Confused

Wendell

I understand the principle behind radiant heat and color. (try heating an aluminum non-stick pan and hold it up in front of your face. The silvery aluminum side will radiate much less heat than the darker non-stick side).



What confuses me is how that would affect radiator efficiency. Wouldn't the total thermal energy that's stored in a radiator make its way into the room regardless of the color. A black radiator would simply cool down faster than a white one, but both would give up their heat to the room eventually?



In my backward way of thinking, wouldn't the color of the INTERIOR of the radiator be more important. If the interior were black it would absorb the steam's heat faster...

Jamie Hall

I can see how the confusion might arise...

but you are forgetting that the system is dynamic.  The whole process is that steam is allowed into the radiator (for the sake of simplification, ignore how much of the radiator is filled for the moment).  Once in the radiator, it is in physcial, conductive and to some extent convective contact with the iron; radiant heat transfer plays almost no part here.  The radiator iron then conducts heat through itself, and that heat is lost at some rate to the space -- partly convective, partly radiation.  Now... the amount of heat which is radiated to the space is, as we are thinking, related to the size of the radiator and its surface finish; shiny equals less radiation.  If we go back up the path to the inside of the iron, and things are in equilibrium, then the amount of heat radiated (and convected) from the outside must exactly equal the amount of heat being transferred from the steam to the iron, by conduction (the heat comes from the latent heat of condensation of the steam).  So... less heat radiated, less steam condensed.



As I said, radiation is almost negligible inside the radiator; what is happening is the steam is convected to the iron, which is fractionally cooler, and when in contact with the iron it condenses to liquid water -- also in contact with the iron -- and the heat released by condensation is transferred by conduction into the iron.



That's when the boiler is operating --  which is what we are really worried about.



You are quite right that once the boiler is shut off, the heat stored in the radiator will all make its way into the space, regardless of the surface finish.  But that's the stored heat, and is a relatively minor part of the equation.

Wendell

Thanks, Jamie

Nicely explained.

Wendell

Just had a thought

I have a silver painted radiator that completely fills with steam during a heating cycle (the return pipe gets hot to the touch while other returns are only just warm). I'll paint it with flat black stove paint and see if the return cools down some... in theory, it should.

Dave in QCA

Clouding Jamie's Clarity

Jamies explanation is perhaps the best stated explanation I have read regarding surface color and finish of a radiator and its thermal efficiency.  Anything that I add will only diminish the clarity and create confusion.  But, I'm going to do it anyway.  



I think that one thing that initially causes some misunderstanding is that when we use the word efficiency, we also think of cost.  So when we talking about improving the thermal efficiency of a radiator, some will assume that by doing so, their heating bill will go down. This rarely would be the case, and if it did occur, would more likely be from other indirect causes.   When we speak of thermal efficiency of a radiator, we are speaking in terms of the total amount of heat that it will dissimenate to the space per sq ft of radiation surface.  



So, if we have a 50 sq ft radiator that is painted with very glossy silver colored paint, it will not put out as much heat as the same radiator with a flat finish dark color of paint.  I don't know how much difference it is, but it's not a huge amount. So, the net result, using rough numbers for illustrative purposes, might be that the radiator now heats the same as a 48 sq ft radiator.  The difference is not lost into outer space, it still makes its way into the room, but just at a slower rate, much the same as installing a very slightly smaller radiator.



The size of the radiator, obviously will affect how much heat is delivered into the room and how much steam is condensed inside the radiator.  But, minor adjustements to the sizes of radiators, as long as the total radiation is still within the range that is a proper match for the boiler, will not affect operating cost to any measurable extent.   (I think.....)

Charlie from wmass

Radiators outputs are measured

in b.t.u. s and time. This is why the color matters. The heat is lost from the structure at a certain speed and needs refilled at a certain speed.

Wendell

Thanks, Charlie

The word "time" is helping me wrap my mind around the dynamics.



It's not as simple as heating a radiator and waiting for it to cool off.



I'm trying to visualize how color would affect my set up (for those that don't know, it's a two pipe system with open returns and no traps).



If my returns remain cold at the end of a cycle (felt at the radiators not in the basement), then all of the steam that has entered a radiator has condensed and converted to heat... and that heat should eventually make its way into the room regardless of color.



If my returns were getting warm, then a color change might help in radiating and condensing heat faster. Am I still seeing this wrong?

Jamie Hall

In a sense, yes

or rather, no -- you are seeing it right.  The faster you can radiate heat away from the radiator, the more steam you can condense in it.  The problem is that changing the colour of the radiator isn't going to make that much difference (it will make a difference, don't get me wrong -- just not a great big difference).



Two pipe systems with open returns are absurdly sensitive to whatever is controlling steam flow into the radiators -- whether it's an adjustable valve or an orifice.  If the pressure is allowed to build at all, they also are very sensitive to that (doubling the pressure -- say from 4 ounces to 8 -- increase the flow about four times, for example).  If all of your returns are warm, I would suggest making sure you are operating on very low pressure and see if that helps.  On the other hand, if only some radiators have warm returns (implying that they are hot all the way across) and others don't, you might try closing the valves somewhat on the ones which get hot all the way across first.  This will reduce the flow of steam to those radiators, and redirect it to some others which, perhaps, are not heating as well as they might.



And, of course, painting the radiators flat black would help too!  Although pretty much any colour paint other than the actual metallics works as well as any other.

Charlie from wmass

The reports list

Terra cotta as the most effective colour.

Wendell

Made a change

I sprayed my front hall radiators flat black. With all else left unchanged, last night's testing showed a shortening of the cycles by three minutes (30 degrees out). The radiators are close to, but around the corner from my thermostat. I think they look hideous… Charlie's terra cotta idea would be easier on the eyes



I've tested my system with both a 1 hour and 2 hour cycle. The overall run time, in a 24 hour period, increases by almost an hour when cycling every hour.



I adjust the cycles by manipulating the anticipator and the hallway radiators. My initial goal was to saturate the living room radiator. I found this takes 30 minutes run time. The return line is just warm to the touch, and roughly 3/4 of the surface area reaches 200 degrees with the remaining area around 185. I use this unrestricted radiator as my base and adjust down the remaining radiators until their returns are also just warm.



Two of my rooms are turned off as we're not using them. When guests stay over, I turn them back on with no apparent affect to the system's balance. The only noticeable change is an extended run time of 7 minutes.



The new painted radiators will need to be choked down slightly to increase my run time back to 1/2 hour.