Snow melt

gpjazz

Can someone help me get info in designing a snow melt system for concrete driveway and sidewalk? What is the best source to find some information? I've done a lot of radiant floor systems (interior), but never did a snowmelt system. I'd imagine the basic idea is the same, but I'm not sure how to size btu's, loop length and size, gpm, ect... And what about insulation under concrete and around perimeter?
This will be a hot water system in western Pennsylvania. Driveway is about 15x50, sidewalk about 3x75. My customer asked me about this and I told him I never did one, but that I'd ask some questions. I'm excited to do this project, but I need a little education first to make sure it's done properly.

Dave H_2

To melt snow, it takes a large amount of energy since you are heating the outdoors and performing a phase change of snow to water to vapor. Typically start at 100 BTU's per square foot.

The second part that is much different than radiant floor heating is that you want to use larger diameter pex. Because of the large load, the gpm flowrate is going to be much higher.

Third, because of the high gpm, loop lengths are going to be shorter resulting in bigger manifolds.

4th, plan remote manifolds near the areas, you don't want to run leader lengths back and forth to the manifolds that will not be melting snow.

And least but definitely not last is controls. This does not operate on an on/off switch. It will depend upon customers expectation levels; accumulation of snow when it starts snowing, a little bit but it is also melting, absolutely no snow on the area or a snowflake turns to steam when it hits the driveway.

Just a quick list to think about when designing and quoting a project.

Dave H.

GGross

ASHRAE says you are at about 110 BTU/h per Square foot plus your backloss which they figure around 30btu/h though this may change with insulation We usually figure between 125-150 btu/h per square foot here in northern michigan

5/8" or 3/4" pex. I prefer 3/4" but we have a whole bunch of systems installed with 5/8" tubing.

Throw out every bit of "rule of thumb" you ever heard about loop lengths, those rules are for space heating, not snow melt. When I design a system I will play around with the loop lengths a bit to keep total head down, some contractors want longer loops and bigger pumps, to each their own, I like shorter loops and smaller pumps. Rough estimate on yours had around 230' max loop length to keep the loop head below 10' Some times I go a bit shorter than that as well, all depends on the other factors impact on the total head loss

Mind your manifold pressure drop, I use different manifolds for snowmelt systems on occasion, the Uponor brass manifolds seem to have better flow characteristics than their stainless manifolds for example. Locate the manifold as close to your snowmelt zone as physically possible, and account for a manifold box. We use the ranbird v-max or something similar designed for irrigation access.

Account for your supply and return tubing head. Don't just size your S/R based on the size of the manifold. If you are going a long ways from the boiler don't be afraid to upsize that tubing to keep your pump smaller. Insulate the heck out of that pipe, don't bury any connections this is a permanent fixture of the property now.

If you are tying to wire mesh consider renting a PexGun they are outstanding tools. https://pexgun.com/collections/pexgun-and-tool/products/pex-gun-complete-tool-kit

Tube spacing may be a bit controversial. I always design on 6" centers unless a contractor tells me otherwise. I have some that only install 5/8" tube on 9" center, those systems work from all reports just fine. I go tighter because we get one chance to do it right, and impact of a possible leaky loop in the future will be slightly less.

Use pex-a or something that allows you to use a heat gun to undo a kink. Tight spacing and larger tubes, it can be quite tricky. Several contractors tell me it is physically impossible to install 3/4" tubing on 6" centers, it is done all the time but some have real trouble with it.

The cost of a snowmelt system may seem really high at first, but thats not really the cost. The true cost of a snowmelt is the cost to operate. Therefore you should put extra care into lowering operating costs. Properly laid out design to keep your SWT lower, ECM pumps, Tekmar 091 090 and 654 for controls, or the real expensive control if you want wifi.

I probably forgot a lot of finer details. There are tons of folks here on the wall who can help with specific details or questions.

STEVEusaPA

https://www.uponor.com/getmedia/ad591b16-0881-4f43-9b71-30a995f0ff73/snow-and-ice-melting-design-and-installation-manual.pdf?sitename=Canada

hot_rod

First decision is how much snow you expect and how quickly you want it gone, or mostly gone
ASHRAE has different class designation. Here is a good primer.

https://www.phcppros.com/articles/8410-snowmelt-design-the-art-of-best-guess

HomerJSmith

I use a counter flow pattern of tube layout because I think it gives a more even heat energy release thru out the slab as opposed to a serpentine pattern . Using a larger tube size lessens the pumping power. There isn't much difference in terms of BTU release between the smaller size tube and a larger size tube, but with a larger tube there is less of a chance of cold spots.

I never use the reflective bubble wrap as an insulative medium. I use rigid high density foam board to prevent backloss. The tubing should be about 1-1/2" to 2" below the top of the slab. I also make my tube lengths some what shorter than the recommendation for tube size. 1/2" tube, instead of 300', I'll go at 250'.

hot_rod

Problem with short loops is you end up with a lot of runs from the manifold. The worksheet shows 200’ for 1/2” tube. That might not even make it to the end of the driveway🧐
The 40% glycol adds to the pumping requirement also.
1/2 is handy for stairways sometimes so you pack in 6” OC

HomerJSmith

Problem with short loops is you end up with a lot of runs from the manifold.

Ya. Life is a series of compromises, I think. We try to make the best decisions to reach our goal. Good decisions depend upon knowledge. Cumulative knowledge says if you do this and this you can expect this and this. Actions have consequences and consequences have further consequences and further consequences have further, further consequences. It's the idea behind the thought that a butterfly in Brazil flapping its wings will create a hurricane off the coast of Africa.

I was told a story about a contractor doing a SIM job. He laid out a 1000' roll of, as I remember, 5/8" pex in a side walk, a commercial job, and was wondering why the snow didn't all melt. That was a mistake cast in concrete. He was offered a solution, I never knew if it worked. That's the kind of mistakes we don't want to make. So, gpjazz needs as much knowledge as he can get.

Hey, when is Idronics going to do a SIMs issue?

hot_rod

HomerJSmith said:

Problem with short loops is you end up with a lot of runs from the manifold.

Ya. Life is a series of compromises, I think. We try to make the best decisions to reach our goal. Good decisions depend upon knowledge. Cumulative knowledge says if you do this and this you can expect this and this. Actions have consequences and consequences have further consequences and further consequences have further, further consequences. It's the idea behind the thought that a butterfly in Brazil flapping its wings will create a hurricane off the coast of Africa. I was told a story about a contractor doing a SIM job. He laid out a 1000' roll of, as I remember, 5/8" pex in a side walk, a commercial job, and was wondering why the snow didn't all melt. That was a mistake cast in concrete. He was offered a solution, I never knew if it worked. That's the kind of mistakes we don't want to make. So, gpjazz need as much knowledge he can get. Hey, when is Idronics going to do a SIMs issue?

Early in my career I made the mistake of doing an entire driveway in a high altitude ski resort town, 1/2”. Rubber hose to boot😫. The Grundfos rep was able to come up with a salvage. It had a huge vertical shaft 1 hp pump on it. The motor and flow velocity alone probably
 melted the snow.

I was impressed on how much you can gain with a flow reverser on long loop systems. Sneak up on the load from two directions. It makes a radiant slab somewhat live-able 

Max has some good SIM experience from his Rehau days. He could come up with a good Idronics issue. We did a SIM Coffee with Caleffi a few years bock

HomerJSmith

Ya, that was the solution, I was told, a flow reverser.

gpjazz

Thank you for all the replies and information... I realize there are a lot of things different with snow melt as opposed to interior radiant floor. So I have a lot of things to figure out before I decide to proceed with this project.
One other thing I've thought about is the higher total btu requirement, and the fact that customer only has propane. So they would likely have a problem with keeping L.P. in the tank. These are things I'll have to discuss with them

hot_rod

gpjazz said:

Thank you for all the replies and information... I realize there are a lot of things different with snow melt as opposed to interior radiant floor. So I have a lot of things to figure out before I decide to proceed with this project.
One other thing I've thought about is the higher total btu requirement, and the fact that customer only has propane. So they would likely have a problem with keeping L.P. in the tank. These are things I'll have to discuss with them

Ballpark the square footage of the melt area, times 125 btu/ sq foot.

So lets say 1400 sq ft X 125 = 181,000 btu/ hr.
A gallon of LP is around 91,000 btu.
So around 2 gallons per hour to melt snow.
500 gallon LP tanks fill to 400 gallons, so that is the rough math on what it might take to fuel the beast.

If you start the slab after snow starts to accumulate it might take hours to see a clear drive. If it continues to snow it could run for days So nearly 50 gallons to run 24 hours.

If LP is 3 bucks a gallon, approximately $6/ hr. to melt, $144 per 24 hours.

Plus the install cost of the system. It's not a cheap date.

leonz

Another thing that should be considered is anthracite coal for a fuel source to fire a small coal stoke boiler.

The boiler could be fired to a high limit of 160 Degrees Fahrenheit and use a covered livestock water tank to place a coil or coils in it to temper the heating loop temperature before it is sent to the concrete.

gpjazz

leonz said:

Another thing that should be considered is anthracite coal for a fuel source to fire a small coal stoke boiler.

The boiler could be fired to a high limit of 160 Degrees Fahrenheit and use a covered livestock water tank to place a coil or coils in it to temper the heating loop temperature before it is sent to the concrete.

Well, the coal is out of the question... The whole reason the customer is interested in this is because they are getting older and can't shovel snow anymore. So no shoveling coal either. LOL

Matt_67

Lots of good advice here, the conversations before hand with the owner are really important.  Manual vs automatic operation, how long it will take snow to melt depending on how and when it is started, that kind of thing.  Some people expect a snowmelt system to keep their slab completely snow free all the time which isn’t likely to happen unless it’s class 3.  Also I had an unfortunate situation where an insulated slab dropped below freezing during a cold fall night and when it warmed up in the morning condensed and froze.  It was the most slippery slab I’ve ever been on.  The concrete that met up to it didn’t freeze so it was only the snowmelt area that was slippery.  It was at a bank and now they idle the slab.  It wasn’t my best day.  So that kind of information helps with future expectations.

hot_rod

Then other issue to consider is where the melt waters will go. If they end up frozen at at the bottom of a sloped driveway for instance, that can become a big problem. An ice dam that the road department doesn't want to deal with either.

CBRob

Hey heating help friends.
Had a couple more questions about snow melt systems and thought I would jump on this existing thread.
Was watching a video from the Watts company about snowmelt systems.
Their technical guy with lots of snow melt system experience made a couple of statements about slab insulation that left me wondering..
I think he wasn't saying that compressive strength of dow board varied by it's thickness.

That one inch dow board would carry less of a load than 2-in dow board.
Wouldn't the compressive strength 
Be the same psi for either thickness?

He also says real world experience shows very little return on your investment by using any foam at all.
That the heat is naturally going to go towards the colder surface, which anytime there is snow will be up..
His little formula suggested it would take 75 years to recoup the money spent on dow board for a typical system.
Is there any science to back that up?
Does your experience support that?

Also been thinking about hydraulic separation for a dedicated snowmelt boiler. 
I think from what I've learned from you guys that the main goal of primary secondary loops on snow melt is to keep adequate flow and/or temp going to the boiler.
If it wasn't for thick glycol and some boilers not liking the cold shock of fluid coming from the slab that it would be feasible to run snow melt on one loop.
So by doing closely spaced tees or a low loss header we are not so much getting our hydraulic separation but just setting up a differential bypass without any kind of pressure regulator.
Am I right on that, 
The end goal could be met just by putting a bypass between the boiler pump and the system pump.

On a side note, looks like the clock might be ticking for snow melt systems going in for my neighborhood.
Crested Butte has already set up a climate tax for anything that heats the outdoors. 
I think it's around 40 or 50 bucks a square foot.
Mount Crested Butte, the ski area jurisdiction is a little less green focused, but I have a feeling it's in the future up here too.
And then the county after that.

hot_rod

That doesn't sound correct it is a pounds per square inch rating. Check with Dow on that.

CBRob

https://youtu.be/0to7wzamTpM?si=j-Gp7AbbdzqNGyvk

39 minutes in is the insulation talk. 

CBRob

I think the speaker just brought thickness into the discussion but it wasn't relevant.
Dow makes different psi rated boards in all the thicknesses.
https://www.bestmaterials.com/detail.aspx?ID=24467#:~:text=Options: Available in Compressive Strengths,40, 60, 100 PSI.