GBA Logo horizontal Facebook LinkedIn Email Pinterest Twitter Instagram YouTube Icon Navigation Search Icon Main Search Icon Video Play Icon Plus Icon Minus Icon Picture icon Hamburger Icon Close Icon Sorted

Community and Q&A

PEX heat transfer plates – extruded or sheet aluminum?

Alison N | Posted in Energy Efficiency and Durability on

We’re retrofitting an existing house with under-floor radiant heat. From my research, it seems clear that we should install aluminum heat transfer plates, but I haven’t been able to determine whether it’s worth investing in the extruded aluminum C-shaped heat transfer plates over the sheet metal (omega or U-shaped). Even when buying in large quantities, the extruded plates are about 3x more expensive than the sheet metal plates.

Healthy Heating has a helpful 2013 study that compares the temperature profile of the extruded aluminum and the U-shaped plates. My big take-aways from that article are that the extruded plates had less even heat distribution across the plate, but lower return water temperature (good for efficiency). I guess the main question is whether that efficiency gain is enough to justify the up-front additional cost.

I realize there are many factors that go into the efficiency performance in a given house, and we will be working with a local heating company to determine the settings for the system. I’m hoping someone who has experience with both plate styles can give us a sense of how much of a difference the extruded plates make compared with the sheet aluminum.

If we do go with the sheet metal, my instinct is to go with the omega shape, considering there is more contact between the metal and the tubing. However, I’ve also heard that the sheet metal plates can be noisy when the tubing expands as it warms, and I wonder whether there is more space in the U-shaped channel for the tubing to expand without affecting the metal around it. Any thoughts on that aspect?

For additional context, we are planning 1/2″ PEX, two lines between the joists (which are 16 OC), insulated from below. Hardwood (oak) floors above, little carpeting. We haven’t decided on a water heater yet. We’re located in Vermont, and in addition to the new heating system, we are also retrofitting the house with exterior foam, air sealing, and new attic insulation.

GBA Prime

Join the leading community of building science experts

Become a GBA Prime member and get instant access to the latest developments in green building, research, and reports from the field.

Replies

  1. Expert Member
    Dana Dorsett | | #1

    In most cases, as long as you go with a higher quality sheet metal version it's hard to rationalize the higher cost of extrusions.

    It can be cheaper to increase total output and to preserve low return water temps during periods of higher load by adding a second stage (or parallel flow) that cuts in some amount of panel radiator or baseboard, using floor-only to cover the average load, but not necessarily the load at the 99% outside design condition. Since the floor is the primary radiator you still have the cush-factor, and it doesn't take much panel radiator to even double the total output to cover the peak loads.

    It takes a lot of detailed information about the system and load it's hard to say whether that would be a preferable, desirable, (or even necessary) option for you. To be sure, system responsiveness goes way up when there is some panel radiator in the radiation mix. There is a real time lag when going through ~1.5-2" of wood with any under-the-subfloor approach, but if even 1/3 of the total radiation out is panel radiator the system has noticeably less lag, and tighter room temperature control, making it roughly as responsive as an above-the-subfloor radiant floor system.

    Both sheet metal and extruded plates can experience some noise during a cold-start of the zone. PEX in contact with wood will also make some noise on a cold-start.

    All heating system decisions start with aggressive room-by-room Manual-J heat load calculations, best performed by a qualified third party such as a certified professional engineer or RESNET rater, but specifically NOT an HVAC company(!). While there are HVAC companies capable of doing that competently, the track record on accuracy is abysmal, even among those who offer that service. With the load numbers run by people who are in the business of running accurate numbers rather than designing & installing heating systems, your odds of getting it right are better. Presenting those numbers to the heating contractor allows them to focus on designing systems to support those specific loads.

    Have the Manual-J calculations done on the "after" picture of your intended upgrades, using aggressive assumptions about eventual air tightness, and R-values, and only at a code-min 68F indoor design temperature, and only the 99th percentile outdoor design temperature. If higher indoor temps or lower outdoor temps are used it inflates the number, which exaggerates the numbers on both the radiation and the heat source/boiler. Since the HVAC designer needs to cover the calculated load, they tend to be conservative, putting in more radiation and more boiler than the calculated load numbers indicate rather than risk endless call-backs from irate customers. So if starting out with an exaggerated load number from a sloppy load calculation, further skewed by additional size inflation at heating system design stage it's way too easy to end up 1.5-3x oversized for the real loads.

    If you have a heating history on the place, even though the house improvements aren't already in, run a fuel-use based load whole-house load calculation from mid-winter fuel use (skip shoulder season fuel use, which suffers large errors induced by solar gain and hot water use, etc.) For an explanation of how that's done see this:

    https://www.greenbuildingadvisor.com/blogs/dept/guest-blogs/out-old-new

    If the heat source is going to be a modulating condensing boiler, run these numbers too:

    https://www.greenbuildingadvisor.com/blogs/dept/guest-blogs/sizing-modulating-condensing-boiler

  2. Alison N | | #2

    Thanks, Dana, this is helpful. It's difficult to tell which of the sheet plates are higher quality, but we're thinking it might be safest to go with a brand name (e.g. Mr. PEX) over some of the generic ones we've found.

    Unfortunately, we don't have historical heating information for the house, as we just purchased it 6 weeks ago. However, we have discovered that the walls currently have zero insulation - just air! Thanks for the suggestion about starting with the Manual-J calculations. We'll look into who in our area could do that for us.

  3. Expert Member
    Dana Dorsett | | #3

    Don't get TOO far ahead of the process by specifying the floor heating plates before you even know what they have to deliver, or what the heat source or heat load is. With a way-better than code house it doesn't take much to deliver design-day heat at condensing temperatures.

    Even specifying radiant floors may be premature, depending on just how much you are lowering the load. It's possible to end up with a load low enough to heat & cool the place decently with ductless mini-split heat pumps, at a fraction of the system cost of a full-on hydronic radiant floor solution. (And far cheaper operating costs than propane or oil boilers.)

    Are there any window updates happening with this major make-over? (If not, what's currently there?)

    How about basement wall insulation?

  4. Alison N | | #4

    Windows were single-pane from when the house was built in 1929, are being upgraded to Energy Star-rated double-pane Anderson windows. No basement wall insulation yet, but as we add the exterior foam, we are going to dig around the foundation to add foam board a few feet down.

  5. user-2890856 | | #5

    Alison ,

    The investment in installing an extruded u or c plate as opposed to a tin foil plate is well worth it . CONDUCTION is KING . The extruded plates make full contact with the subfloor whereas it is very difficult to get that type contact from a flimsy sheet metal plate , actually , it is impossible .
    Dana brought up some very good points that should be heeded . BEWARE of internet peddlers of system components even if they show good cherry picked reviews and especially if they operate out of Vermont . Competent designers are very difficult to find , stay away from folks that use the words specialist , expert and similar . I have made a very good career of following those folks up and fixing their messes and fulfilling their broken promises when possible .

    A properly designed radiant system should never be a cold start and suffer noise issues from the plates either . ODR and maintaining MRT rather than playing a catch up game keeps that all to a minimum and also allows your equipment to operate at elevated efficiencies .

  6. Jumboski | | #6

    Hi Alison,

    I know this is an old post but I'm curious as to what decision you made and how it has performed. I am in a similar position as you were a couple of years ago.

    Thanks,
    Jim

  7. Lazenby | | #7

    Me too! Any more updates on how it went?

Log in or create an account to post an answer.

Community

Recent Questions and Replies

  • |
  • |
  • |
  • |