure differential to close and seal properly, when a hydronic pump shuts down pressure is the same on both sides. Unlike a well pump or sump pump, for example where swing check work best.

Swing checks have a long travel, so they respond slowly, and need full flow to swing the gate out of the way. The Cv is always with the swing wide open, any position less causes head loss, sometimes a lot of head loss!
Always size them by CV not pipe size. Often they are two sizes smaller than the pipe.

They don't get along very well with variable speed pumps as low and or variable flows cause the swing to flop in limbo. A typical 3/4 swing check may have a Cv or 9-11. Anything gpm less than the rated Cv allows the swing to start closing down.

Best to use a soft seat check with a low "pop" spring to close it, just like you see in the hydronic circ pumps these days, .35- .50 psi pop, large face conical disc, soft seal and seat. You can get hydronic specific checks from Watts, Apollo, most any valve manufacturer.
 
I checked and the valve I have is a Tekmar 712, 1 1/4" with a Cv of 21. The next smaller model, the 1" unit, has a Cv of 14, which is on the small side at the design point flow of 16. Thought it would be better to error on the too large side...since my flows will probably rarely be 16gpm, maybe I should have gotten the 1" ???
 
I checked and the valve I have is a Tekmar 712, 1 1/4" with a Cv of 21. The next smaller model, the 1" unit, has a Cv of 14, which is on the small side at the design point flow of 16. Thought it would be better to error on the too large side...since my flows will probably rarely be 16gpm, maybe I should have gotten the 1" ???

You should be fine, if you run 16 gpm you want the 1-1/4 pipe size, puts you right between 2- 4fps, ideal.
 
The potential problem will be when the flow drops down to half that I suppose...

What are your thoughts on this:

Looking at the bigger picture, if the mixing valve is set to output 90F water and my radiant is returning 80F at the design point max heat load of 80,000btu/hr at the design max flow of 16gpm, I'm failing to see what outdoor reset will do for me? Supply temps don't need to increase, and they can't really decrease much, the pumps won't decrease flow unless zone valves close, so wouldn't the system, with a constant supply temp of 90F, without outdoor reset, simply close/cycle zone valves if the temp was met on days when the heating load is less than the design temp? In other words, wouldn't this system just operate more or less at the constant 90F supply temp with flows varying as zone valves open and close-on the design day, the valves will be open most of the time perhaps, and on warmer days without the need for 80,000btu/hr, the valves will cycle......Do I have that right?
 
Status
Not open for further replies.