Welcome! Here are the website rules, as well as some tips for using this forum.
Need to contact us? Visit https://heatinghelp.com/contact-us/.
Click here to Find a Contractor in your area.

Snow Melt

hot_rod Member Posts: 22,121
what about the weight loading on "The Barrier" do they publish a psi load number?

I see they now offer a 3/4" thickness. I may reconsider that for some large shops I have in the works.

Sure is a lot easier to install, and stays together better than sheet foam on irregular subgrades.

hot rod
Bob "hot rod" Rohr
trainer for Caleffi NA
Living the hydronic dream


  • MikeJ
    MikeJ Member Posts: 103
    Snow Melt

    Looking for ideas for a good snow melt system in NE. If you could point me to on line information, or email me with articles or suggestions. It’s going to me a 2300 sq. Ft. Drive way, with 4” concrete. I do plan on insulating under it, is 2” of insulation over kill or is 1” enough. Is it better to go with smaller pex and more of it or larger pex for more length? I have done some in floor, Do you need to cut down on tube length compare to in floor, Thanks Mike J


  • Hi Mike,

    for a driveway that large, you are going to be using 3/4" pipe. Flow rates are huge in snowmelts... 200 feet is long for a 5/8" loop, for example. So see the distance of one down and back on the driveway and you'll see why you will probably need more diameter. 9" o.c. is probably the max you want to use.. 6" if they need fast melting.

    2" of insulation is a good idea, especially if the snowmelt will "Idle". If not, then the thermal break of 1" may be adequate, but I'd do 2" anyway.. clients have a way of changing their mind about how they want the snowmelt to work later on.

    Uponor/Wirsbo has a Snow and Ice Melting Design manual that is very helpful.
  • Glenn Sossin_2
    Glenn Sossin_2 Member Posts: 592
    Snow Melt

    We have had very good success with an insulating product called the barrier. Extremely easy to use - rolls out like carpet padding. It does not have a high R-value like 2" of foam, but for the few hours of the typical snow melt cycle, will perform almost the same. Most of the systems we design are not idling, and are used 2-4 times a month for a few hours at a time. See attachments. I agree with Rob if the potential for idling exists, you may want to consider more insulation.

    Depending on the location, type of use, and physical construction aspects of the driveway, we figure a load between 125 -135 btu/sqft and use 3/4" pex @ 12" centers with a dT of 25F. We will decrease tube spacing to 10" at the foot of the road where the snow plows roam, or on an inclined portion of the driveway.

    We typically design with 3/4" pex loops - either 250ft or 300ft. In a situation like you describe, we may have a 1" or 11/4" trunk of K copper run down the side of the driveway and install a remote manifold installed in a pit to reach the far end of the drive.

    We also use Quip clips - great product. Makes fastening the pex down really fast & easy as long as you run your pipes along the wire mesh grid. I've been after them to come out with an alternate version that would allow you to slide the clamp left to right to get centers tighter than 12" which is ctrs of the roadway wire that we've seen.

    Hope this helped you. Good luck on your project.
  • Glenn Sossin_2
    Glenn Sossin_2 Member Posts: 592

    Don't recall seeing anything on it's crushing strength, - its nothing like the blue board which I believe has designations like T-40,T-60 etc reflecting psi strength values. Typically, we will use it on top of a few inches of RCA mix or tamped earth. We will have a minimum of 2" of concrete(w/pex), some sand and pavers, or approx 4" of concrete(w/pex). The weight of the concrete is uniform so I would think it would compress the insulation evenly across the entire surface area.

    I am aware of the new 3/4" product. I haven't brought it in yet though. I'm not sure it would have a significant effect in a typical installation unless we had an idling situation such as the driveway and steps of a fire house. They don't care about fuel becuase the municipality pays from our taxes.

    If you look at the Ohio power show statistics attachment, you will see no change after 6.5 hours when compared with 1" of structural foam. Most snow melt systems here on LI don't run much longer than that so why pay for 2" of foam. This stuff is a dream to install. Unrolls like carpet padding and has an integral sealing tape on the side. Just roll up the side for your edge insulation, or cut a strip with a utility knife.

    There's no argument that 2" of foam has a surperior R-value. The bigger picture I see is cost justification. The system runs for 5-8 hrs at a time or less. If there is no significant difference in downward heat energy, why be concerned. After all, we're heating thousands of pounds of mass and its less than 30F on the other side of it for a few hours.

    I may bring it in the next time I order it and use it where there is a commercial installation and potential idling. For any given snow melt cycle - 5-8 hrs, for a typical driveway, I think the difference in downward energy loss between 1" of foam, 2" of foam, and the barrier insulation would be of little consequence. I am drawing this conclusion based on their published report.

    The typical customer who puts one of these systems in, isn't going to sweat about running the system an extra hour after they just spent $25K -$60K to put the driveway/snowmelt system in.
  • Mad Dog!!!!!!!!
    Mad Dog!!!!!!!! Member Posts: 157
    Mark Eatherton has designed all of our snowmelts

    We put down 5/8" and keeo the loops down to 250 max. On Long Island, we can melt snow at 90-100 btus p.s.f. You would be doing yourself and your client a great service getting a REALLY good design to work off of. Mark is always willing to work with good folks. Mad Dog
  • Mad Dog!!!!!!!!
    Mad Dog!!!!!!!! Member Posts: 157
    2\" Dow Poly styrene is the best way to go

    60n psi rating. Ive used both...for snowmelt, I would NOT skimp on insulation...fuel costs TOO much. Much
  • make sure u have

    As others said, insulate all around! The edges,etc and make sure you have run off area for drainage! Melting snow will needs a place to go... Checked out one job that had tubings installed and no insulation... I declined installing the heating unit in this attorny laden town due no snow melting runoff to drain or drainage. Sure enough, alleyway all pooled and frozen solid with ice...
  • Glenn Sossin_2
    Glenn Sossin_2 Member Posts: 592
    Point of diminishing returns -

    Is the finacial benefit of thicker insulation worth the additional labor and material costs. Take a location with an ASHRAE design temp of 15F. Is the benefit of R40 instead of R30 in the ceiling worth the additional costs? What about R50? We balance the cost benefit ratio in our design decisions.

    Following that logic, lets look at the snowmelt system. Lets say it runs typically 25 hours a month. How do you think the run time of the boiler, and the speed of melting snow would be affected if we only had 1" of foam insulation versus 2"?

    When looking at the barrier product, given the typical snowmelt useage / cycle times, is there a real benefit to using 1" or 2" of foam. Does the ease of installation and material cost savings outweigh any addtional fuel costs? How much might those fuel costs savings be? I concluded, right or wrong, that using this product would be a better choice than foam in most residential applications. So far we have been very happy with the results.

    With respect to design temps - I was called out to give a 2nd opinion on a job, designed locally by a major manufacturer. Without being told anything, based on the data, I estimated a heat load of approx 850,000 - 900,000 btu's. I found out later, there was just over 600,000 installed. The system didn't work right, only parts of the driveway melted. Pumps were plenty big. Of course you would conclude this was a balancing issue. The plumber spent a whole day balancing the system out within 1-2 degrees on the returns. The mfg rep was out there too. It still didn't work completely. It was desgined for 88 btu/sqft, - way to small in my book. To the best of my knowledge, there is legal action now.

    According to the Viega Snow Melt manual, the 1999 ASHRAE Handbook,table 3.2A, Level 1 for Albany NY is 149 sqft. Being here on Long Island feel very comfortable about the 125 -135 temps I design to. If there is a reasonably steep incline on a driveway, I will design that portion to 160 btu/sqft.

    Most important is the customer. They have an expecation of performance. I don't believe they are concerned whether they burn $20, or $23 worth of fuel for a snow melt cycle. After the major investment they made in the driveway and the system, they want to see the snow melt. While fuel costs are high, we have to keep in mind these systems are not running 24/7, infact out here, in many cases, their lucky to see 24hrs period for a given month.

    I would love to know if there are any control studies that demostrate the effect of different insulation values under identical slabs and identical load conditions.
  • MikeJ
    MikeJ Member Posts: 103

    Thanks to all that post, you all have given me a place to start and lots of ideas for
    installing a good snow melt system.
    It was great to see those pictures Glenn, very nice looking jobs, a
    picture is worth a thousand words.
    Never even consider edge insulation, until now.

    Has any one ever used track railway to fasten down the pex?

    Thanks again
    Mike J

  • ScottMP
    ScottMP Member Posts: 5,884

    is the "Barrier" product along the same lines as Insultarp ? If so I can't imagine its less expensive than 2" foam board. Even if it is, would'nt the better performance and fuel savings in reduced downward loss make it worth the extra money ?

    In your opinion, how much more has the 2" board been on an average job ?


    To Learn More About This Professional, Click Here to Visit Their Ad in "Find A Professional"
  • Glenn Sossin_2
    Glenn Sossin_2 Member Posts: 592
    Blueboard installation costs

    In the comments I made, I am considering not only the cost of the product itself, but the labor associated with it from excavation to installation. The barrier product is approximately 1/2" thick. When compared to 2" of board, we have an additional 11/2" of depth to excavate. Unless you have a very level grade, it cracks when you walk on it. Generally, the foam seems harder to work with.

    While we are a wholesaler, it is not my intent to look like a salesman promoting this product. I think insultarp also is easy to install,- we sell that also. The barrier insulation is apprx 25-30% less than insultarp. We sell a 240sqt roll of barrier for around $150. I don't actually have any current pricing for the foam board product. If it's used, someone else is providing it - we sell the plumbing/pex/boiler side of the installation.

    With regard to the downward loss, I would agree the foam board is clearly superior and would save money on fuel. However, keep in mind, we are looking at a very small window of operation for the typical system here on LI, - it turns on when it's snowing, typcially 10- 20hrs a month for the season. So again, the question is, how much energy is saved during the coure of a snow melt cycle?

    Consider a typcial 20 x 40ft driveway -say the systems runs for 6 hrs using 2" of foam board designed at 125 btu sqft. Thats a load of 100,000 btu's. We can buy this amount of energy for around $2.25 -2.50 depending on the fuel. Even if the system ran for 2 more hours using barrier and the downward loss was 50% higher , how much are we talking about in dollars? $3 - $5 ???

    I am making an assumption that the test data they published is true. The logic/construction of their test makes sense to me. A block of 5" concrete heated to 80F, then the insulation, then 30deg frozen dirt. They montior the temp of each. After 6.5 hours no performance difference between their product and 1" of foam.

    Following this logic, you have to wonder why they came out with a 3/4" product. I will contact them later today and ask for some additional information.
  • ALH_4
    ALH_4 Member Posts: 1,790
    Testing Methods

    I am not sure how this simulation was performed. They heated a 5"x7"x11" block of soil beneath a 5"x7"x11" block of concrete while in a heated room for 5-1/2 hours and it did not warm up? Usually dimensions are LxWxH, so was this an 11" thick chunk of concrete? How do you simulate snowmelt conditions in a 70°F room?

    Glenn, do you have any details on how the testing was performed?

  • I suggest you reconsider your assumption that the test data is true. stop for one second and think. The Barrier is 1/2" of foam. On what planet could 1/2" of foam be "no different" than 1" of foam? As you note, thickness matters, or they would not have come out with a 3/4" product.

    diminishing return, maybe, but that statement is quite obviously false. Also note that the Barrier's manufacturer has been sued for making misleading statements about insulating values.

    That said, drop Insultarp. If the barrier is cheaper, it's at least a 2.5 R value or so instead of the 1.65 you get with the insultarp. no reason to ever use the insultarp in that case.

    But neither is ever equivalent to 1" or 2" of rigid foam, and the cracking of rigid is of no concern. You may not always need the equivalent of 1" or 2" of foam... but never think they are equivalent.
  • Glenn Sossin_2
    Glenn Sossin_2 Member Posts: 592

    I was skeptical at first also. I have learned that testing, depending on how its done, can definetely skew results - particularly when hot box tests are done with reflective insulation products.

    I first saw this product at an RPA convention a couple of years ago in Denver. At first, I was going to keep walking past the booth. There was an older man there. I could see how thick it was yet they were claiming it performed like 1" of blue board. My limited exposure to physics said this was impossible. I would have expected the R-value of this product to be around 2.

    I went up to him, prefaced our converstation by saying I was challenging his claims based on the R-Value of the foam, but I was interested in hearing the how & why he thought this worked. He came out and said he didn't understand it fully, but drew the analogy of a thin foam coffee cup. You have 120F liquid inside, yet through you barely feel anything when holding the cup. Quite different with the paper cups from Starbucks or 7-11.

    Then he showed me the Ohio Power table. Based on this, and pondering the construction of different coffee cups, I decided to give it try. I've been very happy with the results and the contractors that we've sold it to love how easy it is to use.

    I sent a copy of this thread to the rep and to NE sales manager for Northwest Insulation Products, the manufacture of the barrier product. I will find out more information and post the results
  • No Need to bring him here

    Glenn, don't let the sales guys confuse you.

    This product is XPS foam.

    Rigid foam is XPS foam.

    There is only one thing that differentiates them from an insulating standpoint: thickness. Period. End of story. Again, this company has been sued for making VERY misleading statements about K-value, trying to dodge the R-value question.

    The barrier may be easier to install and may include a vapor barrier.. that's great. But from an insulating standpoint, it is only thickness that matters here.
  • Glenn Sossin_2
    Glenn Sossin_2 Member Posts: 592
    Understand and Agree

    Thats why I challenged their salemsan at the show right up front.

    I have a copy somewhere of a DOE report from 1985 showing that a sheet of foil performed like 6" of batt insulation and published the data to support it. We all know this isn't right base on how its applied to the bubble foil stuff. It only works if there is sufficient air infront of it to relfect the radiant component of heat energy.

    These tests, how they're performed and interpreted, can lead to all sorts of claims. I wish it were as simple as IBR ratings. A standard we can all accept and work with.

    As I posted earlier, I will contact them and find out more.
  • Brent_2
    Brent_2 Member Posts: 81

    I would not trust a manufacturer that had the FTC issue it a $104k fine for advertising false R values.
  • Glenn Sossin_2
    Glenn Sossin_2 Member Posts: 592
    Very Intersting

    I wonder how they will try to explain this??

    How about the foil guys similar fines ??
  • Glenn Sossin_2
    Glenn Sossin_2 Member Posts: 592
    Time is the more important variable

    Clearly the R-value is not the same. The question arises, what is the temp difference in real world applications after 5-7 hours of run time. If we could let the system run under controlled conditions, eventually, it would reach an equilibrium in heat flows up and downward - steady state efficiency. How long does this actually take? Then the R-values of the various product could be compared legitimately. The question would be, how much downward heat energy is lost over a 6hr period. Could their chart be correct?

    Over an extended period, I fully agree, this product is what it is - 1/2" of foam. Regardless of the type, I'm sure the R-value is still right around 2.

    I'd like to see a test showing the actual performance of this product compared to structual foam over 6hr,& 10hrs. This would tell the real story. For a fixed set of conditions, how long does it to see a change in the underside temperature with their product versus 1" of foam. This is why we put the product in - to restrict downward energy flow.

    If it takes 8 hrs to see a 2F difference in the underside of the insulation, wouldn't it be worth using? We have to remember, the system is typically on for only a few hours at a time.

    I called and spoke to Wally, owner of Northwestern Ohio Foam. He said he will respond on this thread. I asked for supporting data explaining how the tests were done. Should be very interesting.

  • you're making serious assumptions here Glenn.

    Have you studied the soil quality on the specific site you are talking about?
    Are you sure this particular system is not going to be used in an idle mode? Even if it's not, it's going to have to be on for several hours preceding a storm, and for several hours during a storm.
    Do you know that the increased initial load isn't going to make the system "fall behind" a heavy snowfall?

    Your assumptions are correct in some cases, and not in others. Generally speaking, the cost of insulation is not what makes or breaks a snowmelt project of any size. Why increase risks? They are paying large dollars to make sure the snow goes away.
  • Wally Radjenovic
    Wally Radjenovic Member Posts: 6
    Barrier and BarrierXT

    I wanted to clarify some items here that I've seen posted.

    1. If the Barrier were 1" thick it would have an R value within 10% of extruded foam boards of 5. So 4.5 vs. 5.

    2. The "Powershow" testing was an in house test done as follows and not a steady state heat transfer test: We used frozen dirt that we brought in from outside, put insulation material that we were testing, used concrete bricks that we made that we put in heating cartidges with a temp controller to keep the temp at 80F and put a thermometer between the insulation and the dirt to measure the temp.

    This test showed that there was no temp change in the Barrier or for 1" of extruded foam, it certainly was not scientific, but all the 4 samples were subjected to the same variables so for comparisons sake, it is interesting.

    We had the Barrier and BarrierXT tested as part of a concrete assembly of 3.625" concrete, 3/8" Barrier, 1" of sand and 3" of #57 gravel. The R value of the system was 4 and if you add another 1.7 for BarrierXT (for the additional 3/8") the overall R value goes to 5.7.

    I'm having the test lab that we had our testing done at provide me with the underslab temp versus time so we can see in a non-steady state system like a snow melt how the system performs. I hope they kept the data and I will post it if I'm able to obtain it.

    3. BarrierXT is 3/4" thick and still in roll form 4'x60' rolls

  • Glenn Sossin_2
    Glenn Sossin_2 Member Posts: 592
    No idle mode

    I have never gone to the extent of studying the soil - have never run across someone who has done that for a typical snowmelt. Sometimes we have to take the water table depth into account, this is always a consideration. As far as idling, I typically would only consider this on commercial jobs like the steps/driveway of a firehouse or a service ramp at a car dealership. Idling can be really big money. If someone express an interest in this option, I would not use the barrier product and would spec 2" of foam.

    I'm not clear about the storm comment. I typically use Tekmar 090/091 sensor. Very minimal false triggers. If someone does turn it on in advance of a storm manually, then of course they would burn more fuel - it's their decision.

    For LI, I think we need to take our normal heatloss thinking and set it aside when doing snowmelts. We are not operating the system 24/7 - it's intermittent. The typical snow melt system will never reach a steady state condition.

    I think your making a big assumption that people with these systems are sitting around waiting to turn the system on in advance. As I said earlier, and agree with you, people who pay for these systems expect and want them to melt snow - thats what I try to design them to do.

    Some other people have posted they use 5/8" pex. Unless its a small system, I use 3/4". The cost difference is about 18% but the head loss using 5/8" pex can be 30 -50% or more depending on the design factors. With 3/4", I can use wider centers, less pex, longer loops, smaller manifolds, and potentially less costly pumps - thereby lowering my material costs. I am not trying to imply 5/8" is the wrong pex to use - each designer chooses his weapon - the majority of the time, I choose 3/4" - usually because of the size of the system.

    As far as the customer paying big dollars, I agree fully which is why I figure 125 - 135 btu loads or higher based on site conditions. The last thing I want is a customer to complain the system isn't working to their expectation. Seen that, avoid that.

  • my point glenn is that you are designing a particular type of snowmelt and there are many variables, and the variables impact insulation selection, such as the location of the system. On your long island sand, you probably have less to worry about than in New Hampshire Granite (barring water table issues). On the internet, it pays to be specific about the considerations, because you never know who is reading or where they are ;)

  • Glenn Sossin_2
    Glenn Sossin_2 Member Posts: 592
    Variables galore

    Absoultely agree, which is why I used the word typically and identified LI. Every locale / job has different characteristics.

    My purpose in the post was to give my view/choice of products and support my reasoning. If someone points out an error, or comes up with a better way - I'm far from insulted. We are not born with this knowledge. You can learn from your own mistakes, the experience of others, and the text book/class room. Use them all and this site as a way to further my skills and increase my knowledge. I was just trying to think outside the box. Some good ideas come up that way.

    I'm sure you'll agree, it would be usefull to get an idea of what actually happens under the slab, with different insulation at various time intervals for snowmelt. Then we'd all have the ability to make better recomnmendations and designs.

    Enjoyed the exhange. Look forward to meeting you one day.

  • absolutely right glenn, and I'm as guilty as anyone for knee-jerking out the 2" rigid foam selection.

    That said... I also have a strong resistance to supporting companies that make bread and butter via misleading statements about Rvalues and heat transfer. Basically all slab tarp and roll products meet that bill. Years of BS have left me bitter, I guess.

    But functionally you are probably right about the overall impact in some systems. And, I'm sure we'll get to shake hands some day. Take care,
This discussion has been closed.