Long headers vs dedicated loop homeruns - staple up

ewang

I'm beginning to lay out the staple up (with plates) loops and I'm running into the issue of maxing out loop lengths at 300' for 1/2 pex and trying to avoid punching a larger than necessary hole in my block basement wall for my quad-level house.

One idea is to use a 3/4" header coming from my manifold, which will then tee off into separate 1/2" loops.

One obvious downside is the lack of balancing between the two loops, especially given the odd number of bays I am filling. Another downside is adding in tees in what will become a drywall ceiling.

Any other thoughts? The picture is the most simple example of what I am facing. On the other side of the house, I'm looking at 3 separate loops, so if the 3/4" header idea isn't crazy, I'd save knocking out structural blocks.

From a heating perspective, I have FA zoned backup, so "nailing it" isn't absolutely necessary, but I'd want to have some semblance of an even heated floor.

Thanks!

TAG

How about remote manifold ? drop them down in the middle

I just did a large room with 11 loops and did them in two groups -- 5 and 6 bringing them down in the middle w/ remote manifold.

STEVEusaPA

If you can adjust the flow via flow meters on your manifold, I wouldn't worry about it and adjust there-using the meters and a thermal imaging camera.

ewang

@TAG - By a remote manifold, do you mean one downstream from my main manifold?

TAG

Yes -- you extend the full size pipe out to the area. Where is the higher heat load in the room ? If the loops are 300 feet that room is far away.

ewang

@TAG - By remote manifold, do you mean a secondary manifold off my main manifold that would then be mounted in the room? I guess that's a good option. Would require an additional manifold, but at least I wouldn't have to mount drain valves on this one since it would be in a finished space without a drain.

@STEVEusaPA - My gut tells me that I should be fine with the relatively short loops. I purposefully put the shorter loop at the end of the header, and at the exterior wall where it will be coolest.

Maybe I'm overthinking, but wanted to run it by the experts.

ewang

Sorry, the forum ate my original post and then reposted it, which is above.

@TAG The coolest side is on the right, next to "Zone 1".

TAG

Sometime it's easier to have a remote manifold vs running all the tubes back to a central spot. It does not come off the other manifold -- they branch off the main. I'm assuming zone 1 has a higher heat loss and you are thinking that for a given flow that will stay hotter?

I use that theory -- did so when I just did my plates. But, I also had to as they were the farthest and I will insulate the lines ... I'm not really sure it does much with constant circulation and proper flow w/ correct length lines observed.

You only have two loops

ewang

I was thinking that since Zone 1 would most likely have the highest flux, that it would be better to give it the shorter loop length.

You're right... I'm planning on constant circulation with ODR, so really as long as I'm staying under my limits, I should get good water temps regardless, just varying delta T, which isn't all that bad.

On the other side of my house, I'll have 3 loops, so I'll just pull both manifold off the secondary. Thanks for the advice.

hot_rod

Is it multiple zoned rooms? Is that why the various loop lengths? If not make them all within 10% of one another and a reverse return "long header"

ewang

@Hotrod Sorry, I should have called them zones - they are in fact all loops that are all within one zone. The reason for the different loop lengths is that I have a relatively long run to the manifold and I worried that by combining the Loop 1 & 2, I would exceed limits for 1/2 pex and could have uneven heating given the length.

See below for a full layout. Note that there are areas (in grey) that have ductwork or other obstacles I can't feasibly run pex and plates.

The 3/4" headers and runs from the manifold to the 1/2" loops are intended to minimize pressure drop and the number of pex runs I need to either cross the stairs/beam in sub basement (loops 1&2), and penetrations I must make through a block wall (loops 3&4).

I intend to have an Uponor EP manifold with balancing valves to balance flow between the feeders for loops 1&2 and loops 3&4.

Loops 1&2 serve the master bedroom, which is usually warmer than the areas served by loops 3&4 (office/guest bedroom), due to two bodies and the electric heated floor in our master bath. The master bath will not have staple-up since we already have the electric heated floor under tile, and there's already a lot of piping underneath this area.

My questions are:
1. Given my layout below and estimated loops lengths, would you anticipate I would have adequate flow for even heating?
2. Would you have any reservations of concealing fittings in ceilings finished with drywall?
3. Is there a better way to accomplish what I'm doing?
4. Any more info or pics that would clarify what I've got going on?




Thank you!

ewang

Posting the pic a different way incase you cant read it the way I posted above.

hot_rod

So you mentioned loops 1-4 are all one zone, one thermostat? Generally if loops are within 10% of one another, no need to balance, not your case, however.
Simply putting a manifold in an access panel near "stairs up" with the ability to balance should be the easiest and allow future tweaking. You could also add actuators later if you want to zone those areas.
A 3/4 supply and return to the manifold would supply plenty flow for those 4 loops.

ewang

Yes, planned on having all loops on one zone, one thermostat.

I hope my picture is clear enough. With the plan as-is, my manifold would actually only have two 3/4" connections. One for the header for loops 1&2, one for loops 3&4.

I could probably combine loops 1&2, but my length would probably exceed 300' if I include the header length to the manifold. With filling 7 bays, its difficult to balance them within 10%.

hot_rod

ewang said:

Yes, planned on having all loops on one zone, one thermostat.

I hope my picture is clear enough. With the plan as-is, my manifold would actually only have two 3/4" connections. One for the header for loops 1&2, one for loops 3&4.

I could probably combine loops 1&2, but my length would probably exceed 300' if I include the header length to the manifold. With filling 7 bays, its difficult to balance them within 10%.

Not sure having two separate S&R solves anything. Certainly do not need two for that connected load requirement. The bigger issue is the loop lengths, even on two separate assemblies are are a bit off from the "best practice" as far as lengths. 110 X 10% = 121, but you have a 110- &145' on that zone. Obviously the shorter loop get more flow. That may or may not be an issue, but once you bury a manifold, not much you can do to adjust, balance, tweak.
One manifold with 4 balance valves allows you to dial in the flowrate that your design calls for. Or use the trial and error adjustment method, if you don't have a design showing required flow in each loop.

Or connect loop 1 &2 together, call it 255, now a 3 loop manifold could be used and you have a 225, 240 and 255, that is a better match up.
Reverse return those and call it a day. But no ability to make fine tuning without balance valves.

ewang

I think you talked me into just combining loops 1&2 into Loop 1*, and just biting the bullet and boring additional holes to run dedicated homeruns for loops 3 & 4 incase I run into future issues of balance, or desire to zone. See below for revised layout. Actually, the loops look pretty good. This time, I added in the estimated additional length required to get to the manifold.