Heat Transfer Plates?

DaveGateway

Why use heat transfer plates?

Let's take the increased profit potential, of the longer install time for heat transfer plates, off the table.

I see that you can lower the water temp. by 40-50* due to the increased heat transfer efficiency.
But, can't you get the same results by installing 2 or 3 times as much PEX per square foot? At a much lower cost?

Lets say, for example, that the heat transfer plates cost 5X as much per square foot as 1' of PEX. I don't see the cost/benefit in favor of the heat transfer plates!

I realize that more PEX = more manifols & higher head pumps.

BTW, What % of the aluminum surface of a heat transfer plate is in actual contact with what you are trying to heat? Think about it at the micro-level! I have read, in other threads here, that conduction is by far the primary mode of heat transfer. I suspect some "Fuzzy Phisics"

Thanks for any input BP

Mike T., Swampeast MO

"BTW, What % of the aluminum surface of a heat transfer plate is in actual contact with what you are trying to heat? Think about it at the micro-level! I have read, in other threads here, that conduction is by far the primary mode of heat transfer. I suspect some "Fuzzy Physics""

Granted there can't be perfect contact between plates and surfaces--if there were it would be like the non-magnetic stainless steel "bars" made for ultra-precise reference measures--they actually "stick" together! What I can say for certain though is that one plate will have MUCH more contact that even 6 cylinders contacting a plane! Remember that even this "line" will be far from perfect contact!

"But, can't you get the same results by installing 2 or 3 times as much PEX per square foot? At a much lower cost?"

Maybe you could, but say you've used plates on 8"± centers in a normal 14½" wide joist cavity. No biggie to bend most PEX. Triple that and you have about 2.3" centers. Are you going to use hundreds of expensive PEX fittings--you'll never bend to a radius even close to 1.15". The other option would be "interleaving" three runs of PEX in the same joist space--sounds like a nightmare to me!

Say you needed 10 loops in the structure when using plates. Triple that and you now have 30 loops. Not only do you need a manifold with more connections, but one with triple the flow capacity. Add the cost of the additional PEX, triple the circulating power, plus all of the added labor to install the tube and you'll probably find that the final cost is greater than using the finest, heaviest plates. Granted you can say, "that bare tube won't be giving off as much heat per foot as the tube in plates, so I can just make the loops longer." To a degree this is correct, but you're forever stuck with the electric bill from forcing circulation through a constipated system.

Alan(CaliforniaRadiant)Forbes

Radiant Heating

Good radiant heating is done at low water temperatures and in the case of staple-up, that means with plates.

Why?

1) The tubing will last longer.

2) Lower temperatures means less heat loss through the boiler, primary piping and tubing.

3) Fewer expansion and contraction noises.



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

Daryl

RE: Aluminum transfer plates

Buffalobill,

The use of transfer plates in under floor heating is (in my opinion)the only way to go. If you just put in more tubing per square foot your still relying on convection rather than conduction. The plates are designed to transfer the heat across the entire surface. With just the tubing, your heating the air space and trying to heat your living space via the hotter air under your flooring. Although the tubing is in direct contact with the floor, your not going to be able to get more than 8 inches on center with 1/2. Unless you plan on running several tubes down one bay, thus increasing your costs on manifolds ect. The btu's that are distributed with the plates and astro foil 5/8 of an inch under the plates yeilds 46 btu's per square foot. Please refer to www.radiantmax.com, under radiant heat transfer plates.

Sincerely,

Daryl High
National Sales Manager
Eagle Mountain Inc.

DaveGateway

Heat Transfer Plates?

Mike, you make some very good points! Thanks

I know that the PEX would have even less contact & thus less conduction of heat to the floor than heat transfer plates (HTP ok?). My point was that conduction plays a minor role compared with radiation & convection!

I don't see that interleaving 3X loops of PEX, either under the sub-floor or on top between sleepers, as a "nightmare"! I think that it would go quicker than HTP.
I'm not talking about Warmboard or RauPanel here.

I don't see a need for tight radius bends or extra fittings! Yes, more manifold connections, but I wouldn't exceed the 330' loop lenghth that Wirsbo allows for 1/2". Thus I don't see it as "constipated" (I like your choice of words, sounds like the way I talk).

Your point on increased pump power usage is valid!
But realize, that all of the energy used is converted into heat, which helps heat the building! Just at a higher cost per BTU.

Thanks BP

DaveGateway

Heat Transfer Plates?

Alan, Thanks for the input

3X PEX vs. HTP

I think that the required temps. would be very close.

PEX "early end of life" don't look like a problem, to me!

Doesn't PEX usually make less noise than HTPs?

Thanks BP

DaveGateway

Heat Transfer Plates?

Daryl, thanks for the input.

I'm not counting on direct contact, you don't need it!

I'm designing for only 20btuh/ft2. Wouldn't 46 btuh/ft2 give a floor surface temp. of about 93*??? & cook TJI's & subfloor etc.??

Thanks

Mike T., Swampeast MO

Convection/Conduction/Radiation

"HTP" is kind of taken already for "Heat Transfer Products" aka maker of the Munchkin Boiler, but as long as you state your specific alphabet soup each time I guess it doesn't matter...

Remember that convection is a form of conduction--it is conduction that occurs more rapidly than it "should" because of physical movement in the medium. Because air is so diffuse, convection plays a very significant role when it is heated.

Bare tube convects with natural, free-air convection in the entire joist cavity. While the contact between a conduction plate and the floor certainly isn't perfect, it's quite close and the convection is mainly in the tiny cavities between the plate and the floor. Temperature difference is the driving force of natural convection--those tiny areas heat quite evenly with little difference to drive convection. The distinction between "conduction" and "convection" blurs...

This is one of the reasons you want to insulate DIRECTLY against the plates--you're trying to keep those tiny air gaps as isolated from the rest of the space as possible.

With bare tube however, you don't want to insulate making such a sandwich. Why? Because you WANT to introduce temperature difference in the joist space to encourage air movement to encourage heat to leave the tube and enter the floor.

DaveGateway

Heat Transfer Plates?

Mike, thanks for the input! I have a HTP Voyager in my home.

There is a very good description of the 3 modes of heat transfer, starting on pg. 6 in John Siegenthalers' book:
Modern Hydronic Heating.

I don't agree with your comment: "While the contact between a conduction plate and the floor certainly isn't perfect, it's quite close"

They are "quite close" but a very small % is in actual contact! Think in terms of a microscopic view of the 2 surfaces! The sub-floor would be quite uneven!

This reminds me of the "outdated" idea that staple up jobs must have "alot" of staples, to keep it in contact with the sub-floor.

It seems to me, that an air gap of several inches or more below, with a radiant reflective surface, would beneficial to either type! What is your source of data, that would support having the insulation up against the HTP?

Thanks, got to go get a green beer!

heatboy

Fuzzy Physics?

Please elaborate. From someone who has installed miles of extruded aluminum heat transfer plates above and below the sub-floor, I can assure you they are, undoubtedly, the premier heat transfer medium available.

http://climatecadvanced.com/news/RPA-2002-10.pdf

Of course, I'm an eletist and my views are just that. Mine ;-)

Warm Regards,

hb



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

hr

It's the conduction!

Try this, fill a sause pan with water hold it 2" above an electric burner. See how long it takes to boil.

Then try it with the pan on the burner. The conduction transfer is much, much more powerfull.

Sane with the radiant floor, wall, or ceiling. As a result you get quicker response, better heat spread (less striping) and equal output with lower water temperature.

It's been calculated, modeled, experienced first hand, and here are some color IR photos to show the difference after a 6 hour run time. The 3 methods shown are staple up EPDM, pex suspended below 2", and ThermoFin heavy aluminun extruded plates with pex.

hot rod

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

hr

It's conduction

a much stronger heat transfer method than convection or radiation. In fact stronger than convection and conduction combined!

Try this, hold a sauce pan of water 2" above an electric burner. See how long to boiling. Then try the same pan and water sitting on the burner. Enough said.

The difference in heat transfer has been calculated, modeled, photographed, and felt first hand by radiant fans

Here are three examples I built and videoed over a 6 hour time period.

See if you can pick out the EPDM stapled to plywood. Notice the strong conduction forces at work where the metal staples tightly contact the tube wall!

The pex suspended 2" below the plywood.

And the pex in heavy gauge extruded ThermoFin transfer plates.

These photos are taken at the end of a 6 hours run time. Same supply temperature, same tube length, and same gpm flow rate.

hot rod

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

Joel_3

Hot rod!!!

that's cool no it's hot!!! you've got an infared video! Way to go!! Nice to be able to see what we already suspected. Bill running more and more pex would not be cheaper than using plates not by the time you fought with the tube to get it in there on tight runs and had to drill lots more holes. Plates go up pretty quick once you get used to it. I wouldn't worry so much about getting the cost down, just biuld the best system you can and charge accordingly.Upfront cost really shouldn't be a concern as opposed to lifetime cost. I'm trying to figure out what to do with a house done without plates 3 years ago. It doesn't work, and it's not going to work unless we tear all the insulation down and add plates and repipe the boiler. Fixing the mess is going to run probably double what they payed to have it installed the first time, skimping is just not worth it! "biuld it and they will come"

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

bill clinton_3

For what it's worth

I've always had a blind unreasonable prejudice against plates: Just born that way I guess. I have the following to contribute:
1. Can't deny plates (good extruded tight fitting ones at any rate) give superior heat transfer at any given temperature. It's true that the fit isn't perfect, so there is a layer of air most places at the interface of plate and wood, so most of the heat transfer, plate to wood doe happen by conduction (the space being too tight for convection) The trick is that resistance to heat flow is proportional to distance the heat must be conducted--in this case measured in thousandths of an inch, so that air layer doesn't slow things up much. Since the plate doesn't cover all the exposed underfloor, some convection would also occur. (Although tight insulation would put the kybosh on that.)

2. I have tried increasing the amount of suspended tube. Late in the game, as installation was starting, the customer came wanting absolute assurance his very leaky house would make his warmth loving wife happy. So I halved the spacing--down to 4" centers. They got through what in these parts passes for a cold winter (design temp=38 degrees) and were very happy. Customer promises he will install the insulation below the tube sometime before next winter! Installation was quite easy: Had to be easier than plates tho as I said above, I'm not experienced in that area.
3. Both pex and plates can be noisy. Mostly it's the pex with the EVOH oxygen barrier which tends to be "tacky" and will drive you nuts if it's laid so two tubes press against each other. Separate them with a piece of insulation and use plastic isolators where pex goes through wood and full circle plastic hangers where you attach it. Listen to the other guys around here re plates.

4. I don't have exact figures, but my very subjective guess is that tube on 4" centers operating with 150 degree water will probably match plates at 115 or so degrees. If that satisfies your needs, I'd say go for it.

By the way, I've been doing suspended tube for a dozen or so years and found that, in situations demanding something around 15 btu/square foot, suspended tube on eight inch centers running 150 derees does a nice job EXCEPT if theres a nice thick carpet and pad above it. THAT would call for doubling up!

Bill

Mike Kraft_2

Hot Rod/Fred McMurray/ Flubber

Great pics HR.What was the water temp?Did you do a comparison "shoot" to design for the same surface temps.?

Thanks

cheese

hr

Show us the numbers, Bill

without them you are guessing. I disagree adding additional tube increases the output on suspended. It's the temperature in the joist space, and the floor covering that limits the surface temperature. Even if you could warm that cavity to 180°, which makes all the wood manufactures squirm. The limit becomes how hot you need to get the "air" in that space, to transfer through the R of the wood subfloor and coverings. I think.

Increasing supply temperature would increase output, not, to my thinking, adding more tube at the same temperature??

For me, no one has done more actual research, testing, modeling, FEA analysis, and actual operation comparison, than the boys at Radiant Engineering.

They have the docs to back up the claims. That's the difference.

I'm not quite clear on the contact area either. New Advantech type subfloors are very very flat and smooth. I feel a lot more conduction contact than air film is available??

I agree, however there is a place for suspended tube, and in California mild climates I suspect you can meet the loads often. Seems that 12- 15 BTU/ hr number is believable Although still at higher temperature supply.

A lot of issues that come across this list on a daily, or weekly basis could be cleared up if the other product and installation method promoterd would "show us the math" to back up the claims. I think you know of whom I speak

hot rod

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

cruizer

very easy.....

Running with plates means lower water temps. Lower water temps mean less fuel used. Less fuel used means less money sent to the gas company. Less money sent to the gas company means happier customers.

Higher supply temps are less efficient, why would you want to sell less efficiency to your customer?

JOTI

HOT, HOT, WATER

I have to agree with using the plates in most of my designs and those pictures from hot rod are great visuals. Ihave heard of a company that runs 180 degree water with less tubing running across the joists, you can look under ULTRA-FIN by McDuffco on the web. I think the idea is a little far fetched but who knows,without wierd ideas where would the fun be.

hr

When

I used to do rubber installs we always used counterflow piping. It was so easy to do! Pull the entire loop length off the coil, connect both ends to the S&R manifold and string the hose with a supply next to the return on every pass.

This, I felt, was the key to the nice even floor surface temperatures that the old Solaroll systems provided. In that application the supply tube was "connected" to the return in the make up of the material. The eveness of heat was unsurpassed. The warmest supply temperatures were bumped against the coolest return temperatures! Floor temperatures were amazingly consistent.

If the delta T drives the output, then the loops along the outside walls will produce the higher output, as they see the coldest temperatures in the structure.

So in theory you could pack all the needed tube into a 2 foot wide band along the outside walls and heat the space, just fine.

There is very little loss, or need for tubing, in the very center of the room if the load is "covered" in the perimeter. However in terms of comfort to the foot the tubing spread across the entire "radiant panel" is more desireable.

Then again a unit heater hung in the corner could handle the load, also.

It's the fine, or maybe coarse, balance between just plain heat, and ultimate comfort, I suppose And to what extreme you want to chase the "perfect 10"

hot rod

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

bill clinton_3

Numbers? we don't got no stinkin numbers!

Don't be so tough on me Hot Rod. Thought I made myself pretty clear that I was doing seat of the pants guessing. I would like to someday do the work of getting real numbers, but haven't yet: just stating my experience over a dozen years of cautiously experimenting with suspended tube and the sense of the thing derived from that, AND, I thought, putting plenty of qualifiers on my statement.

I will argue with you on one thing: Doubling the tube certainly does increase output simply because it doubles surface area (emphatically it does NOT double output, maybe something like 20% and yes this is a guess). Increasing surface area is, after all, the rationale behind
using plates: That's their function. So long as there is a temperature differential between a surface and an ambient, increased area will give increased heat transfer.

I was really disappointed with that ASHRAE sponsored study showing comparative performance of various methods at a given temperature and concluding plates worked better than naked tube. We all knew that already. The strategy with the naked tube is to run higher temperatures (within reason--my self-imposed limit is 160 degrees.) As you point out, we need numbers.

As an aside, and offering a club for suspended-tube critics to bash me with; I am becoming alarmed at the cavalier use of naked suspended tube. I am now getting calls from homeowners and general contractors specifying that's what they want. No analysis has been done; no caveats have been issued; inexperienced and untrained plumbers are doing it with no caution at all. Scary!

OK, you can hit me now.



Bill

Mike T., Swampeast MO

"What is your source of data, that would support having the insulation up against the HTP?"

Engineering data supplied by the makers of nearly every such product and advice given by those who have installed 1000s upon 1000s of feet of heavy, extruded plates.

My "super-conductor" plate experiment--copper tube in Thermofin type "C". Insulated with rigid foam both between the "Cs" and on top of the entire assembly. This results in nearly infinite delta-t (return nearly ambient temp) for many hours on cold startup with the return temp climbing as the system nears stasis with its environment. Is there convection in this transfer sytem? Yes, but not in the manner that you are talking about.

Lurker_2

re

What about the necessity of trnasfer paltes on an above the subfloor system? Had planned to run the tube between sleepers.

cruizer

Same deal. No striping, the heat is more evenly spread across the floor. Lower water temperature. Etc, etc....

jerry scharf

you invited people to be tough on you

> BTW, What %

> of the aluminum surface of a heat transfer plate

> is in actual contact with what you are trying to

> heat? Think about it at the micro-level! I have

> read, in other threads here, that conduction is

> by far the primary mode of heat transfer. I

> suspect some "Fuzzy Phisics"

>


I think people are pushing back on you for this statement. You have put up nothing but your belief, and others have used given clear evidence and documentation to the contrary. You can either accept it or not, but don't expect people to be light on you when you call their work "fuzzy phisics."

jerry

Radiant Wizard

Just look at

Dan's book on Pumping Away. We have all been through this argument time and time again.

Reason why - Lower Water Temps, Quicker Response Time, A more even floor surface temp. These are the only 3 reasons for using plates.

This question is like asking should the circulator go on the supply or return? The system works both ways but one is better than the other.

hr

Bill

I thought I had included enough smiley faces to indicate I was just yanking your chain. Not really coming down hard

Keep in mind when talking extra transfer surface area the aluminun (a conductor) as well as the thickness of it, plays into the heat transfer ability.

Or else we would have transfer plates made out of pex!

The thickness of the aluminum, the alloy type, and the tube to plate fit are all part of the transfer "connection" Thin aluminum, with lose fit and poor connection to the subfloor due to twisting under temperature, are a far cry from heavy extruded plates!

Or thin transfer plates that can only be fastened on one side. What a concept, and marketing ploy that was! Transfer plates that are not connected to the subfloor. Sounds more like a pex "bracket"

hot rod

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

DaveGateway

Heat Transfer Plates?

Jerry,

I fully expected these responses!

BTW, I didn't call anyones work: "fuzzy physics"

What I tried to say was: that "conduction" does not move heat across an air gap, however small, Please look up the definition of "conduction". And that: a flat plate of aluminum will make only a small % of actual intimate contact with the rough surface of a subfloor!

Therefore, I still believe that the primary mode of the thermal energy movement from a heat transfer plate to the subfloor is Radiation & convection!

"Physics has little regard for opinions"

Thanks for the input, BP

hr

Cheese

I chose 130° for ther supply temperature to all. Would have liked to use 115°, but output on suspended or staple up gets a little harder to see and compare at that low number..

130° on bare plywood like this is way beyond floor surface comfort range, actually.

In the video of all these applications there is a temperature window in the picture to indicate what the various colors show. This makes comparison a bit easier. Yellow and white, for instance are hotter than the red and darker colors. For some reason in the translation from video to still pics the temperature gauge disappered

hot rod

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

DaveGateway

Heat Transfer Plates?

Heatboy,

I read your reference, and am underwhelmed!
Seems like a statement of opinions, to me!

I have no doubt, that high quality HTPs are the premier
"dry" heat transfer medium available! I never said that they wern't!

I want to get some discussion on is the "bang for the Buck"

Thanks for the input, BP

DaveGateway

Heat Transfer Plates?

Hot Rod,
I know what you meant. Conduction is the strongest method of heat transfer.

For "conduction" to occur, there can be no air gap, period!
If there is ANY space between 2 materials, heat transfers via radiation or convection, whichever is easiest! to cross the gap!

This morning I heated my coffee pot by holding in the blue flame of my LP stove. It also heated up quite quickly!
With 0.00% conduction. Look up the definitions.

Great pics. Each is worth 10,000 words! It is easy to tell which is which! The HTPs are obviously much better at moving the heat to the subfloor above them. What I also find striking is how little lateral movement there is in the subfloor itself!

Now, could you re-run the experiment as follows:
Take your total cost per sq.ft. for the HTP + PEX case.
Divide that cost by 2.
Put in the number of runs of PEX those $$ would pay for.
I suspect that the PEX would be about 1" OC.
How do you think the IR pic. heat distribution would look?

Thanks for the input, BP

DaveGateway

Heat Transfer Plates?

Keith, are you dyslexic? (:
I am planning the same thing. 2x4 sleepers 12"OC on top of the subfloor with 6 runs of PEX between each. So they would average 2" OC. across the rooms. With a hardwood floor above & R13 with a reflective layer on top, below the subfloor, with a 4" air gap. I think that it would deliver my design 20btuh/ft2 just fine at 1/3 to 1/2 the cost of high quality heat transfer plates & PEX 8" OC!
Thanks BP

bill clinton_3

sleepers/pex

I think you'll hit your output target easily. No, I ain't got no numbers Be careful about nailing: Even well placed nails can follow the grain of the wood and squirt out the side of your sleeper causing you know what.

Also be careful how you lay that pex: if it's barrier tube and it winds up rubbing against itself, it can be noisy. If possible, run hot water through as a test before nailing hardwood.

Bill

DaveGateway

Heat Transfer Plates?

Hot Rod,

IF you coated the HTP with a thin coat of silicone sealant, before screwing them into place, then I would agree that the primary method of heat transfer to the subfloor was conduction! I don't know if heat transfer would get better, stay the same, or get worse, though!

Thanks BP