TheSamba.com :: View topic - Bare Metal vs Painted Metal- Which Cools Better?

Tue Sep 17, 2019 5:10 pm

It's actually a well-documented phenomenon that painted metal surfaces shed heat better than bare metal surfaces, but a lot of people believe that painting engine components will actually prevent heat from escaping. I can see why someone might think that- paint is not a very good conductor of thermal energy. However, conduction is only one corner of the thermal triangle. The other two are convection, which involves a cooling fluid (such as air or water) flowing over a heated object; and radiation, which involves the emission of heat in the form of electromagnetic energy. Paint, or rather dry paint, is not a fluid, so obviously it can't remove heat through convection. But it has an enormous capacity to radiate heat- black paint especially, because it radiates heat in all of the wavelengths of the spectrum.

A few days ago, I posted a little experiment I did on my cylinder tins, comparing a bare metal one, to one painted black. I won't go into detail, but it was fairly interesting, so if you're curious you can read it here:
https://www.thesamba.com/vw/forum/viewtopic.php?t=724533.

However, it was only a very basic experiment, meant only to show proof-of-concept, and there were some comments as to the validity of the results. So, I decided to do a more thorough experiment. As I mentioned in the other post, I'd already done some other research on this topic, which was inspired by an article entitled Do Darker Objects Really Cool Faster written for the Dept. of Physics at Trinity U. in San Antonio, Texas. Basically, this experiment is just a recreation of what was done in that article, with identical results (as I fully expected them to be).

I started with three identical metal cans. One was left bare, one was painted flat black, and one was painted with VHT black wrinkle coat. Here are the paints I used:

The painted cans were covered inside and out, top to bottom. Two light/medium coats of the flat black were applied, and three heavy/medium coats of the VHT black were applied, which is how I normally use these particular products. Here is the result:

Notice in the pic with all three cans stacked together, how the inside of the bare metal can is clearly visible, but the insides of the black cans are not visible beyond the first inch or so. Very little light is being emitted from the interiors of these cans- they almost appear to be true blackbodies! In reality, though, enough light is emitted for the naked eye to see the bottoms of the black cans, but only just, and the VHT black is harder to see than the flat black.

In the first experiment, each can was separately filled with 300 ml of boiling water and sealed inside a plastic zip-lock bag. The can and bag were then placed on top of a cloth heat pad, on top of a wooden cutting board. The idea is to prevent any convection and/or conduction from aiding in cooling the can and its contents, as this is mainly a test of thermal radiation. A digital thermometer was then inserted through a small hole in the plastic bag and into the heated water. Here is the heat pad, cutting board, and thermometer that were used in the experiment:

Once the temperature of the water had cooled to 200°F (to ensure that the metal can and the water had reached thermal equilibrium), a timer was started and measurements were taken at one-minute intervals until the water reached 150°F. Ambient temp was approximately 76°F. I originally wanted to see how long it would take for each can to reach ambient temp, but I started with the bare metal can, and after six hours it was still twenty degrees or so above ambient, and I still had two more cans to do, so I said screw it, 200°F to 150°F is good enough for the Samba! Here is the rig:

And here are the results:

As you can see, the water in the painted cans cooled significantly faster, with the VHT black slightly edging out the flat black. I believe this to be a good argument for painting things like oil coolers, engine cases, and radiators, as these are basically metal cans filled with a hot liquid (it makes no difference what that liquid is, be it oil or water or hot cheddar cheese). And as I mentioned before, my results were consistent with the results of the experiment done at Trinity U. of San Antonio:

My second experiment was slightly different than the first. It also involved placing a heated object inside the respective cans, but this time the object was not allowed to come into direct contact with the cans. For this, a steel cylinder (actually, a stack of Mustang axle-shaft spacers from a Truetrac LSD, all wired together), was heated to 212°F and placed upright on the heat pad:

The three respective cans were then immediately placed over the steel cylinder, and temperature readings were taken with an IR gun at two separate locations on the can, one at the top and one on the side:

Ambient temp was again approximately 76°F, and infrared readings were taken at thirty-second intervals for 15 minutes, starting 30 seconds after the can was placed over the steel cylinder (Note: this starting point is marked as zero on the chart, even though technically 30 seconds had already passed. Each can started the experiment at 76°F). I stopped after 15 minutes due to the rapidly reduced rate of cooling, and a strong desire to do something else other than stand there watching a soup can lose 0.2°F every 30 seconds for the next six hours.

The idea for this experiment was to simulate the cooling tins surrounding the cylinders and cylinder heads of an ACVW engine, and how painted tins (as the stock tins were) are potentially better for helping the engine shed heat than bare metal tins (i.e. chrome or cadmium plated). This experiment is also a very simple one meant only to show proof-of-concept. To make it better, I would have liked to measure the rate of temperature drop in the steel cylinder at the same time as the increase and subsequent decrease in temperature of the can, but I didn't have a proper way to do that at the time. If this weren't the Samba, I would simply invoke the law of conservation of energy, meaning that if the surface of the can got hotter faster, then the source of the heat inside the can got that much colder by the same degree. However, this is the Samba, and I respect the fact that some people need to see it to believe it, and even then only if they get hit in the head with it, kicking and screaming, so I'm sure the doubting Thomas's will pounce all over this. At any rate, here are the results (feel free to question them). To plot the chart more accurately, an average was taken between the two infrared readings from the top and side of the can, to help smooth out any anomalies. As I was using an IR gun to measure the surface temp of the cans, I couldn't cover them with the zip-lock bag, so convection played a part in the rate of cooling this time:

Again, as you would expect, the black cans got hotter, faster AND shed that heat faster. In case that's difficult to see, here is another chart showing the same information, but this time superimposed on top of each other, with the start and end temperatures shown as absolute values, rather than in degrees Fahrenheit:

Interestingly enough, the flat black did a little better than the VHT black on this one, although not by much. I would have imagined the opposite to be true, what with the greater surface area of the VHT and convection playing a part. Hmm, I'll have to do some thinking on that one.

So there you have it. I've said it before and I'll say it again, if you want to be cool, do like Mick says and paint it black!

Tue Sep 17, 2019 5:37 pm

Interesting and valid experiments, thanks!

Black not only sheds more heat, it also absorbs more heat. So in actual practice, a consideration of the immediate environment is needed. If the metal needs to shed heat, but is in an area where it can receive radiated heat, it may or may not cool better when painted. The two effects mitigate one another, and it's not clear which prevails.

The sole VW-sanctioned comparison is their arctic valve covers which are chromed to NOT shed as much heat.

Hmmm.........

Tue Sep 17, 2019 7:00 pm

Well done, you have an experimental mind. do you gave a measurement on the paint thicknesses?

You may wish to do this experiment with forced air cooling and compare to your still air test. The test I read about used vw finned cylinders rather than a can, and had a blower ducted at them. He also filled the cylinders with hot liquid and measured cooling rate under forced air.

Bug On!

Tue Sep 17, 2019 7:02 pm

KTPhil wrote: Interesting and valid experiments, thanks!

Black not only sheds more heat, it also absorbs more heat. So in actual practice, a consideration of the immediate environment is needed. If the metal needs to shed heat, but is in an area where it can receive radiated heat, it may or may not cool better when painted. The two effects mitigate one another, and it's not clear which prevails.

The sole VW-sanctioned comparison is their arctic valve covers which are chromed to NOT shed as much heat.

Hmmm.........

The best tea kettles, that hold in the heat the longest, are polished metal, chrome. Just like Arctic valve covers, traps the heat

Tue Sep 17, 2019 7:08 pm

Very cool! Would be interesting to see different colors and finish...might have to stock up on soup cans though

Tue Sep 17, 2019 11:28 pm

Very interesting study. An expert in air-conditioning pointed out that IR thermometers tended to "read better off black surfaces." Does that skew the results in any way?

Wed Sep 18, 2019 1:35 am

Excellent and easy (enough) experiment.

I shared with our 7th grade science teachers-it's one of the standards on heat absorption and radiation.

So next question does it answer these two other topics on it:

https://www.thesamba.com/vw/forum/viewtopic.php?t=706783

https://www.thesamba.com/vw/forum/viewtopic.php?t=642565

Hmmm....

Wed Sep 18, 2019 4:51 am

Something not mentioned but very valid in my view is the fact that bare metal rusts with time and rust insulates. That alone is reason enough to want to give your cylinders a light coat of paint.

Try your experiment again with a rusty can!

Wed Sep 18, 2019 4:58 am

YDBD wrote: So next question does it answer these two other topics on it:

https://www.thesamba.com/vw/forum/viewtopic.php?t=706783

https://www.thesamba.com/vw/forum/viewtopic.php?t=642565

Hmmm....
My opinion is that the color of your tins, which is what these two threads are about, will have very little measurable effect on the engine temperature as thier sole purpose is to contain and direct air to the engine. The only time that they would effect the temperature I believe is after the engine is shut off. Then the color may change the cool down time.

Wed Sep 18, 2019 6:01 am

oprn wrote:
My opinion is that the color of your tins, which is what these two threads are about, will have very little measurable effect on the engine temperature as thier sole purpose is to contain and direct air to the engine. The only time that they would effect the temperature I believe is after the engine is shut off. Then the color may change the cool down time.

Ah, but the black engine tin in an enclosed VW has little airflow, thus radiating the heat away from the engine to the surrounding body and possibly the air moving in front of the engine to mix with the fresh cool air coming in through the louvers. Especially cylinder tin and exhaust tin.

Question: Can we redo the experiment with a fan on it? and then compare the data? Will the black painted can cool faster with air flow than a metal or other colored can with air flow and then will those cool faster than just static air?

Wed Sep 18, 2019 8:36 am

So my class in heat transfer was many decades ago, and I only remember enough to be thoroughly ignorant :wink:

I do remember that radiant heat transfer (ignoring convection/conduction) radiates normal to the surface (perpendicular). The cylinder fins are roughly parallel with each other, meaning much of the radiant energy is going to be re-absorbed by the next fin. Obviously there are still a lot of surfaces that radiate out from the cylinder, but for between the fins, I would guess it's close to a "wash".

It's a cool experiment, keep going !

Wed Sep 18, 2019 9:09 am

Chad1376 wrote: So my class in heat transfer was many decades ago, and I only remember enough to be thoroughly ignorant :wink:

I do remember that radiant heat transfer (ignoring convection/conduction) radiates normal to the surface (perpendicular). The cylinder fins are roughly parallel with each other, meaning much of the radiant energy is going to be re-absorbed by the next fin. Obviously there are still a lot of surfaces that radiate out from the cylinder, but for between the fins, I would guess it's close to a "wash".

It's a cool experiment, keep going !

With forced air cooling, cold air absorbs some of that radiant heat and is carried away. In still air, convection may move the heat away. in either case, fins greatly enhance cooling.

Wed Sep 18, 2019 9:18 am

Radiant heat transfer and convection are two completely different modes of moving heat energy.

Radiant heat is light - just a wavelength your eyes can't see. Shine a flashlight against a wall. Now, blow a fan across the light beam. The light still hits the wall. If air was opaque, it would absorb the radiant energy and carry it away, but air is pretty darn transparent.

Same thing with those radiant heaters. They beam the energy through the air. They don't heat the surrounding air, only objects that absorb the radiation.

Yes, the fins increase the surface area, and provide a surface for heat to be picked up and carried away by convection. What I'm saying is the color between the fins may not be important in radiation.

Wed Sep 18, 2019 9:51 am

That's why irregular surface is important -- so not all goes perpendicular, but gets scattered more. That's why the VHT textured paint did better. So, even on the fins, that effect will be there.

Wed Sep 18, 2019 10:04 am

bnam wrote: That's why irregular surface is important -- so not all goes perpendicular, but gets scattered more. That's why the VHT textured paint did better. So, even on the fins, that effect will be there.

Between the fins - I doubt it. They're too close together. It's not like crinkle paint is going to aim all surfaces so there's a straight line of sight out of the fin area.

More than likely, the crinkle paint is increasing surface area a tiny bit, aiding convection, or the paint itself has a bit higher thermal conductivity.

Wed Sep 18, 2019 10:24 am

Chad1376 wrote: bnam wrote: That's why irregular surface is important -- so not all goes perpendicular, but gets scattered more. That's why the VHT textured paint did better. So, even on the fins, that effect will be there.

Between the fins - I doubt it. They're too close together. It's not like crinkle paint is going to aim all surfaces so there's a straight line of sight out of the fin area.

More than likely, the crinkle paint is increasing surface area a tiny bit, aiding convection, or the paint itself has a bit higher thermal conductivity.

I meant that the cast surface of the fins itself is textured and radiation will not be perpendicular as the simplified picture shows. Some of if will come out. Less than on non-finned areas - absolutely. But, not all trapped.

Wed Sep 18, 2019 10:40 am

bnam wrote: Chad1376 wrote: bnam wrote: That's why irregular surface is important -- so not all goes perpendicular, but gets scattered more. That's why the VHT textured paint did better. So, even on the fins, that effect will be there.

Between the fins - I doubt it. They're too close together. It's not like crinkle paint is going to aim all surfaces so there's a straight line of sight out of the fin area.

More than likely, the crinkle paint is increasing surface area a tiny bit, aiding convection, or the paint itself has a bit higher thermal conductivity.

I meant that the cast surface of the fins itself is textured and radiation will not be perpendicular as the simplified picture shows. Some of if will come out. Less than on non-finned areas - absolutely. But, not all trapped.

Same-same. In either surface, the path of radiation is going to be random, with as much being directed back toward the cylinder, or opposing fin, as out from between the fins. The net average is still perpendicular, unless you had some crazy highly engineered and machined surface that tries to direct radiation out from the fins. At the end of the day, the surface texture is going to have such a small effect on radiation, it's not worth worrying about. Surface area and convection is where the magic happens.

Wed Sep 18, 2019 11:22 am

Did you chemically or mechanically remove the existing lining from the non painted steel can?

Wed Sep 18, 2019 2:29 pm

bnam wrote: That's why irregular surface is important -- so not all goes perpendicular, but gets scattered more. That's why the VHT textured paint did better. So, even on the fins, that effect will be there.

But does the irregular surface effect the airflow, maybe rough surface brings on tiny turbulent flow for better mixing of hot and cold air

Wed Sep 18, 2019 2:33 pm

Chad1376 wrote: Radiant heat transfer and convection are two completely different modes of moving heat energy.

Radiant heat is light - just a wavelength your eyes can't see. Shine a flashlight against a wall. Now, blow a fan across the light beam. The light still hits the wall. If air was opaque, it would absorb the radiant energy and carry it away, but air is pretty darn transparent.

Same thing with those radiant heaters. They beam the energy through the air. They don't heat the surrounding air, only objects that absorb the radiation.

Yes, the fins increase the surface area, and provide a surface for heat to be picked up and carried away by convection. What I'm saying is the color between the fins may not be important in radiation.

air is not fully transparent to IR. thus it will heat from IR. but most of the cooling is conduction to the air, cold air, cold moving air, convection.