piping recommendations for best balance

roundrightfarm

I am getting ready to run the pex for 19 wall hung rads. This is a 75k BTUH designed system for a 4 level
house. Each rad will have a TRV and get a homerun to the boiler room. The BTUH for each rad at design temp is small enough to do 3/8" for each rad, but the lengths for the upper floors get quite long and so I've decided to run 3/8" for the lower two levels and 1/2" for the upper two.

I've attached a photo showing how I'm connecting all the pex to two 12 port manifold sets, leaving 5 ports for possible future connections. My question is what is the most advantageous way to connect the pex so that the system has the best balance built into it before any flow adjustments are made. I've copied how each rad measures out below.

I thought for simplicity, I might just put all the 1/2" on one manifold and the 3/8" on the other. Maybe this is the best way since each pipe will be sharing a manifold with other pipes of the same size and of a similar length. It also turns out that piped this way, each manifold would handling about half the total design BTUH.

If it was piped this way, does the order matter for each manifold, i.e. should I put the longest runs or most BTUH closest or furthest from the pump. Does this even matter?

Any considerations I'm missing? Your thoughts are greatly appreciated




BASEMENT Manual J
Bathroom 940
Bedroom 5530
Entry Room 9540
FIRST FLOOR
Bathroom 470
Office 2740
Music 4280
Main 6200
Main/2 6200
SECOND FLOOR
small bedroom 1260
large bedroom 2700
Bathroom-kids 50
Bathroom-Master 1050
Landing 4460
Master 6450
ATTIC
West Window 3700
North window 5540
South Window 5540
East Window 5540
bathroom 400

Rich_49

If that is in fact a picture of your manifolds I would make every attempt to get both sides as equal as possible in GPM and head , also have the highest head circuits farthest from the circ and declining back towards the Tee .

Remember , what goes into a tee must come out of a tee . You probably should have oriented that tee differently . I recommend turning those manifolds over also .

Ironman

Pipe the feeds to the manifolds in reverse return like this:

hot_rod

Here is what reverse return means, so the 1st supply off the manifold is the last returned to the opposite manifold.

I assume all radiators are above the manifolds? No problem with the manifolds facing up, Caleffi offers that version called inverted. Add an IN to any manifold order # to get inverted version.

The way you piped into the manifold header is considered bull head, not my favorite method but will probably work. The flow slams into the tee connection and you sometimes get inconsistent flow to the two opposing directions.

This drawing shows RR as well as parallel piping manifolds off the S&R lines.

roundrightfarm

Thanks Bob for the suggestion. It won't be too much work to swap one of the supply and return manifolds and sweat in the
extra tees and elbows.

Yes, all rads are above the manifold, except the ones on the lowest level, but going under the ceiling is still the only way to get to those.

How does the length of each circuit factor into placement on the manifolds? On the supply manifold, should the longest run be closest to the circulator or furthest from it?

Another question about flushing these manifolds before I attach the pex. Is there a way to flush them through the purging valves for cleaning purposes without capping off every outlet. I assume that I can close the isolation valves on the return lines, but can the flow setters close off the supply manifold outlets?

Rich_49

How much air will get to those vents in this configuration ?

Highest head or longest length usually correspond . If a circuit with the highest resistance is in the middle why would fluid go through it if the easiest path is ahead of it ?

roundrightfarm

Hi Rich,
I'm not sure I understood your response. The longest length circuit has the highest head, so would it go closest to the circ or furthest from the circ?

your first post said " have the highest head circuits farthest from the circ and declining back towards the Tee" but doesn't this put the easiest path in front of the longest/highest head circs?

hot_rod

roundrightfarm said:

Thanks Bob for the suggestion. It won't be too much work to swap one of the supply and return manifolds and sweat in the
extra tees and elbows.

Yes, all rads are above the manifold, except the ones on the lowest level, but going under the ceiling is still the only way to get to those.

How does the length of each circuit factor into placement on the manifolds? On the supply manifold, should the longest run be closest to the circulator or furthest from it?

Another question about flushing these manifolds before I attach the pex. Is there a way to flush them through the purging valves for cleaning purposes without capping off every outlet. I assume that I can close the isolation valves on the return lines, but can the flow setters close off the supply manifold outlets?

If the loops are different lengths use the flow meter/ circuit setter to adjust for proper flow to each loop.

Turn the white plastic caps clockwise to shut off the branch. On the flowmeter, pop the white plastic ring off, flip it over to use it as a wrench. Turning clockwise will shut off that branch.
So you can do a loop by loop purge.

The float vents will not add much value in that location, hopefully you have a good purger back at the boiler. You may hear them bleed as you start filling the boiler.

roundrightfarm

Thanks for your help. The vents came with the manifolds and were configured in that location. Since I will redo some piping, is there a way to put the vents in a better location?

hot_rod

leave the vents there, they have hygroscopic caps, so they should never leak. they are a BSP thread so it would not be easy to move them or plug the openings.

Gordy

Remember you have trvs. Chances of them all being full flow are next to none. As a matter of fact the upper floors will probably be seeing less flow most of the time which would be your higher head runs. With out knowing which loads are on those upper floors hard to say how things will react. Just to throw a slider into the mix

What is your circulator selection? Hopefully a delta p, or t ecm variety.

Mark Eatherton

For the benefit to others reading this post, you should NEVER use a bull headed tee, as shown in your original photo. Parallel reverse return is the best way to simply balance flows, assuming flow requirements are nearly the same. The TRV's will also aid in thermally balancing out the zones. Internal zones satisfy first, and external zones then get more flow. A piece of trivia, the very first article that Siggy wrote was on why not to use bull head tees. In that scenario, the flows never really "lock in" but keep switching flows back and forth. It actually used to be illegal in the mechanical codes, but when primary secondary piping came about, instead of trying to educate the inspectors, the code officials decided to take it's prohibition out of the code. It still matters. With potable water, it doesn't make much difference because you have 70 PSY of driving force. In heating system you have 7 PSI of driving force, and again, it does matter. Don't do it.

ME

roundrightfarm

Thanks Guys,
I'm gonna redo the piping. In my installation, the return lines are not bullheaded due to the flow direction, right? I've attached a drawing of a design that would allow me to keep my manifolds where they are. Is this an ok way to go?


As for the PSI of the system, my radiator supplier suggested running it at 20PSI due to the height of the installation. The five upper level rads will be almost 40 feet above the manifolds. He said that, even in a closed system, any air bubbles will expand as they rise and the higher PSI will help keep any of the rads from getting airlocked if this happens. Does this make sense?

My circulator selection is a Grundfos Alpha2. I'm planning to run this system as continuous circulation with ODR and buffer tank thermometer as the only feedback controls. This circ will be fine without a bypass, even if all the TRVs close themselves off completely, right?

hot_rod

air vents on the high rads are a good idea. With TRVs you will have a nice system, it should not need much manual balance.

Ironman

That's better, but it's still not reverse return.

Imagine it's like a ladder: first in, last out. That's reverse return. First in, first out; that's direct return.

Water takes the path of least resistance. So with direct return, the further down the ladder from the supply, the less flow through the rungs (individual circuits).

Ironman

With 40' of elevation, you should actually have 22 psi static fill. That will leave a safety cushion of about 5 psi at the top of the system. How did you size the expansion tank for the system? With the elevated fill pressure that's necessary, and the increased water volume from the buffer tank and rad's, a standard Xtrol 30 will not be large enough. Have your rep size the tank using the long formula that takes these things into account.

The Alpha is fine: it needs no bypass.

roundrightfarm

With a 22psi makeup water pressure, will I still be ok with a 30psi pressure release on the boiler? The boiler has a max operating pressure of 60psi

Ironman

If the expansion tank is properly sized, yes.

Mark Eatherton

Ironman said:

That's better, but it's still not reverse return.

Imagine it's like a ladder: first in, last out. That's reverse return. First in, first out; that's direct return.

Water takes the path of least resistance. So with direct return, the further down the ladder from the supply, the less flow through the rungs (individual circuits).

Turn these drawings on their sides, and it will make more sense to you. The rungs of the ladder are your connected loads. They just don't show.

It's really easy to do, chews up a minor additional pipe, but will be as balanced as it can be if done right. As long as each "rung" has nearly the same pressure drop (+/_ 10%) you'll be fine. If greater than that then you MIGHT have to made some adjustments, but doubtful. No one, ok very few people commission their systems, and they work just fine. Fill'em, purge 'em and blow and go. Don't even bother to show the consumer the emergency shuts offs.


ME

Mark Eatherton

Gosh that drawing looks familiar

ME

hot_rod

A true hydronic perfectionist, would size the pipe in a RR to decrease as the branches take off Then each crossover, or branch would have balanced flow.

This pic also show the difference between a dead end RR and a ladder style that ME showed.