I have been getting quotes for a hydronic system and I am concerned about the gauge of the aluminum heat plates provided with one system. They are so thin that they are affixed under the subfloor with a stapler. Seems pretty flimsy to me. I thought the idea was to disperse the heat not wrap it like a baked potato. Any thoughts?
Edited 9/30/2009 6:54 pm ET by Stex
Replies
Radientecs plates are about the guage of flashing. I have used them in my house and on a couple of jobs, and can say that they work perfectly fine. Don't know if the heavy plates that others sell may or may not be better, but they sure are more expensive.
These plates are really just radiator fins. Have you ever seen how thin the fins are on baseboard radiators, or AC condensers, or even in your car radiator? They don't need to be thick to work properly. The idea is not for them to retain heat (requiring a lot of mass), but to radiate the heat (requiring less mass, but a larger surface area). The staple-up fins are fine.
Mike Hennessy
Pittsburgh, PA
Everything fits, until you put glue on it.
The only advantage the heavier plates have is needing fewer fasteners to hang them.
Welcome to the
Taunton University of Knowledge FHB Campus at Breaktime.
where ...
Excellence is its own reward!
this is completely wrong. Heavy plates outperform lightweights in several areas, including water temperature reductions/output increases, and noise reduction.Heavy plates will very significantly outperform lightweight plates in a joist application. In most cases, the water temperature differential between the two, to achieve the same output, is more than 20 degrees. that's 20 degrees cooler you can run with heavier plates. it is more like 30 degrees typical compared to plateless applications.there are people on the web who claim to have done testing that say this is wrong. They are wrong. the performance differential between heavy and lightweight plates is well known in the industry and supported by thousands of active projects in the field. I'm not necessarily maligning the people doing the "testing", but energy flow testing is very difficult to do well, and they are not doing it well, nor are they reporting very meaningful results. Some big names in our industry have made similar mistakes in their analysis before, so they are in good company, but it does not make them right. It makes them unfortunately wrong to a very wide and largely uneducated audience.We typically advise either heavy plates or no plates depending on the heat load of the space and floor coverings. lightweights in a joist only get you about 10 degrees over plateless in most cases, and really poor fitting plates out there probably don't even get you that. and then with lightweights you need to worry more about noise issues (well, if the plates FIT at least), and switch over to a PEX-AL-PEX which is harder to work with in a joist bay.-------------------------------------
-=Northeast Radiant Technology=-
Radiant Design, Consultation, Parts Supply
http://www.NRTradiant.com
I know this is your ar4ea of expertise so I will accept what you say. I was repeating what I've heard from local 'experts' who have only heard it from others without first hand knowledge. That is on the heat transfer.But on noise, my experience is that noise comes from the pex sliding through the plates as it expands. A plate that is well fastened is not going to make noise itself, and a lightweight plate is more likely to get fastened well, because it needs to be
Welcome to the Taunton University of Knowledge FHB Campus at Breaktime. where ... Excellence is its own reward!
It is certainly possible to make noiseless light plate systems, but it's much harder than with heavy plate systems, which are much more forgiving in that respect.-------------------------------------
-=Northeast Radiant Technology=-
Radiant Design, Consultation, Parts Supply
http://www.NRTradiant.com
Mr. Hennessey is right ... don't need much metal MASS, just a way to transfer the heat quickly away from the tubing ... actually the lighter the better in that respect. No need for 'beefy' .... it's just a waste of money and materials.
this is also wrong. metal thickness determines volume of heat transfer capacity. solid contact with the pipe determines how effective the conduction is.light plates restrict heat flow just like small wires can only conduct small amounts of electricity and small pipes can only carry small amounts of heat.-------------------------------------
-=Northeast Radiant Technology=-
Radiant Design, Consultation, Parts Supply
http://www.NRTradiant.com
"volume of heat transfer capacity"I'm sitting here pondering this thing and the physics of it.If that phrase is the essence of the arguement, it seems to me that in the real world, capacity is not 10% as important as actual delivery. It is the amt of sq ft of surface in contact with the element being warmed that determines heat transfer, not what it is possible to carry thru the plate, right. So how much does it matter if the thicker plate can carry more btu/hour to it's outer edges, if the same amt of surface is in contact with the subfloor above.Not trying top argue, but to understand. I'm sure I could be missing something here.
Welcome to the Taunton University of Knowledge FHB Campus at Breaktime. where ... Excellence is its own reward!
imagine we had a 1/4" pipe to heat with.the ability of that pipe, to transfer heat, is not that much different than 3/8", IF the pipe is all a uniform temp.what temp the pipe actually would be in a given heat flow situation is changed, however, by the reduced ability of the pipe to carry flow. IF you needed a flow greater than what the 1/4" could deliver to maintain your temperature across the pipe, then upsizing the pipe would help fix that.this is, to my best understanding, the situation we are in with heavy plates. also, beyond that the heavy plates have vastly superior direct tubing contact which enhances heat transfer significantly.I have not done the plate engineering myself, so this is interpreted knowledge, not direct knowledge: but I do know what the hundreds of each type of job I have out there can do and the fact is, heavy plates are very superior to lightweights in most joist applications. so even if I have a detail or two a little off, I'm still very confident in calling anyone who claims that heavy plates are not significantly better than lightweights in *most cases* wrong.-------------------------------------
-=Northeast Radiant Technology=-
Radiant Design, Consultation, Parts Supply
http://www.NRTradiant.com
Rob, a related question that might shed more light: We have a client willing to pay a bunch of money to add cast iron baseboard radiators to her house. Our HVAC guy thinks it's silly, and says that the heat output (BTU's) are the same for cast iron as for fin tube. The cast iron certainly looks better, but is this an analogous situation?
I've seen guys install those heavy plates, BTW, and it looks like a LOT more work than staple-up. With an air space, a good air seal and plenty of R-value below the tubes, is the heat transfer capacity really that important? Doesn't the whole thing work as a system?
not using plates means you max out at about 20 BTUs/sq ft with something like 165 degree water under a wood floor.Using heavy plates means you can do that much output under wood with a 120 degree water temperature. Yes, that's a 45 degree water temperature differentialThat's a typical output requirement for a home. that's the difference between
-condensing and not condensing on a condensing boiler:
-mixing and not mixing for a basement slab on the same system:
-keeping up and not keeping up if you have a higher load than that.
-being in range for geothermal/heat pump systems and not being in range for slightly lower loads.
-response time is faster if you care about such things. t
-joist cavity temperatures differences that can really effect the loss out of the rim joists and/or downward.
-the ability to dump the heat is important for cycling control of the boiler in many cases... heavy plates win again by dumping the heat instead of sending it back to the boiler, which shuts it down faster.
-The temperature you are running at, on a conventional boiler, also affects what you can do for boiler cycles and the low temp method here can run the boiler much more efficiently, with less cycles, even if dispersion were the same. this is because you can draw the boiler down to a very low temperature before firing it again, then fire it longer... essentially, you use the boiler as a buffer tank.In short, it matters in almost all cases, and it matters quite a lot if efficiency is a goal and if actually heating a space is important. it is critical that a system have completely redundant backup, or the numbers are run so hard they scream if a plateless system is done.. and most of the time, it's being done to save money, so neither of those things are true. It's USUALLY the sign of a hack system. and lightweights less than about 5 years old would say about the same thing in a joist cavity (heavyweights haven't been around forever).for your question, CIBB is mostly radiant transfer, and Convective baseboard is mostly convection. so CIBB is usually more comfortable and it should stratify less. but it won't necessarily use a much lower water temperature: it can, perhaps a bit, but you can size either one to get water temperatures down to a lower range, if you're willing to put more in. aesthetics matter too. Personally I like radiant ceiling for cheap, radiant, low temperature systems.-------------------------------------
-=Northeast Radiant Technology=-
Radiant Design, Consultation, Parts Supply
http://www.NRTradiant.com
Thanks!
"I've seen guys install those heavy plates, BTW, and it looks like a LOT more work than staple-up. "I have done both and I gaurantee you it is more work to install the heavy I meqn both heavy and light plates - never done the sealed tubing stapled only, though I have seen it.
Â
Welcome to the
Taunton University of Knowledge FHB Campus at Breaktime.
 where ...
Excellence is its own reward!
Edited 10/2/2009 2:57 pm ET by Piffin
I have several contractor clients who would disagree with you on that as well. If you use a heavy stapler to put up the heavy plates and a palm hammer to put the tubing in, they are typically seen as easier than lightweights: you don't have to fight the tubing and the plate at the same time.some guys have sophisticated light plate techniques so I wouldn't use this as a selling point or an arguement for heavy plates, but neither is it a slam dunk that heavy plates are harder than lightweights."staple up" tubing should NEVER be done... staples can chew through the pipe over time. suspended tube with roomy clamps to allow for tubing movement need to be screwed in place, which is slower than plate installs of any kind I believe.-------------------------------------
-=Northeast Radiant Technology=-
Radiant Design, Consultation, Parts Supply
http://www.NRTradiant.com
But we are not talking 'high voltage' or amperage here. So I disagree. This is low temp heat transfer. Your theory ... if I had 1/2" plates it would work better ... not the case. Too much mass will slow the heat transfer. Low mass transfers the heat faster.
I suppose there is a science balance here where we are both wrong, but I suspect that light plates move the low temp heat quicker to the end of the plate because of the low mass. Once heavy plates are warm, they will transfer fine, but there is no need to have the expense of heavy plates.
Don't really want to argue the physics, (NRTRob is right), but the heavy plates from wirsbo/Upnor, are quicker and easier to install. The plates go up first and then the pipe snaps into the plate. They also have better contact between the pipes and the plates. And better contact between the plates and the floors.
And a reduction of water temps by 20 degrees is about a 7% savings in energy costs. (Every 3 degrees equals 1%)
Absolutely right on.....anyone who actually works with this stuff on a regular basis would know that!
"better contact between the plates and the floors."I'd have to agree with that point, which is probably the most important in my hiking.
Welcome to the Taunton University of Knowledge FHB Campus at Breaktime. where ... Excellence is its own reward!
The point you are making here is more relevant. Quicker and easier to install and better contact translates to better performance. Although I used fairly thin plates on mine (stapable?) and I think I got good contact between pipe and plate. It's not like the plate was deformed enough to not ensure good continuous contact.
Edited 10/2/2009 7:27 am ET by Clewless1
I apologize for not sugar coating this, but you are totally wrong. As in, you're not even in the ballpark of being correct... this isn't even left field. mass does not "slow heat transfer". you transfer heat to the same rate, given the same conductive properties, from pipe to plate, and plate to floor, no matter what the mass is... it only depends on conductivity and temperature differential.the time it takes to heat up the mass changes, true, but heat transfer rate does not. If I have 1" of concrete, and 10" of concrete, with the same pipe in it, I can transfer the same amount of heat into both at the same speed... the amount of mass didn't change that, and it is heat transfer rate that determines your water temperature requirements for a given application. all mass does is time shift CHANGES in heat transfer rate... it has to cool down to slow down the transfer rate, and it takes time to cool down a mass, and vice versa, because of heat storage capacity. but you have not changed your RATE of transfer, just how fast you change your rate of transfer. in other words, you might add 20 minutes to the time it takes to get up to your peak transfer rate, but once you are there, mass doesn't matter at all anymore. The plate could have 500 pounds of mass... that would have no affect on its ability to deliver at low temperatures. If that were not true, concrete systems would require the highest water temperatures of all... and instead, they are among the lowest. why? because they offer decent conductivity, good heat flow characteristics and excellent conductive contact.note the total lack of the word "mass" in that description. You confuse temperature with energy and mass with conduction. and in doing so manage to ignore an entire industry's direct experience with a particular product while you are at it.the fact is heavyweights outclass lightweights. but there is a limit to what is helpful and going up to 1/2" would be very expensive and wouldn't get you much more gain... it's not linear improvement with thickness.-------------------------------------
-=Northeast Radiant Technology=-
Radiant Design, Consultation, Parts Supply
http://www.NRTradiant.com
You don't have to apologize ... we are simply having a discussion and an exchange of points of view.
I understand what you are saying ... the rate of transfer is the same for the same material. I think that was my point, though ... as the system CHANGES, heavier mass is slower to respond. If the system is constant (which generally it would be in an IDEAL situation), there is little change and the two applications are equal ... other than as was said ... better contact means better distribution of heat transfer.
If I try to heat up 10" of concrete, it will receive heat at the same rate as 1", but to get that heat spread out ... or transferred to another place, it will take time. If the concrete is already up to temp, that's different. I was speaking of the dynamics of a situation and you were referring to a static (steady state) condition.
However, the mass of our two products relative to their conductivity is probably small and so either way it may not be much of an issue.
Assuming a fairly steady state condition, wouldn't a poorly contacted plate be similar to a well contacted plate? Thinking out loud here. If the pipe is transferring energy to a plate, but through some small air spaces (that 'insulate'), but air spaces that are very close to the pipe temp, the plate will also be that temp. But no, just like pipe insulation, the air gaps aren't good for heat transfer as well as it would be with good contact. But are we talking really a small impact in this kind of situation? That is the detailed science of radiant plates that I admitedly don't know much about.
But also, I think you had made the point that it was more the better contact with the heavier plate that gave it the performance/ability to distribute heat, not the mass of the plate. Assuming good contact with a lighter plate, you should get the same (or similar) performance (i.e. distribution), right?
Ideally, a radiant floor won't be too dynamic ... it should chug along providing just the right amount of heat for the load. The controls would adjust the water temp just enough that the circ pump would operate continuously. This doesn't really happen, though as most systems aren't that sophisticate and the dynamics of the load can be too quick (e.g. when the salesman comes to the door unexpectedly and talks your leg off or a cloud goes in front of the sun and shuts down your passive solar gain).
I wasn't really arguing that massive plates are worse than lighter plates. The science says they have to be slower to respond to dynamics, but that may be a small issue in the context of our discussion. The real point of masive plates is a better fit for better heat transfer (at least that was what I was getting from the discussion) ... which has nothing to do with the fact that they have more mass ... although the mass provides the property to be more rigid and thereby more consistent in it's contact w/ the tubing. I know little or nothing about the crossectional profile of the heavier plate. If it isn't designed to hug the tubing continuously around the circumference and along the length, then they may be no better than lighter plates (from a heat transfer standpoint).
This isn't much of an exact science anyway, is it? The heat transfer plates aren't installed continuously anyway, so there is always a large variation in heat transfer from one end of the tubing to the other.
here is a pic of joist trak. We use the double ones, two tracks per plate, less work installing.
the difference between good tubing contact and poor tubing contact is significant. AND the ability of the metal to transfer heat to the edges in sufficient quantity is also significant. Heavier gauge plates... up to a point, though it's definitely a very significant point at the levels we're discussing with currently available plates... beat lightweights significantly in both respects. You're right, it's not that the heavy plate has more mass that is important: it's that it's thicker and can conduct more heat as a result, again, just like bigger wires and bigger pipes can transmit more heat from one end to the other. bigger wires and bigger pipes also have more mass, but that isn't the reason they conduct more heat. imagine you were trying to conduct heat between two metal globes. would a wire be just as effective as a thick bar of metal? the answer is no, and the reason is that the thickness of the conductive medium, in conjunction with its actual conductivity, is what determines heat transfer. thin media would be a "bottleneck".'Dynamics' we call "response time" which is a separate issue. at some point, mass would increase response time, but in this case, heavy plates decrease response time because the difference in plate mass from that standpoint is negligible: the increased heat transfer to the subfloor vastly offsets it so heavy plates allow a joist system to react quicker than light plates and much, much quicker than no plates because they allow much greater heat transfer to the floor assembly. I don't regard that as important because a properly controlled system as you note shouldn't have to worry about response time, in most cases, but thems the facts. Steady state performance is what primarily determines the required water temperatures for a system to operate effectively. In that case, everything that slow downs heat transfer (not rate of change of heat transfer to the room) hurts your water temperature requirements. bad tubing contact, additional layers of resistance, thin conductive paths, all hurt you.I don't know what you would call an "exact science". It's very clear that heavy gauge plates perform significantly better than lightweights and wildly better than no plates at all, within any range of normal variation in a radiant project. These are not small differences, these are big, noticeable, efficiency and effectiveness changing differences. you will not find a light plate joist system that outperforms a heavy plate system in the same envelope. though, in very low load situations, the differences between all methods shrinks.-------------------------------------
-=Northeast Radiant Technology=-
Radiant Design, Consultation, Parts Supply
http://www.NRTradiant.com
it's not that the heavy plate has more mass that is important: it's that it's thicker and can conduct more heat as a result, again, just like bigger wires and bigger pipes can transmit more heat from one end to the other. bigger wires and bigger pipes also have more mass, but that isn't the reason they conduct more heat. imagine you were trying to conduct heat between two metal globes. would a wire be just as effective as a thick bar of metal? the answer is no, and the reason is that the thickness of the conductive medium, in conjunction with its actual conductivity, is what determines heat transfer. thin media would be a "bottleneck".
Bigger provides the capacity to transfer more heat, but at these low temps, I question that we need that huge transfer of heat. A large bar between two globes will transfer more heat, but if I don't need to transfer but a little heat, I question that the bigger bar wouldn't actually be a deterrant. Do huge wires transfer e.g. 12 volts and .5 amps as well as a relatively fine wire? I'm guessing that it doesn't.
But we aren't talking large differences, but I am talking about your own logic. It would be different if it was e.g. steam heat or high temp H2O. I operate my radiant floors at around 110 degF just fine. I'm using the thin plates.
I'm not arguing that the heavier plates aren't better, just not necessarily within the logic you are using.
and again, at very low load the difference between all methods is small. If you're running 110 with light plates, you could have run NO plates at about 120 and heavy plates at about 95, assuming equal temperature drops in all cases and whatever flow adjustments you'd need to make that happen.. that's the range we're talking about. if you were running a geo system, then getting to 95 would be worthwhile. If you're running a mod/con, you might as well have skipped plates. Those are the kinds of decisions this kind of thing decides: but you're talking like the difference is negligible, and it's not, even with low water temps. It's very significant, whether you think so or not.at approximately 12BTUs/sq ft max load (assuming your water temp is design day, and not today) I would have questioned doing radiant floor at all unless you are over a cold space. at any rate, I don't know what to tell you. the volume of conductivity through the plate is important. You can refuse to believe me if you like, since I'm not going to bust out a lengthy calculus session here to prove it. but the fact is it's about more than just tubing contact.-------------------------------------
-=Northeast Radiant Technology=-
Radiant Design, Consultation, Parts Supply
http://www.NRTradiant.com
Now I apologize. I was simply trying to think about some science about a topic I know little about even though I did install a radiant system in my own house. I didn't mean to imply I thought the difference was negligible. I think I was trying to say that I thought the performance of the heavier plate may be due more to the design of the plate's ability to make a better connection than the shear mass of the material itself. I may be wrong.
I've never studied the detailed science of one plate type vs the other in terms of heat transfer. I used the thin plates, but I was installing on top of the floor sheathing, not under. A manufacturer went to a lot of trouble making a cadillac product, I suspect they may know more about performance differences than I.
Try this. Thick plates move more btu's. It's the btu's being transferred from the boiler to the space that counts.
So, rule of thumb design temps are fin tube convectors 180, cast iron rad 160, staple up 140, thin plate 120, thick plate 110 and warm board 100. (temps approximate)
Ignoring first costs, the important thing to remember is for every 3 degrees you can drop your water temp, you save 1% in energy costs.
Edited 10/4/2009 11:22 pm ET by rich1
Edited 10/4/2009 11:23 pm ET by rich1
Sort of hinted at, but not expressed exactly in prior posts, is the "fin efficiency" associated with extended surfaces attached to a pipe through which a heat transfer fluid flows. IIRC, this is the area of the extended surface (fin) that would deliver the same heat transfer rate (BTU/hr/degree F) if the whole fin were at the temperature of the tube wall, divided by the actual fin area. It's either that or the actual heat transfer rate divided by what it would be if the fin surface temperature were uniformly at the tube wall temperature.As Rob pointed out in #20, the temperature of the extended surface falls off with distance from the tube, as it loses heat to the surroundings (floor above). There must be a temperature gradient, or no heat would flow outward through the fin.With more heat capacity, or greater thermal conductivity in the fin relative to that of the floor above, fin efficiency goes up because the temperature drop outward from the tube is decreased. However, as has been mentioned, a heat distribution system with more heat capacity (concrete, thicker plates) takes longer to heat up. Thicker finned surface attached to the tubes means an initial greater temperature drop in fluid temperature as it moves along the path through the tube.
With more heat capacity, or greater thermal conductivity in the fin relative to that of the floor above, fin efficiency goes up
Heat capacity relates to mass (Btu/lb/Fdeg). Conductivity relates to heat transfer rate (i.e. R-value btuh/Fdeg). Sounds like you are using them interchangeably.
They are not interchangable, but each has some effect on fin efficiency. As the fin gives up heat at some rate, depending on temperature difference and the conductivity of the floor it contacts, the effect of that loss on the temperature of the fin at that spot is less if the fin is more massive (higher heat capacity, BTU/F). The efficiency of an ultra-thin foil will be relatively low, compared to that of a thick plate, for a given material.The efficiency also goes up as the conductivity of the fin is increased relative to that of the floor to which it gives up heat. An aluminum fin trying to transfer heat through a "floor" consisting of a layer of foam insulation would be far more isothermal than if the same fin contacts a highly conductive floor material.
"this is also wrong. metal thickness determines volume of heat transfer capacity."
Well, so much for what I thought I knew. ;-) "You da man" in this area, but I gotta admit I'm a bit puzzled. With good fin to tube contact a given, metal thickness is proportional to heat transfer only to the extent you can dump the heat somewhere. Once the metal heats up to the water temp, there will be no additional heat transfer except to the extent that it's radiated away, cooling the metal. At that point, it's surface area that counts and the more surface area, the faster the radiation cooling. And I would think that the plate's ability to radiate is the limiting factor here, no?Mike HennessyPittsburgh, PAEverything fits, until you put glue on it.
I think you expressed what I'm thinking better than I thunk it!;)Must be a perfect balance there someplace.
Welcome to the Taunton University of Knowledge FHB Campus at Breaktime. where ... Excellence is its own reward!
you are assuming you can just make the whole plate a certain temp. this is no more true than it is with a piece of pipe with heated water in it.let's stick with a pipe because that's more intuitive. If I have 100 feet of pipe and I push 1 GPM of 100 degree water into it, great. but the whole pipe is not 100 degrees: as water travels through the pipe, it cools down because we are pulling heat out.If I double my flow rate, I cut the amount the temperature of the pipe drops in half from start to finish. so, say before it was going in at 100 and coming out at 80 (average 90). If I DOUBLE my flow rate, it will only drop ten degrees... and my average temperature across the loop is now 95 degrees. Note I did NOT double my ability to transfer heat: I just increased it by whatever a 5 Degree rise of average temperature does for me in the application I am using.this assumes heat loss stays the same, though in reality the rise of temperature would push a little more heat out of the pipe and so this is a simplified description.same thing with a plate. it doesn't stay one temp from middle to edge when it's working under a load. it cools down as it leaves the source of heat. thicker plates cool down less because they can transmit more heat energy from source to plate edge, so the average temperature is higher.. just like when I increased my flow rate in the pipe above, which drives more heat transfer. thin plates can't really transfer all that much heat. they are better than nothing, and moreso in some other applications perhaps, but they aren't high powered transfer media. If you want to bust out numbers on conductivity and such this is all calculatable, but it takes some doing.the higher your heat load, the more important this is. You're correct that the contact issue is important as well, even good lightweights (and there are a LOT of crappy ones out there) can't contact as nice and firmly as a heavy gauge plate can.-------------------------------------
-=Northeast Radiant Technology=-
Radiant Design, Consultation, Parts Supply
http://www.NRTradiant.com
Gotcha.
My motto is field experience always trumps theory -- especially when it's MY theory. LOL!Mike HennessyPittsburgh, PAEverything fits, until you put glue on it.