copper in-floor tubing vs PEX

previn

I am installing a radiant floor in my home, I have extensive hardwood and tile. I have done my heatloss calculations (55KBtu/hr). When I came up with the tubing to use it showed I needed tubing on 8" centers between joists using a PEX product. I want to use copper. Since copper is a much better conductor of heat than PEX, what is the equivalent copper use compared to PEX or even PEX-AL_PEX?
Copper also is readily available and close in cost to PEX.
any advice anyone?

chris smith

copper

it going to be awful tough to pull 250' loops of copper, why not use heat transfer plates with pex if you want more eff. heat transfer

chris smith

paradise porter maine

previn

more info

I want to run hard copper in the bays, soft copper in the curved transition areas like between the joists....
Pulling "soft" copper is not what i had in mind.

Mark Hunt

Attatchment method?

> I want to run hard copper in the bays, soft

> copper in the curved transition areas like

> between the joists.... Pulling "soft" copper is

> not what i had in mind.





To Learn More About This Contractor, Click Here to Visit Their Ad in "Find A Contractor"

Mark Hunt

Attatchment method?


How are you planning on attatching the copper to the sub-flooring?

I am going to assume that you plan on having the copper in direct contact with the sub-floor, if this is the case you should keep in mind that copper has a pretty high expansion coefficient. Pex tubing does too, but Pex-Al-Pex is lower than both.

As for the conductive differences between copper and Pex products, I really doubt that you would realize any substantial savings with copper.

All that being said, it is your house and you can do whatever you want. Your control strategy should include variable speed injection and perhaps constant circulation to keep expansion noises to a minimum.

JMHO

Mark H

To Learn More About This Contractor, Click Here to Visit Their Ad in "Find A Contractor"

Tom Hughes

Be Careful

Underfloor copper systems are notorious for leaking after years of expansion/contraction. I have 1st hand knowledge of one project that had horrible water problems. I'd stick with PEX.

Art Pittaway

Two things,

1) Look for the written warranty of copper vs. PEX (the good stuff, not H.D.)
2) You are planning to insulate under the tube and know how much and where?

heatboy

Considerations

By doing what you propose, the supply and return piping scheme will have to be a reverse return configuration. It is the only way that flow rates will be quasi-equal in the joist bays. Do you know what that is? A direct return set-up will cause short cycling unless some type of flow regulator is used for each bay.

What are you using to suspend the copper in the joist bay? I would recommend "clics". #15 for 1/2 copper or 1/2 PEX.

hb

To Learn More About This Contractor, Click Here to Visit Their Ad in "Find A Contractor"

Mike T., Swampeast MO

Convection really is the major method of heat transfer in any joist bay system that uses bare tube. In practice it can be difficult to hold the tube directly against the floor throughout its entire length and even if you do there is only the tangent line of direct contact (no, it's not a true line with infinitely small width, but it's not very wide either).

In general, such systems operate at rather high temperature (160°-180°--at least one person uses copper at 200°-220° however) in order to get the convective currents moving. It can be difficult to get enough heat in cold climates and have quick enough response in climates with rapid temperature change. Both of these problems can be negated to a great extent by using conduction plates--remember heavier IS better.

I'm not aware of any "scientifically verified" data regarding bare copper "staple-up" or any "correction factor" to use in comparison to similar PEX systems.

Years of success has been reported (in New Jersey I believe) by:

1) age the copper outdoors to enhance its radiation potential

2) pre-make the loops for each joist bay

3) attach tight to floor with copper straps

4) insulate THOROUGHLY AND ACCURATELY maintaining a consistent air gap!!! (foil-faced rigid insulation preferred)

5) run temp as high as you want (without producing steam of course)

The person doing this reports no problem on the finish floors despite the high temperature and very rapid response.

While some (usually contractors) certainly report good comfort and efficient operation with bare-tube joist bay systems others report the worst disasters imaginable. The former seem to come mainly from those highly familiar with the local climate requirements and who have developed their own "system" for sizing and installation. The latter seem to come mainly from inexperienced contractors and homeowners who rely on engineering data from the "oh so simple" radiant firms out there.

Steve Ebels

Where's Ken

When you need him for a good old fashioned scrum.

KEEENNNNNNN ohhhh KEEEEEEEENNNN, Where AAARRRRRRRREEEE YOOOUUUUUU??

hot_rod

I'll be doing a copper radiant

sometime this summer. Here is my plan. Using my REMS Curvo I'll bend 8" OC loops of 20 foot copper lengths. To jump joist bays I will use some prebent U bends with swaged ends to minimize solder joints. I will prefab the entire tube mat in my shop. Not unlike the ols Revere Copper Company system from years gone by

Either ThermoFin U or C fins will be used to fasten the tune to the underside of the subfloor.

Here's what copper will do for you in this installation. Zoom, zoom, zoom! The heat transfer from the fluid to the tube wall to the plate and onto the floor in almost instant! I documanted this with IF cameras. Warmboard was faster still!

The tube wall of PEX or rubber never does reach the temperature of the fluid inside the tube. I measured the pex at about 5 degree delta T, and about 9 on the Onix product, after a 6 hour run time.

Go to www.pmengineer.com and click on the April 03 issue for an excellent article on this subject "Back to Basics" A Study of Heat Transfer, Pipe to Pipe" Dr. Birol Kilkis. My actual measurment show his calcs to be real, real close.

So, if the pursuit of ultimate heat transfer is your goal copper is a clear winner. The graph in the above article shows a temperature drop of "pratically zero" in the 1/2" copper L example.

Combine it with transfer plates, a condensing boiler, a good insulation package, and you may just have the ultimate radiant floor "fossil fuel to warm floor" transfer package

In this photo I had forgot to screw the transfer plates to the plywood after I crimped the pex onto the copper. You can spot the two sections where the plates are not in contact with the floor!

hot rod

To Learn More About This Contractor, Click Here to Visit Their Ad in "Find A Contractor"

Unknown

I remember once

replacing a boiler , and the customer installed Argo baseboard elements in the joist bays underneath his living room floor . No insulation between the basement and elements . He said he got some good heat out of it - me I'm not so sure , especially without insulation . But most of the aluminum elements were in contact with the floor . Anyone ever run across something like this , and how did it heat ?

Hot Rod , great pic of the actual heat transfer .

kevin

i suppose...

it would eventually heat up and it would act as a floor warming tool but w/o the insulation that basement would be the hot place...kpc

Boilerpro_3

Triad systems were something like this

Triad (still in business in Chicago) used elements in a radiant ceiling system in an attorney's home near heere built in the 50's. System apparently has worked well for all these years, right down to the original boiler. I fixed one leak in a supply line where a long ago driven nail finally wre its way thruogh the pipe and I beleive a line froze up in the attic once. Triad was really ahead of its time, right down to pumping away from the bottom of thier boilers in this home and others I've seen.

Boilerpro

hot_rod

Conduction trumps

convection and radiation. Some experts say conduction betters convection and radiation combined. I think finned element's "bottleneck" would be the absense of air current movment in a tight well sealed joist space. Hot water baseboard works by pulling cool air in the bottom, sending warm air up and creating a swirling air current in the room.

In my mind, contacting the surface with the best conductor, maybe copper and aluminum, is hard to beat.

The assembly closest to you in this shot is Warmboard with pex. Notic where I routered across the aluminum surface, in two spots, how it effects the heat transfer to break the "conduction" Can you pick out the pex/ plates, suspended pex, staple rubber assemblies? My testing indicated Warmboard at 110 supply equaled the IR picture of the suspended at 140.

Still, once the floor surface gets to the max of 83-85, that's as good as it gets. The floor doesn't really care what is driving it there, really. Over and over the results show conduction and conductors move the heat to the floor fastest, and with the lowest supply temperatures. Putting the conductor on the top of the subfloor takes the plywood, or whatever, R-value out of the equasion also.

Plenty of way to fry a fish, or floor

hot rod

To Learn More About This Contractor, Click Here to Visit Their Ad in "Find A Contractor"

Steve Ebels

I have

uninsulated fin tube under the floor of my kitchen, (one of many ongoing hydronic experiments in my house) it
DEFINITELY warms the floor. The contents of which are 3/4" plywood, 1/2" cement board and ceramic on top. When the stat opens the zone valve it takes about an hour to feel the floor warming. So, does it work? temp regulation wise? Poorly. Speed SLOW. Warmth and comfort? Not bad at all when the zone is running halfway steady. Basement hot? No, not at all. Of course, there is no other heat in the basement except heatloss from the boiler and piping.

Just for your information

[Deleted User]

Wow...

I didn't pick up on that during the first round. Thats pretty pronounced. Makes it about the same as with plates under carpet eh...

Great stuff. Post some more pictures and give us some more tours HR. I knew I was missing a lot.

ME

hot_rod

Every picture tells a story

You really need to study each shot for a while. The shots with the crosshairs help pinpoint temperatures across the panels. In every shot half of the 3/4" Advantech underlayment is covered with a thick berber carpet (no pad)

With a good close up shot you can actually see the fibers of the carpet warming.

I hope to make all the various shots (better quality, and closer views) available thru the RPA to all members. Somehow.

I think they would make excellent brochures, truck decals, add slicks, home show video clips, presentation to homebuilders, etc, etc. It truely shows first timers, pros, and homeowners the real gist of how radiant warms, and the various installation methods performance.

hot rod

To Learn More About This Contractor, Click Here to Visit Their Ad in "Find A Contractor"

Mike T., Swampeast MO

Photo of just that

(copper tube in fin) I did pre-make the sections for each bay with a "hook" to make it through the joist, but only had my little spring tubing bender so it's not particularly neat.

Tubing extends greatly beyond the plates as this floor (fully internal bath) shares a little reverse-return "mainfold" serving another bath (one exposure with 2 windows). [see other photo--don't blame me for top plate butchering--it was done LONG ago for warm air ducts rapidly abandoned for a gravity water system] Made the tube runs identical in length with same number of bends for good measure.

NO t-stat--just running off of constantly circulating mains. Ball valves at connections to mains for balancing. Operation STILL amazes me. The internal bath floor is left uninsulated and the exposed insulated with strips of styrofoam between then another layer over all.

Internal bath only has about 4' less plate BUT its delta-t is about 10° while that of the exposed bath is about 30°. Floor in internal bath runs only about 2° above the other but BOTH result in a room 5° above the temperature of the rooms surrounding! VERY low flow for these two baths and on cold startup delta-t is essentially infinite (return nearly ambient) for many hours.

GREAT STUFF HOT ROD!!! You don't have to give it all away, but would be curious if you'd tell me how that educated guess ranking I sent you came out. Looks like the "appear to run out at the edges" for warmboard was on the money.

Mike T., Swampeast MO

Hey Hotrod!

When you were first talking about this GREAT test you said something like, "...and, just for fun, copper in thermofin".

Did you try an egg on this example? ;>)

Mike T., Swampeast MO

From front to rear...

...warmboard, staple rubber, suspended pex, pex/plates

previn

wow

I guess as one of you guys put it "more than one way to fry a fish (floor)"
the verdict? copper out performs PEX, but one question arises, how is expansion of the copper handled? and does any one have a photo of a loop to manifold connection and a "bay jumping" shot?

nice IR photos too, hotrod

Mike T., Swampeast MO

Expansion

of copper in aluminum doesn't seem to be a problem. Of course you want to be extra careful to smooth/flare the ends of the fin. Have never heard a single noise from this system. Remember though that it uses constant circulation at low water temperature. Don't know if higher temp/digital control would cause problems, but again ZERO noise even at cold start.

My rather crude method was to fashion the bay loops as illustrated and use a coupling. 3/8" soft "L" tube by the way. "Snapping" in into the fin 11' above the floor was NOT fun...

If you're wondering about my weird pipe insulation it's improvised but seems effective. Split plain 1/2" wall pipe insulation and wrapped diagonally around mains/main fittings. Then wrapped this with 3-M arc-proofing "tape" (only sticky for the last 1" or so) that I got from a University surplus auction. As you see, I did have to use some duct tape--sorry.

Other photo is attachment to mains--was simple as used the old takeoffs to the iron panel rad that used to be in the exposed bath (the route lines to that rad took did SEVERE structural harm by the way).

p.s. will fill these perfect fire chases (PRE-EXISTING) with an appropriate product eventually but there's still some temporary wiring to remove, temporary lolly columns to replace with permanent steel, etc. Currently have about 40 adjustable posts (plus 2x6 frame walls that were the 2nd step of leveling/stabilizing) holding things up and have things raised just a bit higher than final state of "level" (remember it's a 100-year-old house) to allow the permanent steel to "slip in." Still have a VERY distorted doorframe in bearing wall of 1st floor to fix when I figure out how to do it without throwing things above out of their current balance. 1920s plumbers/tin hacks caused the vast majority of the structure problems. Floors are literally 2 1/2" thick and it takes LOTS of time for them to "adjust" after bearing points are leveled/reinforced!

hot_rod

Loop end ideas

I did this copper ceiling retro fit last fall. I used pex for the loop ends to allow expansion movement. You actually end up with the same amount of solder connections.

The best way to avoid expansion noise is to avoid wide and sudden temperature swings. Probably a very good application for constant circ or at least outdoor reset control. With these logics the ramp up and temperature extremes will be limited.

The grip of the plate to the tube is another concern. The tighter the better! I've found pex tube with the EVOH barrier on the outside actually make noise when allowed to move in the "looser fit plates" typically the thin flashing gauge ones.

I still feel PAP is the perfect match for limiting noise potiental, if your going plastic! No noisy external barrier, and the aluminum layer restricts the "growth" somewhat, as the tube heats. Furthermore the PAP's are a bit larger OD so the tight fit gets even tighter.

hot rod

To Learn More About This Contractor, Click Here to Visit Their Ad in "Find A Contractor"

hot_rod

The panels in the picture

Closest to you is the Warmboard. Next is copper in ThermoFin, then staple up Onix, suspended Wirsbo hePex, and the furthest is Watts Radiant EpexB in ThermoFin. As you can see the Onix staple up does benefit from some conductive transfer compared to suspended.

hot rod

To Learn More About This Contractor, Click Here to Visit Their Ad in "Find A Contractor"