GBA Logo horizontal Facebook LinkedIn Email Pinterest Twitter X 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

Hydronic Panel Rads

WPmichael | Posted in Mechanicals on

I have a steam heating system and the boiler is old and rather than replacing the boiler. I am considering a hydronic system using panel radiator instead of baseboard units.

Does anyone have them installed in there home and what is your opinion of the panel radiators? What has been your experience ( good or bad) with these radiators. It seems like the big two brands out there are Buderus and Biasi, are there any other brands I should consider? Any comments on the quality, efficiency etc of panel radiators are welcome.

Thanks in Advance
Mike

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

    Myson has decent quality value priced panel rads. There are several others at different price points & appearances (Runtal, Pensotti, just to name a couple).

    QHT has dropped the Biasi brand name for their rads and are now calling them EcoStyle brand, but they've kept the Biasi brand name for their boilers.

    Low mas bi-metal or aluminum convecting radiators (popular in Asia & eastern Europe) can also be a good value, provided the thermal mass isn't needed to suppress short-cycling of a fossil burning boiler or hydronic heat pump.

    All radiation solutions start with a room by room heat load calculation, then comes the decisions around water temperature (lower= higher efficiency , but more radiator) , microzoning, heat source, and it all has to work together.

    Panel rads and bimetal rads have fairly linear and predicable response even down to 90F-100F water temperatures, which what it takes to get good efficiency (or even sufficient capacity) out of air source hydronic heat pumps and reversible chillers in cold to very cold climates. With fossil burner boilers it doesn't have to go that low- anything below 110F yields only very marginal gains, despite a huge improvement in combustion efficiency dropping from 140F to 120F, and further incremental gains dropping from 120F to 110F.

    I don't have panel rads in my current home, but when I lived in NL I had an apartment heated with ~120F water in large panel radiators below a large viewing window- it was nice! My house has been set up for heating 125F water with a diverse set of heat emitters in different zones, including radiant floor, hydro-air, and (properly sized for temperature) cast iron radiators & cast iron baseboard more architecturally appropriate for a 1920s bungalow.

  2. WPmichael | | #2

    Thanks for the info Dana, all food for thought. I noticed the difference in BTU output vs water temp. Going from 180 F to 140 F nearly halves the BTU output ( EcoStyle Rads)

  3. Expert Member
    Dana Dorsett | | #3

    The halving of output when dropping from 180F to 140F is generically true across a large range of heat emitter types, including fin-tube baseboard and hydro-air coils, eg:

    https://www.slantfin.com/images/stories/Technical-Literature/ratings_fineline30_r.pdf

    ...as well as cast iron rads (see the nomograph on p.2):

    http://www.columbiaheatingsupply.com/page_images/Sizing%20Cast%20Iron%20Radiator%20Heating%20Capacity%20Guide.pdf

    You'll note that another rough halving occurs when dropping from 140F to 120F. So it takes ~4x as much radiator to deliver the heat with 120F water as it does with 180F water.

    The cost of radiation why most old-school hydronic systems were designed (if designed, rather than hacked) to be able to deliver the heat with 180F water, but also with a 1.5-2x oversize factor. But between the original oversize factor and updated air sealing, windows and air sealing most of those legacy radiation systems can heat the house comfortably with <140f water.

    Since you're replacing the system and have a heating history on the place, you are in a position to run a fuel-use based heat load calculation for ballparking both the total radiation requirements, and the boiler size, using the steam boiler as the measuring instrument. The distribution & standby losses of steam systems are large, especially if some of the plumbing is located in exterior walls. When running fuel use against heat load on steam systems using the boiler as the measuring instrument, discount the final load number by 25-30% to reflect what it needs to be with a low-temp system. See:

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

    While you're at it, estimate the square feet of equivalent direct radiation (EDR) of each of your radiators, and the total. (The radiator document link, above.) You may already have enough radiator (which would need to be re-plumbed for hot water) to heat the place at the 99% outdoor temperature with 140F water. If that turns out to be the case, cleaning up and re-plumbing the steam boilers might be the greenest thing to do. If they have radiator covers and/or are painted a silvery or bronze color they need to be uncovered and re-painted (any non-metallic color) to work as a low-temp radiator.

    A bit of background on the silver-painted & covered-over steam radiators: After the flu pandemic of 1918 the thinking among public health officials was that massive ventilation could prevent large numbers of influenza transmission, and heating systems were designed to be able to heat the place in the dead of winter with the windows open (!). With the windows closed they were 3x, sometimes 4x oversized for the load, even in uninsulated buildings with single pane windows.
    This of course led to massive overheating discomfort problems. Painting steam radiators a metallic color cuts the heat output by about 15-20%, which was enough for managing the hot-flash discomfort issue in some cases, but in many cases it wasn't enough, spawning a radiator-cover retrofit industry to reduce the convective heat transfer off the radiators too. This trend lasted throughout the 1920s, but they started backing off in the 1930s & later, though 2x-3x oversizing of steam system was still more common than right-sized steam, especially after insulation became common.

    Many of these intentionally-oversized steam systems for the current design load when emitting 240BTU/hr per square foot EDR, which means when converted to lo-temp hydronic they can often cover the peak space heating load with 125F - 130F water, which is still within the condensing zone for modulating condensing boilers, and can run water temps WELL within the condensing range during the average winter load.

  4. WPmichael | | #4

    Great info and an interesting piece of history.

    It seems condensing hot water boiler are very popular because of the +90% efficiency. Looking at the specs it seems you lose those efficiency once you have water temps over 130 F. Is it pointless to use these types of boiler if I use 180 F water for the rads?

  5. Expert Member
    Dana Dorsett | | #5

    If you actually needed 180F radiators when it's -10F outdoors, you'd still only need ~140F water and starting to hit 90% efficiniency when it's 35-40F outdoors. Most modulating condensing boilers have a programmable "outdoor reset curve" so that the boiler output temp to be adjusted to the minimum necessary to keep up.

    It's the entering water temperature at the boiler that is most critical for combustion efficiency. If the system isn't over-pumped you'd be getting 125-130F water returning from the rads when the boiler output setpoint is 140F.

    Run the napkin math in the links on both the heat load and the radiator EDR, then report back. Almost NO heating system (even retrofitted steam radiators) really NEEDS 180F water even at design conditions.

    With high volume (=high mass) radiators there is often sufficient thermal mass to be able to micro-zone without much risk of short cycling the boiler into low efficiency or an early grave. More on those issues can be found here:

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

  6. WPmichael | | #6

    Ok, will do Dana. According to this chart the outdoor design temp would be 12 F for White Plains, NY is this correct?

    https://www.captiveaire.com/CATALOGCONTENT/FANS/SUP_MPU/doc/Winter_Summer_Design_Temps_US.pdf

  7. WPmichael | | #7

    I used Slant Fin Ipad app to calculate the heat load for each room. And calculated the BTU for the existing steam rads. See attached spreadsheet.

  8. Expert Member
    Dana Dorsett | | #8

    The 99% outside design temp for White Plains is estimated at +12F in this document too:

    https://articles.extension.org/sites/default/files/7.%20Outdoor_Design_Conditions_508.pdf

    The SlantFin tool tends to run higher than reality by 15-25%, which is perhaps too conservative, but it'll give you the right ballpark.

    The spreadsheet doesn't contain the formula from which "BTU - Radiator" was calclulated. What do those numbers even mean? Is that just EDR' x 240 BTU/hr for steam? (looks like it, but it's not entered in the spreadsheet.)

    It looks like you have 134' EDR and a (padded) load of 27,575 BTU/hr which is a ratio of 27,575/134 = 206 BTU/hr per foot EDR, which would need ~195F average water temp (AWT) at +12F. If we assume the SlantFin tool overestimates by 20% (probably does) that would be more like 165 BTU/hr per foot EDR, which can be achieved at 170-175F AWT, and the actual load is more like 22,000 BTU/hr

    Assuming the indoor design temp used was 68F and the outdoor design temp of 12F, you have 68F -12F = 56F heating-degrees. A linear approximation of load with temperature difference then has a slope of 22,000/56F= 393 BTU/hr per degree below 68F. At the average wintertime temp of ~35F (see: https://weatherspark.com/m/24859/1/Average-Weather-in-January-in-White-Plains-New-York-United-States#Sections-Temperature ), you would then be at 68F-35F = 33F heating degrees, for an average mid-winter load of 393 x 33F = 12,969 BTU/hr.

    With 134' EDR that's a ratio of 12,969/134= 97 BTU/hr per EDR', which can be delivered at a AWT of 145F, which is pretty close to the condensing zone. When it's in the 40s outside you'll definitely be in the condensing zone, and from a seasonal average performance you'd probably hit the low-90s with that radiation, given a reasonably dialed-in reset curve.

    Run a fuel-use load calc to sanity check the SlantFin tool. You might be pleasantly surprised. Whatever the load is, it can usually be cost effectively reduced by quite a bit if the house has yet to undergo a round of blower-door & infra-red guided air sealing. If the foundation has no insulation, bringing the basement wall-R up to current code min (R10 continuous insulation, for Westchester County, US zone 4A) will typically knock 10-25% off the whole house heating load, which would allow you to reduce water temperatures even further. Before dropping $12-20K on a hydronic condensing heating system, pluck as much of the low-hanging fruit on the building envelope end first.

    A fuel use load calc on a steam system will typically be higher than reality for a number of reasons. The standby and distribution losses are high, and the high temperature of radiators located near windows raises the actual load, due to the higher air temperature at the windows, and the parasitic 24/7 infiltration driven by the atmospheric drafted boiler flue is worth a few percent too. If the fuel use calc aligns perfectly with the SlantFin tool's numbers, it's really 25% or more higher than it would be with a lower temp system. But run the numbers anyway to put a firm stake in the ground defining the upper bound.

    [edited to add]

    If the radiators are painted silvery, the 240 BTU/hr per foot EDR drops to about 190 BTU/hr per foot EDR, which means the radiators can only deliver ~25,728 BTU/hr, which is LESS than the load calculated by the SlantFin tool. Are they painted silver? If yes, do they keep the place warm when it's 12F (or cooler) outside?

  9. WPmichael | | #9

    Thanks Dana, my current system is steam, so I think the output for the CI rads are correct. Yes the house stays warm when it is 12 F but when it gets down to the single digits I can barely keep the home at 62 F. The boiler will shut off for 5 minutes at most and then startup again. The last 4 winters we have had temps down to zero, just a rash of cold weather lately.

    You have been generous with your time and knowledge. One last question, if I go to hot water system with panel radiant heat how it is generally plumbed in practice. For instance I would have 6 rads for the 1st floor.

    This site outline different plumber options, what is used most often in the field, series, mono-flow etc. I seen some formulas for instance 1/2" pex with 2 GPM and temp delta of 20 degree can delivery 20000 BTU enough for roughly 2 - 3 rad based on the heat load of each room. If you could shed some light on piping design I would be grateful.

    https://highperformancehvac.com/boiler-water-loops/

  10. Expert Member
    Dana Dorsett | | #10

    A flow rate of 2gpm is very close to 1000lbs per hour. A BTU is the amount of heat it takes to raise a pound of water by one degree Fahrenheit, so yes, 2 gpm x 20F = 20,000 BTU/hr. A 2 gpm flow or even 4gpm in half-inch PEX or copper, and it doesn't take a lot of pump, and the radiators present less pumping head than the plumbing, or the heat exchanger in a typicl condensing boiler. Low mass boilers have minimum flow requirements that might not be met with 2gpm, but there are near-boiler plumbing fixes when the boiler and radiation have dramatically different flow requirements.

    A key thing to get right is the proportional size of the radiation to the heat load of the rooms being heated, to achieve reasonable room to room temperature balance, and zone to zone water temperature requirements. Looking at your spreadsheet it looks like your existing load/EDR ratios range from a bit less than 100 BTU/EDR in the Living Room to a bit more than 200 BTU/EDR in the Bedrooms. If replacing them with panel rads, try to do better than that.

    Many panel radiators have limitations on how many can be plumbed in series, to 3 or so, but it's fine to run parallel loops and and hand-tweaking the flow balance with ball valves or globe valves. Many panel rads have adjustable internal bypass plumbing for tweaking the room-to-room temperature balance too. Note that if designing for a 20F delta-T, the first rads on the loop are going to have warmer water than the last rads on the loop, so reducing the load/radiator ratio a bit along the path can make the balance pretty good even without tweakable radiators.

    Most panel radiator specifications will list the output at two or more temperatures, typically 180F and 140F. If you design the radiation to deliver the design day heat at 140F or a bit lower you'll be able to operate in the condensing zone nearly all the time, with mid-90s type efficiency. The larger panels would also deliver higher comfort, since it raises the MRT (mean radiant temperature) of the surfaces in the room more than a smaller, higher-temp radiator. See the output vs. temperature charts for radiators in these short-form specs:

    http://www.mysoncomfort.com/static_files/int/media/downloads/myson_radiator_brochure_2015-06-05.pdf

    http://www.qhtinc.com/wp/wp-content/uploads/2017/03/ECO-and-Compact-Lit.pdf

    If designing for a hydronic heat pump it usually takes 110F or cooler water at design condition to hit the seasonal average efficiency of air to air heat pumps, which ends up being a LOT of radiator doing it with panel rads at heat loads your size. Radiant floors & ceilings are often employed Io help get the water temp requirements down.

  11. WPmichael | | #11

    You have given me a wealth of info and food for thought. Thanks for taking the time to answer my post.

  12. Expert Member
    Dana Dorsett | | #12

    Another thing to bear in mind with condensing boilers, particularly in multi-zoned systems it to have sufficient radiation on each zone to be able to emit the minimum-fire BTU/hr output of the boiler. Even better is to select a boiler that has a minimum firing rate well under half the design condition load and well under the average wintertime load, so that it can modulate running long, nearly continuous burns rather than cycling on/off. With a well tuned outdoor reset curve it's possible to get it down to fewer than 10 burns per day. This mode of operation simultaneously yields the most stable room temperatures, and the lowest possible fuel use, taking every marginal bit of condensing efficiency that it can get, and with fewer ignition cycles the maintenance issues are lower.

    More on sizing modulating condensing boilers here:

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

    With a design load in the 25,000 BTU/hr range ideally you'll want the minimum fire output to less than 25,000/4 = 6250 BTU/hr, if you can. Most boilers have a much higher minimum firing rates than 6250 BTU/hr out, but there are several with min-fire output less than 8000 BTY/hr to choose from. With a min-fire output ~1/4 your design load it can still have long nearly continuous burns without being forced into cycling whenever it's under 55F or so outside. A boiler with a min-fire output of 20,000 BTU/hr would have to cycle whenever it's above 25F or so (which is even cooler than your mean outdoor temp in January). Oversizing a modulating boiler to the point that it rarely or never modulates sort of defeats the purpose, robbing you of both efficiency and comfort.

  13. SwitchgrassFarmer | | #13

    I have two panel radiators on the second floor of my house. They are fed hot water from my ground source heat pump (geothermal).

    Hottest water I have made so far, when it was -4 outside, was around 133 degrees. That takes ... awhile.

    I have that second floor area set for priority on the pump controller, it will get hot water first. Fortunately it's a guest bedroom and rarely used, so normally I have its thermostat dialed back. (We run outdoor reset using Taco mix valves on each of the radiant slab pumps for the rest of the home.)

    The picture I attached is of the second panel radiator on that floor. It's on the same manifold as the one in the room with the thermostat, but we use this thermostatic operator to fine tune its output.

    A high volume low pressure sprayer was used to paint these radiators to match the wall and trim colors.

    John Siegenthaler's original design for my house had panel radiators on our first floor too. I thought that was great from an engineering perspective; engineering lost out to aesthetics with another household member.

    All in all I think they work well.

  14. WPmichael | | #14

    Dana that is another parameter for the equation. I am Mech, Enginner by education but in IT now. So these topics are not completely foreign but it was decades ago and no practical experience. I have gotten around to do a full heat load calculate on the entire house.

    1st Floor 29700
    2nd Floor 18500
    Finish attic 9100

    Total 57200

    As a sanity check does this sound like a heat load for a 1920's house with 2x4 construction with recent blow in cellulose insulation. The window are large 5.5' x 2.5' double glazed glass with aluminum frame. The metal conducts heat more than plastic types, 9 windows per floor. The house footprint is 45' X 20'. Is nearly 60K BTU seem reasonable?

    About the minimum fire output, so with 57200/4 = 14300. I should look for a boiler that can obtain a minimum fire rate in this range?

    Also I noticed many mod-con boiler tend to be 80K or 105K BTU units. The 80K units have a Net (I=B=R) capacity of 62K BTU. Which is slightly about the house BTU, would this be an appropriate boiler?

    PS Thanks for the comment and photos Andrew.

  15. Expert Member
    Dana Dorsett | | #15

    A load of 57-60K @ +12F for a an insulated 2x4 framed house might be about right if the size of the house was north of 3000' of fully conditioned space.

    A more likely load for an 1800' house with an UN-insulated 900' basement and 450' of insulated finished attic would be under 40,000BTUh/r @ +12F, probably somewhere in the low to mid 30s (which would be consistent with your radiation EDR estimates),. With air sealing and basement insulation it could be in the mid 20s. The SlantFin tool came up with 27,575 BTU/hr, literally half your calculated number, and that tool consistently overestimates reality, NEVER under-shooting by 50%!

    There's something seriously wrong with either your load calculatons or your EDR estimates. 57,200/134' EDR= 427 BTU/hr per square foot of existing radiator, whereas the rads can only deliver 240 BTU/hr per 'EDR. That means with the existing radiators the house can only be heated to 70F at temps of about 32F and higher.

    A fuel use based load calculation would put a firm upper bound on it, so run those numbers!

    https://www.greenbuildingadvisor.com/

    Using the IBR output capacity for sizing a boiler is only valid when the boiler and a significant fraction of the distribution plumbing is on the other side of an insualted wall from the conditioned space. If it's in a basement without insulation in the basement ceiling/first floor joists, use the DOE output. Air sealing and insulating the basement walls still makes sense, but in a 2.5 story house + full basement the reduction the total heat load would be smaller as a percentage than it would be on a 2000' rancher with a 2000' basement, with 1.5-2' of above grade exposure on the foundation.

  16. WPmichael | | #16

    You have a good eye, the Slant fin app only had the first room at the outside design temp of 12 F.. The remainder of the rooms defaulted to zero and I did not catch that. I used an indoor design temp of 70. I did not use the slant fin app for the attic, the wall are slanted etc, the radiators calculated to 9100, I think a better estimate would be 5000. The numbers now are:

    1st floor 24700
    2nd floor 12898
    attic 5000

    total 42598

    I think the large windows (5.5' X 2.5' ) increased the load beyond the mid 30's you anticipated.

  17. user-7047834 | | #17

    Dana, you are a god! Thank you very much

Log in or create an account to post an answer.

Community

Recent Questions and Replies

  • |
  • |
  • |
  • |