Snowmelting System Preferences
JoeArchitect
| Posted in Energy, Heating & Insulation on
I’m currently working on a remodeling and upgrade of the exterior grounds of a Cathedral in Chicago. One of the elements in this project is to provide a snowmelting system under the new concrete stairs, paver courtyard, and concrete handicap ramp.
I’ve looked at electrical vs hydronic. I’ve read some of the systems installation specs and techinical material is why they all feel their system is best. I am leaning towards the electrical snowmelt system.
Just wanted to hear if anyone out there has any opinions and past experience with installation of these systems and any problems after installation.
Replies
Electricity will cost more to run over the life of the system. WHile electricity is 99% efficient it costs more per BTU than Gas. The Carnot equation is immutable in the conversion of energy from one source to another.
Carnot's rule, named after French physicist Sadi Carnot, is an early formulation of the second law of thermodynamics.
It sets essential limitations on the yield of a cyclic heat engine such as steam engines or internal combustion engines: they can extract only a certain proportion of mechanical energy from the heat of the working fluid; this maximal amount is realized by the ideal Carnot heat engine.
All of the above are links to Wikipedia. Essentially Carnot said that energy used to make another form of energy (ie Thermal, to mechanical, to Thermal...etc.) Looses a predictable amount of energy that cannot be transfered.
We burn coal or gas to turn a turbine in a power plant. That is AT BEST 60% efficient in being turned into electricity. Power transmission losses can subtract 10 to 20% loss again. Then you are turning it back into heat in the concrete steps. That quirkily enough is pretty efficient. In the end, 45 cents worth of electric heat at the sidewalk was created from one dollars worth of gas at the power plant.
System wise, it might be less costly to make an electric heating system but the customer (and the environment) will pay for it in the long run. (maybe even the short run.)
Edited 1/16/2007 1:44 pm by booch
One of the pros for an electrical system is that you just flip on switch to turn it on if you know it's going to snow, sleet. You may have a very mild winter and never need to turn it on. On the other hand, as I have read, with a hydronic system in the Chicago area, you'd have to turn the system on in by mid to late November and leave it on until late February or mid March. So, in a mild winter you don't need to turn on the electric system, no energy being used. Savings at that point.
No reason why you couldn't turn a hydronic system on and off with essentially the same frequency as an electric one.
Half of the harm that is done in this world is due to people who want to feel important. They don't mean to do harm but the harm does not interest them. --T.S. Eliot
On the contrary, the system has to be turned on to keep the fluid from freezing, thus, you need to keep the system turned on longer even when you don't have freezing snow or sleet.
Never heard of antifreeze?
Half of the harm that is done in this world is due to people who want to feel important. They don't mean to do harm but the harm does not interest them. --T.S. Eliot
Joe, have you heard of antifreeze. We use it all the time in our cars up here in the north. Actually even for cooling in the south.
You do not have to have this thing turned on all winter, polyprop glycol is the answer to freezing problems. they also make snow sensors that are set in the slab to turn it on. Go to THE WALL at heating help .com , those guys love snowmelt and radiant ?
I'm not sure (not being an electric expert). Start up for a hydronic system in snowmelting conditions can take awhile. I would imagine that electric might be a much quicker startup though again I'm not sure.I do want to note that the very good energy usage analysis posted above does not take into account heat losses. A square yard is 9 square feet and snowmelts are typically designed to product 120-150 BTUs/sq ft/hr or 1080-1350 BTUs/sq yard/hr.This would lead me to believe that the large majority of output is actually just lost heat since I can't imagine a typical snowmelt yields 10lbs of ice over a square yard in an hour, and I would at least double those energy usage estimates, maybe more, to account for those additional losses. Certainly not a scientific analysis, but I have to think that the waste heat quotient on a snowmelt is very significant.-------------------------------------
-=Northeast Radiant Technology=-
Radiant Design, Consultation, Parts Supply
http://www.NRTradiant.com
Yeah, it probably takes 10-15 for a hydronic system to get "up to speed" (a major variable being whether the boiler is already hot), while electric heat can be applied instantly. But that difference isn't very significant in most cases.
Half of the harm that is done in this world is due to people who want to feel important. They don't mean to do harm but the harm does not interest them. --T.S. Eliot
10-15 what? Hours? I've never seen a snowmelt go from cold start to melting in 15 minutes..-------------------------------------
-=Northeast Radiant Technology=-
Radiant Design, Consultation, Parts Supply
http://www.NRTradiant.com
Minutes. From the time you flip the switch until significant heat is flowing through the under-pavement coil.
Half of the harm that is done in this world is due to people who want to feel important. They don't mean to do harm but the harm does not interest them. --T.S. Eliot
"Certainly not a scientific analysis, but I have to think that the waste heat quotient on a snowmelt is very significant."
The waste is so great that some states forbid the use of any "primary" heat (i.e. that which is not waste heat recovery based) for snowmelting. Wisconsin's energy code does not allow snowmelt except from a waste heat source.
Very true. The problem is the heat transfer material. Water as a transfer agent between the boiler and the walkway is a problem because it would freeze or solidify in the tubes.
I haven't kept up with the alternative boiler fluids available but that would be a question to ask. Jack of all trades and master of none - you got a problem with that?
The usual approach is to have a intermediate heat exchanger, allowing a separate loop of antifreeze solution. This drives up the installed cost a bit, but doesn't significantly affect efficiency.
Half of the harm that is done in this world is due to people who want to feel important. They don't mean to do harm but the harm does not interest them. --T.S. Eliot
DanH. If a guy wanted to do a very small residential hydronic snow melt system for say a walk way, could you just use a water heater, filled with an antifreeze mix? Of course you need an expansion tank and reflow pump or pumps.
Why would you need to turn a hydronic system on in Nov and leave it on, while an electric system could be turned on and off at will, as you say? Both need to heat up the mass of concrete before snow or ice will melt, and both require X BTUs. ALl should be equal, unless, I suppose, the electircal ssystem is built very shallow into the concrete and therefore more energy is more immediately available to teh surface of the concrete, but even here teh entire mass of concrete will be heated over time.
??????
Carnot or no carnot, that is the question.
My guess would be that your original leaning toward electrical is correct.
Let us see how the numbers play, pretty easy to calculate
- ORD averages 30 inches of snow a year.
During the 1967 Chicago blizzard (last year I lived in the area!) I wanted to know how much water was in a foot of Chicago snow, so cut a cubic foot of snow out of the yard and melted it down, was left with just under 2 quarts of water.
So, in one year, your walkway needs to melt 30 inches of snow or the euqivalent of about 3 inches of ice, we can round that off to about 140 pounds of ice per square yard per year.
It takes 144 BTUs to melt a pound of ice, there are thus about 20,000 BTUs needed per year to melt snowfall.
Add 25% for re-melting tracked in snow, gives a need for 30,000 BTUs per year or about 6.6 kW-hrs per year per square yard of walkway.
Commonwealth Edison (I've not bothered to look up the current rate) hit you for say 12 cents kW-hr, so electricity would cost about 80 cents per year to melt snow, less if any type salt is also used as a de-icer.
This asumes that there is an embedded temp or snow sensor so that someone does not just turn it on and leave it on for weeks at a time.
So, say we have 300 sq yards to clear each winter, that is $240 per year.
Assuming capital costs 10% per year for the parish, the cost differential justified between the electrical resistance heaters and for boiler hot water or somthing like a heat pump at 3.0 COP for 0 F to 40+F is only $2400. Now add in maintenance cost for other than electrical system.
You know your equipment and maintenance prices in the area, go from there.
Hey, with global warming (a woodshed thread) you wont even need the system in a few years.
edit PS: re turning it on and off.
Yep, you can get the alter guild ladies to flip a switch, probably not to go into the boiler room (that still has coal dust from the old stoker??) and turn a big wonkin' valve.
Edited 1/16/2007 4:19 pm ET by junkhound
Wow! I'm impressed with your numbers. Thanks for your input.
impressed with your numbers
thanks for the compliment, but pretty crude actually, just a ballpark estimate - if you have a really big area to do the economics on, such as a whole parking lot or some such, then there is a whole 'nuther set of numbers to crunch on the issue of insulation under the slab or allowing ground heat (which also cools down the heaters in really cold weather, etc. ) - very climate dependent.
Also, from your comments to Dan, I'm guessing the vendors you havae contacted for takeoffs on an existing heating system have very simple systems to try to compete on first cost. Like Dan implied, you can have solenoid controlled valves where you do only have to throw an electrical switch. Plus, glycol mixtures are good to -56F, below any Chicago winter air temp much less ground temp, but that adds another heat exchanger and cost to the system. Probably if the snow melt area is more than a few hundred yards, you would want to look at higher complexity systems to acheive life cycle cost savings. +-
Junkhound is right in many respects... But Heating the concrete is not free. Count that as part of the melting ice equation. Convection from the local air temperature above the concrete will also impede the melting of the snow/ice.
The math is good but incomplete.Jack of all trades and master of none - you got a problem with that?
The math is .... incomplete.
Correct, see post #10.
For a complete analysis, one would need to do at least a Maxwell 2D or equivalent FEA including thermal interfaces with the soil, air convection, heat excahnger efficiencies, thermal gradients in the soil, differentials in radiation loss beteen night and day, solar gain in the day, time of day an day of year, sky conditions, differences in melting point with salt additions in various concentration, wind speed, sublimation cooling effects, relative humidity, etc. Those are just the secondary effects 'off the top of head'.
Think you'd get charged a few hours eng time for a complete analysis <G>
But the big question: Will Schrodinger's cat survive?
Half of the harm that is done in this world is due to people who want to feel important. They don't mean to do harm but the harm does not interest them. --T.S. Eliot
Schrodinger's cat
Ah, a no-brainer.
If glycol is used in a hydro system, the cat will likely die from glycol poisoning anyway due to a leak. Thus, the cat has a slightly better quantum chance with an electrically heated walkway, as long as it is not near Chernobyl.
Throw out a few more terms and someone here will have to send you a check. It would be hard to justify the time of a full analysis unless you were putting that sidewalk in front of every church in the land.
It would be safe to say the short term expense of the heater wire in the concrete will win out. I sit on a parish council in Wisconsin and witness these decisions every month. The short term expense wins out.
Jack of all trades and master of none - you got a problem with that?
One factor that isn't in your figures (and I can't readily evaluate) is the heat lost to heating the pavement. Before you can melt snow you must inject enough heat into the pavement to raise its temperature above 32F, and, after the snow is melted, that energy is lost into the atmosphere. Plus there's a certain amount of loss (probably fairly constant per hour of operation) into the ground via conduction.Before making a decision on the system I think it would be wise to get some "real world" numbers for operating cost -- numbers that take these factors into account.
Half of the harm that is done in this world is due to people who want to feel important. They don't mean to do harm but the harm does not interest them. --T.S. Eliot
Yep, I left out the heating of the pavement, couldn't do it all in my head as I wrote, so assumed it all started out at 32F.
It takes 144 BTUs to melt a pound of ice, it only takes 12 Btus to bring a pound of pavement and a pound of ice from 22F to 32F. Heat of fusion is the biggie. Conduction into the ground is a bigger factor which was also neglected - touched on in the following post about insulation under the pavement.
God melts snow and ice better then I.
A man must recognize his limitations.
hrm, really? it takes 10 btus just to raise a pound of water from 22 to 32 (if it could remain liquid, of course). I know ice is less dense than water, but are you suggesting pavement is too?-------------------------------------
-=Northeast Radiant Technology=-
Radiant Design, Consultation, Parts Supply
http://www.NRTradiant.com
specific heat of concrete much less than water; H2O = 1, masonry = 0.2 .
1.2X10 = 12
You're right, I was thinking equal volumes (where density comes into play) but you were talking equal weights of concrete and ice.*smacks forehead*-------------------------------------
-=Northeast Radiant Technology=-
Radiant Design, Consultation, Parts Supply
http://www.NRTradiant.com
junk, specific heat? I forget my physics. I recall that it takes 1 BTU to raise one gram of water one degree C. Are you saying that it takes only .2 BTUs to raise one gram of concrete one degree C?
Units are off. It is one calorie with a small "c" that it takes to raise one cubic centimeter of water 1 degree Centigrade. The Big C calorie is the food unit we use for measuring the caloric content of food. I that is 1000 times as many little c's.
My wife's water bed in the early years had a heater that took a dump. I sat one night figuring how much body heat it would require to raise the queen sized bed 10 degrees from ambient. (thus comfortable to sleep on) The result was ridiculous. 27,000 servings of broccoli with cheese sauce would have to be consumed to do the task. I bought a heater. 47 bucks for a heater seemed so much less expensive than the sheer quantity of toilet paper that would be required with the broccoli.
Jack of all trades and master of none - you got a problem with that?
yes, yes, but the man question was does it that less energy to raise one unit of concret one degree than an equivilent unit of water? I think that was the question. grams, cals., Cals, or Kcals, BTUs, or btus, etal. LOL.
Bump this onto a satire track now:
IT takes EXACTLY the same amount of energy to add one jewel (sp on purpose) to any size or shape of water or concrete.
Thermodynamic trivia Quiz: What was the name of the waterfall where JP did his famous N-m to J experiment ??
Sallenches
Congrats, ya get the milkbones!
They were good enough reward for my dog. I'll take 'em.
mass is mass, duh!
No, really it's: "a Joule is a Joule" . It really does take a diifferent amount of energy to raise the temp of a gram water than a gram of stone.
ok, perhaps then asking a question for the thirty-seventh time will receive an answer. What is the difference in energy required to raise X quantity of concrete vs the same quantity of water.
So everyone here is a genius, but knows nothing?
You gotta define the units of your quantity, weight or volume.
By weight, concrete takes only 20% the energy for the same temp rise as water.
junk, pretty sure that in eachof the forty-seven times I have asked I have referred to weight. If not it is pretty obvious that weight is the relevant factor, actually mass if you wish to get technical. Thanks for the reply. Not sure I had a forty-eighth post in me.
http://hypertextbook.com/physics/thermal/heat-sensible/
Half of the harm that is done in this world is due to people who want to feel important. They don't mean to do harm but the harm does not interest them. --T.S. Eliot
It takes 0.2 Btu to raise the temp of a lb of concrete 1 degree F
It takes 1 Btu to raise the temp of a lb of water 1 degree F.
It takes 0.8 more Btu per lb to heat the water than the concrete by 1 degree F.
You can figure this out for any substance by looking up the specific heat.
mike. I love math, physics, science or all kinds. It is just a shame I can't keep up with all of it. As for Physics, it has been 25 years since cracking the last book. So you tend to forget the terms. Specific heat! Ya yes.
Do you know why a bottle of propane being discharged into a camp stove or barbe frosts up? I recall my chemistry better. PV over T = PV over T!
The propane valve frosting is because of the throttling done by the valve. When the valve necks down it forces the gas velocity in the valve to go up. The increased kinetic energy, due to the high velocity comes at the expense of the thermal energy (enthalpy) and the temperature drops.
Would happen even without a change in velocity. Standard Carnot cycle stuff -- adiabatic expansion: http://en.wikipedia.org/wiki/Carnot_cycle
So convenient a thing it is to be a reasonable Creature, since it enables one to find or make a Reason for everything one has a mind to do. --Benjamin Franklin
Carnot Cycle is a theoretical heat engine. The throttling process is different; no mechanical work being done, a constant enthalpy process.
The purpose of the Carnot Cycle is to tell you that the theoretical efficiency of a heat engine is fixed by the temperatures. It's natures way of telling you that not only can you not get something for nothing, you can't even break even.
The good news is that you can catch massive Striped Bass at the power plant outfalls as they reject all the heat that the Carnot Cycle demands.
Rapid expansion is roughly adiabatic, since heat can't flow in fast enough to maintain constant temperature.
So convenient a thing it is to be a reasonable Creature, since it enables one to find or make a Reason for everything one has a mind to do. --Benjamin Franklin
Adiabatic means no heat transfer. That's different than constant temperature, which is called isothermal. In theory the throttling is adiabatic, not really though because we notice the frost formation. It's not isothermal though.
If I were to insulate a throttle valve and run the propane through it the process would be adiabatic, but not isothermal. The thermodynamic conditions before and after the valve would be identical. No work was extracted. Pass steam through a turbine and it's a different story. Big difference before and after the turbine, lots of work extracted.
Pass ...... through a turbine and it's a different story.
Viola, 20 seer ac with turbine for TXV. Some large cryogenic systems do just that.
It wouldn't get cold unless there was a large degree of adiabatic expansion.If you insulated the valve then it would be closer to pure adiabatic and the gas exiting the valve would be even colder.The passage through the valve releases energy and actually heats up the gas slightly (though not anywhere near enough to offset the cooling due to adiabatic expansion). The potential energy in the compressed gas is converted to kinetic energy and then heat via turbulence in the valve.
So convenient a thing it is to be a reasonable Creature, since it enables one to find or make a Reason for everything one has a mind to do. --Benjamin Franklin
PV over T equals PV over T. As the gas is released from the bottle the presure in the bottle is reduced. Since PV divided by T has to equal PV divided by T the equation must be balanced by a reduction in Temperature.
The ideal gas law (PV/T) is not the explanation for the temp drop at the throttle. For a bottle of liquid propane the ideal gas law does not really apply because there is phase change at work as the propane boils. What applies here is the Bernoulii Equation :
½ V2 + h = constant
The "h" is the enthalpy which is a function of the temperature. V is the fluid velocity. That's the difference between this and the ideal gas law. Velocity up and enthalpy (temperature) goes down. When the fluid velocity goes back down on the other side of the valve the temp goes back up. It's all about the velocity.
I don't beleive you are correct. The bottle is what is frosting up, ie it is what is cold. There may indeed be all kinds of things going on at the throttle, but the bottle clear is dropping in temperature. PV/T = PV/T.
The bottle frosts because the propane is boiling as the pressure drops. The propane gets its heat of vaporization from the surrounding environment.
> The ideal gas law (PV/T) is not the explanation for the temp drop at the throttle. For a bottle of liquid propane the ideal gas law does not really apply because there is phase change at work as the propane boils. The "ideal gas" law obviously doesn't apply to any "real" gasses except the noble gasses. But the basic relationship between temperature, pressure, and volume still holds, it's just not a simple linear equation because the "compressibility" of the gas is not the same at all temperatures.The phase change in propane normally occurs in the bulk of the upright bottle, and this is why the tank will "sweat" below the liquid line. But generally only gas is fed through the valve (except with a hand-held propane torch) and so the cooling of the valve is due to gas expansion. It's this same principle that makes a freon refrigerator work, using a gas that is very similar to propane in its thermal characteristics.Certainly the Bernoulli principle is at work, but the high-velocity gas is only found in a small area of the valve, and any temperature drop caused by the accelerating gas would be matched by a nearly identical temperature rise as the gas decelerates, so no significant net change in temperature would occur.Saying there's a difference between the two laws is obvious -- they're only related because they both deal with gas thermodynamics and both assume an "ideal" gas. They don't describe the same relationship. (The "V" is volume in one case and velocity in the other.)
So convenient a thing it is to be a reasonable Creature, since it enables one to find or make a Reason for everything one has a mind to do. --Benjamin Franklin
It's melting the ice that sucks up the BTUs, because it changes state......Heating the pavement & heating the water can almost be ignored for simplicity (though it gets tricky in that melting snow might not be the same as melting ice?)
jrnbi,
It is true that it takes 80 cals. to melt frozen water and only 1 to raise the temperature of water 1 degree (and per someone else on this thread 1/5 of a cal for concrete.
However, while you are heating that 20 degree concrete up to 32 degrees and beyond, costing you 12+ cal, the heat is radiating into the air continually! That can't be cheap and over a not to long period of time is likely to dwarf the cost to melt the occasional snow.
I suppose this is why it is important to know whether it is affective to turn such a system on and off as needed. Did we answer that question yet?
well, once the snow is melted you will radiate a lot of waste...but with the snow on top it's a real good insulator (my house stays much warmer with snow on the roof) and with ice on top any lost heat is going into the ice...so, I think the main energy savings are in not running the thing-electric or gas-without something to melt on top of it....
Sorry to be pedantic, but...
Firstly, "centigrade" went out years ago. "Celsius" is what the rest of the world uses.
Second, it takes 1 joule (not 1 calorie) to raise the temperature of 1 millilitre (1 cubic centimetre) of water by 1 degree celsius. 1 calorie equals approximately 4.18 joules.
1 nutrition calorie is actually a kilocalorie. Thus, 1 "food" calorie equals about 4,180 joules.
If a waterbed is 100 litres = 100,000 mL * 10 degree C / 4180 = 240 nutrition calories.
I think maybe a cup of broccoli with cheese sauce should have done the trick. :-)
Regards,
Tim Ruttan
junkhound. Nice figuring. Are these electric systems some kind of surface mount system rather than imbeded in concrete? if not, I think your figures would have to in clue cost of heating the concrete up to the melting point. and unless you then turn the system off while snow is STILL on the pavement you would be losing heat to teh air long after the snow is gone.
I've been involved in a couple projects that used electric heat tracing and/or snow melting equipment. The best online reference I've found is the Tyco Thermal website. http://www.tycothermal.com/usa/english/snow_melting/
which one? electric of course, as I have no experience with plumbing.
I have installed both types
ice melting cables ...on roofs and in downspouts, these used an exterior t-stat for control
and mats, that were embedded in the pour, these were controlled by a snow sensor. that was either in the pour as well or remote, and mounted above the roof line,
....which is probably out of the question on the Cathedral...
The drawbacks to mats is the length of the tails...and the splices that may need to be made up outside of the building .....
.
.
.Imagination is more important than knowledge.....
Albert Einstein..........Wer ist jetzt der Idiot
?
The BTU output of a hydronic snow melt system can be as large as you'd like to design. More tubes, larger tubes, higher flow rates, higher temps all contribute to faster response time.
Snow melt systems have an antifreeze filled loop that can be shut off or turned on as needed.
There are many more electric radiant heat setups that have gone kaput than hydronic. Not an easy thing to fix. The hydronic is a much safer system. Much easier than trying to explain the risks to a ticked off client just out of warrantee.
Whichever route, insulate under the concrete or the cost to opperate the system can be prohibitive.
If there are concrete steps involved quicker response time can be had by minimizing the top concrete cover. Instead of a simple massive clump of concrete, rigid foam can be used to take up space and insulate the bottom. The concrete crew will need to be watched carefully or shortcuts will be taken...guaranteed.
At least locally, the price of 3/4" pex has come down so much that the cost per btu is least expensive with this route vs. 1/2".
Good melting.
Beer was created so carpenters wouldn't rule the world.
Thank you all for your responses, calculations, and opinions.
You forgot "prejudices", "suspicions", and "accusations". ;)
Half of the harm that is done in this world is due to people who want to feel important. They don't mean to do harm but the harm does not interest them. --T.S. Eliot
Of course, that's all understood.
The hydronic sounds better but if any piping leaks it is a big expense to replace it, unless you use medium sized pavers or such. Again you would want to watch the forming and pouring operation to prevent damage to the plumbing. I would think elec wires would be more failsafe. Why do elec systems fail so easily? Barmo
The hydronic sounds better but if any piping leaks it is a big expense to replace it, unless you use medium sized pavers or such. Again you would want to watch the forming and pouring operation to prevent damage to the plumbing. I would think elec wires would be more failsafe. Why do elec systems fail so easily? Barmo
Prior to the pour, each section of the pex tubing is pressurized to 100 psi and must hold that for at least 24 hours. This will show any pinhole leaks, even those too small to leak water. Then the pressure is watched prior to, during, and after the pour.
If a hole is made in the tubing during the pour it will bubble up through the concrete showing it's location. The concrete is pushed back, a union fitting installed, the pressure is run up and a little water will show any small leaks in the fitting as bubbles. I've never been on a job where this has happened, but we always have the tools on hand just in case. Electric either works or it doesn't and it's never energized while workers are installing it.
If a leak occurs down the road with pex, it's easily found by the wet spot. Electric doesn't give any clues.
As an example of how tough pex is, Warmboard radiant heat decking is installed prior to the wall framing and the exposed pex (recessed so the top of tubing is at floor level) is pressurized and simply checked occationally during the rest of construction. Litterally months of construction and foot traffic are on the exposed pex and it rarely causes a problem.
As for the reliability of the electric systems, it's been our experience that a certain percentage of the grids fail in the first year, say 2%, no probably closer to 1%. How lucky do you feel? Pouring concrete over 20 grids and the odds of one of them going south goes way up. How do you fix 'em? What if it's your bathroom instead of the hall? You can't fix it short of demo.
What other system has an acceptable failure rate that high with replacement costs so high? None.
Ask any electrican that installs them how often they hear of failures by "other electricans". Every electrican is someone's "other".
Pex is less likely to fail simply because it's flexible. A crack may occur and the slab can shift a slight amount and the pex is fine, while the electric wires have been sheared in half.
The electric grids are popular because they are easy. Simply burry the mat during the pour, run the wire to a relay hooked to a thermostat and it's instant radiant heat.
As with many choices in life, do you want easy or good?
Good heating.
Beer was created so carpenters wouldn't rule the world.
Snowmelting preference?
I have had really good results with JP-5, but you need a little gasoline to get it going.
All the mental masturbating aside, a few simple rules of thumb apply for basic snowmelt systems in Chicagoland and surrounding areas. Some of the BS figures produced here are complete nonsense and have no basis in reality. This is not rocket science and it has already been figured out by many, many people throughout the years by trial and error more so than by engineering.
30-50 btu/hr for each square foot of poured concrete (insulated with 1" EPS, 4-6" thick, std portland "6-bag mix") is required to melt snow in this area under typical conditions, in a reasonable time period. Those conditions would be 15 degF OAT, 35 degF slab surface temp. Snowmelt systems have glycol in them to prevent frezing. Usually at 30-40-% mixture is good to -20 degF. These system require experienced system design. When started, the huge mass of concrete at sub freezing temps will destry a boiler setup incorrectly, even condensing boiler HX's will twist, warp and crack. The shock is severe.
So when talk about electrics, consider a 10,000 sf (500 x 20) drive. At the middle of the practical range, that is a requirment of 400,000 btuh of output or 500,000 btuh of input for a standrd cast iron boiler. 400 MBH is the equivalent of 117 kW. In single phase, 240 volt power that is 488 amps, not a realistic option. Even for a much smaller surface, 1000 sq.ft., your're still looking at a 50 amp draw.
destroy a boiler? Is this because teh return water is quite cold and therefore socks or cracks the boiler?
Yes, that is exactly the reason. Normally, boiler operate on a finite temperature difference of 20 to 40 degrees. Also, because of the large capacity requirements and the infrequent use (in most cases) non-condensing boilers are selected for smowmelt applications. During stratup, the return water from a comfort heating application comes back at 60 degrees of greater, and it warmsup quickley. In a snowmelt system, that is more like 20 and it heats up very slowly. That kind of temperature difference will cause extreme stress and most metals will fail.