Radiator Paint Color, ...what if?

R Dougan

Okay. I've just re-read the Dept. of Commerce's 1935 findings on metallic radiator paint. So now I'm trying to wrap my head around the physical attribution for this. Is it because the microscopic metal flakes act in the same manner as a metallic first-aid blanket that reflects body heat back to an injured victim ? If so, then how does placing a top coat of non-metallic paint restore the surface to the original radiation capacity? The metal flakes are still there, albeit now under a layer of titanium oxide base, pigment and binder.

Could it have something to do with the metal flakes themselves being in contact with the atmosphere around the radiator? Possibly conducting heat from the surface back toward the source?

If so, ...then what is happening when, once again the rad is covered with a nonmetallic paint? Is the heat energy now transferred via conduction around each individual tiny metal flake to the surface of the top layer where no flakes are in contact with the atmosphere, ... so its conduction only goes in one direction instead of two?

Could it be because the specific heat of the paint binder and the specific heat of the metal flake are different?

Still with me?

With all this being thrown out there for feedback, my main reason for asking is this, ...could a clear-coat perform the same function as a non-metallic paint top coat? The layer would contain no metal flake that would penetrate to the surface yet it would still preserve the look of the metallic paint visible to the eye.

Is the roughly 17 % heating loss inconsequential. (...seems pretty consequential to me ....its 17 %, .... I do like the metallic look, ....not sure if its worth a 20% increase in fuel use to get back to 100% radiation output.

Any thoughts?, ...experience with?

Bob D.

nicholas bonham-carter

Now this is a topic we can get our teeth into!
Air cooled engines were found to run cooler if the fins were anodized, or painted black.
The dark colored paint will be heated up by conduction from the base coat of metallic paint, and then will radiate the heat outwards better. I don’t think the clear coat would do as well. I’m surprised by that 17% figure.—NBC

R Dougan

The 17% I believe, if I read it correctly, comes from the progression that 2/3 of the heat comes from the convection of air passing over the rad and 1/3 comes from actual radiation. The metallic paint was seen to reduce this by half in smaller radiators: the number drops as the radiators increased in size.
1/2 of 1/3 is 1/6 which comes out to .1667 or roughly 17%.

Now if you view your initial input as 100% when the rads are putting out 100%, this input would remain the same as the input for the metallically painted rads. So now, to bring the diminished outputting rads of .83 or 83% back to 100% the input would have to increase to 1/.83 or 1.2048 or by an increase of 20.5% Wow. That's a lot.

acwagner

Interesting article. As I understand it, this only comes into play when the radiator becomes fully heated and stays that way. The energy it takes to raise the temperature of the cast iron hasn't changed because of the paint type, merely the rate at which it is released to the room.

If a radiator is only heated to say 75% of the way across, all that will happen is a radiator with metallic paint will release that heat back into the room at a slower rate than a non-metallic painted radiator. They both release the same amount of energy to the room in the end. A 17% increase in time doesn't seem like that would adversely effect the comfort of the room in this situation. If the time is doubled or tripled, then it probably would. Just my guess, though.

But, if your radiators are heated fully up to temperature and stay that way for an extended period, then the metallic painted radiators become derated because they'll condense steam at a slower rate than similar non-metallic painted radiators. That could cause a comfort issue, or at least an issue with some rooms being hotter than others.

I know I'm not answering the exact question posed, but it seems like it depends on the application if the paint type even matters. Are the radiators you're proposing doing this to small radiators that heat up fully under normal calls for heat or large radiators that only heat up partially?

R Dougan

@acwagner ..."If a radiator is only heated to say 75% of the way across, all that will happen is a radiator with metallic paint will release that heat back into the room at a slower rate than a non-metallic painted radiator. They both release the same amount of energy to the room in the end. A 17% increase in time doesn't seem like that would adversely effect the comfort of the room in this situation. If the time is doubled or tripled, then it probably would. Just my guess, though."

I don't think this is what the finding was, it wasn't slower radiation, it was lower radiant capacity. In other words if the radiator only got hot across 75%, then that part of the radiator only emits 83 % of it total capacity at any given instant. 83% of 75% would mean the radiator is now operating at 62% of its total potential capacity* (*100% hot across all the sections and painted a nonmetallic color)

This would seem to magnify the paint color choice because to satisfy the call for heat, the burner must stay on longer. Or maybe not, ...if the radiator isn't heating all the way across this is because the call for heat is being met. But......, the call for heat could have been met sooner had the radiant capacity been higher in which case the radiator would have only had to get hot across to 62 % its sections. If the velocity of the steam is constant then it would follow that a lower volume of steam would be required for equivalent radiant output between the two. Less steam>>>>>>>>less fuel burning.

Or am I in left field.

acwagner

I don't think you are in left field. I agree the radiant capacity is decreased and generally with what you are saying. Here's how I see it: if the radiator emits heat 18% slower but still has the same amount of BTUs delivered to it compared to an identical radiator, then is there really an issue? Both give off the same total amount of heat, just at slightly different rates. The only issue comes up if the slower radiator cannot emit all the BTUs it was given before the next call for heat, or it becomes maxed out (100% full or at design day).

I guess my theory is oversimplifying reality, and every system runs differently. On most days, my boiler fires maybe once every 1.5 to 2 hours to maintain temperature, so the radiators are usually at ambient temperature by the next call for heat. If a radiator took 10 minutes longer to emit all its heat compared to another, I don't think it would make an appreciable difference, all things considered. It's probably debatable if both radiators would receive the same amount of BTUs. I think they would assuming they are only partially filling, which is the case for me.

I think the article was written during a time when the boiler firings were so close together (or it was a coal system) that an 18% change it radiant capacity could throw things off. For some people now, that's still the case. My house has had insulation and building envelop improvements so that my radiators are way oversized for the building heat loss typically seen.

For what it's worth my bathroom radiators are painted with metallic paint, and I've compared the room temperature in them to other rooms, and it's pretty uniform--within a few tenths of a degree even on cold days. I've noticed that the bathroom radiators stay warm to the touch longer, that's all.

Jamie Hall

The fundamental point is that the radiation which is emitted to the room or space is a function of the surface temperature of the object and a quality of the surface called the emissivity. That is a physical constant of the material. Dull paint has very high emissivity (any colour). Glossy paint a little less. But metallic surfaces have low to very low emissivity.

The loss from a real life radiator with metallic paint is due to that low emissivity. You have lost most of the radiative heat transfer, and are left with the convective heat transfer from the hot metal.

The only layer that matters is the outermost layer -- so if you have metallic paint as that outermost layer, you've lost. On the other hand, if you paint with normal paint over metallic, you've gained.

I suspect a clear coat would be somewhere in between. Haven't seen actual measurements on clear coats over metal flake.

R Dougan

@acwagner …they stay warm to the touch longer? That’s interesting. I wonder if that’s an insulating property of the paint or maybe its because if the radiator can’t give up the heat, it stays in the metal. Sort of a conservation of energy characteristic. That is very interesting.


@Jamie Hall Emissivity. That's the thermodynamic property I was looking for.

Color, to the human eye, is the wavelength of light that is reflected back and that which the eye detects. Light is electromagnetic radiation. What we see as the color black is the result of no visible wavelengths reflecting back to the eye, its absorbed. But black is the best at emitting radiation. White on the other hand, reflects all visible wavelengths back to the eye and has a lower emissivity.

Heat is mostly in the infrared region of the electromagnetic spectrum and probably must obey these same rules. So this must mean that the metal flake paints' emissivity (or lack thereof) is based upon the properties of EM wavelength absorption and reflection (?)

My guess is, if my progression here is actually right, a clear coat will probably have no effect whatsoever because the color at the surface would still have the same light absorption and reflection properties.

Emissivity.

Dark grey is starting to look pretty good Lol.

Thanks everybody.

Canucker

> @R Dougan said:
> @acwagner …they stay warm to the touch longer? That’s interesting. I wonder if that’s an insulating property of the paint or maybe its because if the radiator can’t give up the heat, it stays in the metal. Sort of a conservation of energy characteristic. That is very interesting.
>
>
> @Jamie Hall Emissivity. That's the thermodynamic property I was looking for.
>
> Color, to the human eye, is the wavelength of light that is reflected back and that which the eye detects. Light is electromagnetic radiation. What we see as the color black is the result of no visible wavelengths reflecting back to the eye, its absorbed. But black is the best at emitting radiation. White on the other hand, reflects all visible wavelengths back to the eye and has a lower emissivity.
>
> Heat is mostly in the infrared region of the electromagnetic spectrum and probably must obey these same rules. So this must mean that the metal flake paints' emissivity (or lack thereof) is based upon the properties of EM wavelength absorption and reflection (?)
>
> My guess is, if my progression here is actually right, a clear coat will probably have no effect whatsoever because the color at the surface would still have the same light absorption and reflection properties.
>
> Emissivity.
>
> Dark grey is starting to look pretty good Lol.
>
> Thanks everybody.

Flat black for me. Haven't done it yet but I'm hoping to feel the difference when I do

acwagner

R Dougan said:

because if the radiator can’t give up the heat, it stays in the metal. Sort of a conservation of energy characteristic. That is very interesting.

Yup, that's what my theory is. The heat is still in the radiator, just released slower. The heat isn't lost or destroyed. It's still there and available to the room. But, only when the radiator is partially filled condition. At and after 100% filled, the radiator can only accept as much heat as it puts out and will lag behind other radiators. Thus, the paint color comes into play at that point.

Kahooli

The radiation component of having a radiator based heat system is pretty important for human comfort. The air temperature can be lower and the same comfort obtained. In my opinion, Low E paint should never be used on radiators. Pipes, sure, if you don't insulate them.

R Dougan

@Kahooli I've never known paint to have E - rating data associated with it . I'm assuming the E stands for either energy or emissivity (both of which are probably directly proportional to each other). I think the actual physics of the phenomenon we're discussing can be found in this wiki article:

https://en.wikipedia.org/wiki/Kirchhoff's_law_of_thermal_radiation

...well above my paygrade but accessible enough to get the gist. At some point, one has to accept the end result of the sky being blue even if there is a complicated particle scattering, wavelength filtering mechanism that makes it so. The "how" answer to the "why" question. The same is true for this diminished radiance due to the emissivity of the surface layer. That's the take away here, ...the big "ah-ha" if you will.

Just because something sounds too fantastic to be true, doesn't mean it isn't. My problem is that I have that age old problem particularly found in 5 year olds, ...that is, a good grasp of how to recite the word "why". After all this, I feel like I can finally now say, "Ohhhhh ....now I get it".

Again thanks to Mr. Hall for pointing the way towards this answer. I'm now leaning toward the clear coat being useless in mitigating this effect. I think Canucker's plan would be far more effective, flat black, based upon this new understanding.

So as a practical example: the hotter a color gets sitting out in the sun, the better it will be at radiating the latent heat offered up to the cast iron of the radiator by the steam. Emissivity and absorption the two halves of the input/output factor of paint color.

Interesting.

Bob D.

Mark Eatherton

And then there's this thing called AUST (average unheated surface temperature)... It too affects heat transfer by radiation versus convection. And what about incident angle?

So many facets, so little time.

I can tell you that I had a nice bright shiny copper ceiling radiator in my office. You could feel heat coming from it, but it didn't feel as strong as it should have, so I documented its performance (2 degree F differential on its fluid side) and painted it flat white and DOUBLED the fluid differential to 4 degrees F. Flat white latex paint, as opposed to bright shiny copper... Whooda thunk?

ME

R Dougan

@ Mark Eatherton

Wow, ... the rabbit hole goes deeper. More proof that in application the model performs as expected.

I'd be curious to know to what degree flat black would have had on your temp. diff. Seems likely it would have been as marked an increase as was the copper to white. Did you find yourself saving energy on the input side or did energy saving result from not having to increase the output setting to meet the desired comfort level?

Mark Eatherton

I think it would have been about the same for flat black as it was for flat white. Didn't really see any decrease in energy consumption,but did see a huge increase in comfort. Energy costs are relatively negligible anyway.

ME

Canucker

I would have to agree with @Mark Eatherton on the colour effect. I suspect the paint being flat and not a glossy finish would have a larger effect than just the colour. It definitely has an effect on an IR camera

mikeg2015

Fill a white mug and black identical in size with hot coffee. Insert thermometer. Black mug will cool off faster. Did this in 7th grade science class 28 years ago. White reflect back more radiant energy. Conduction to surrounding air is the same.

srmaietta

Slightly off topic, but what about polished metal sheet behind radiators, against wall? That was mentioned in a book somewhere, reflecting more heat into the room..?

My radiators are flat white.. well, the couple I've gotten around to. And yes, they just feel great. They were painted with Zinnser BIN, the red shellac based version. May have to get some tinted dark brown to match the rooms woodwork, should make it even a little better.

~Steve

Mark Eatherton

Yes, spectral reflectivity works. This is a great application for the jun science of bubble foil bubble insulation. I've seen before an after infrared photos from the outside of the home and it does make a difference.

ME