How much hotter is a black car than a white car in the sun?
Everyone knows that feeling of getting into a car that has been sitting closed up in the sun, the closest many of us will come to the flames of hell. Nobody likes it, and everyone knows that the darker the paint color, the hotter it will be. So is it completely crazy to buy a black or dark colored car? Probably millions of people have made car color purchasing decisions based on the idea that dark paint will just be too hot. Some of those people liked the look of dark colors and would have bought one if it weren't for the heat factor.
When I was buying my Toyota Prius, I was caught in this quandary. The two colors that I liked were the light Silver color, and a darker green-grey color called Tideland. My only concern was that the Tideland being darker might just be too hot. I searched the web and asked lots of folks looking for some kind of scientific, quantitative number... just how much warmer will dark colors be? Nobody seemed to really know, and I couldn't find any test results for different paint colors on the web.
So I did the tests myself.
For those who just want to know the results of the test, I'll tell you. The glass temperature (which I assume to be a proxy for interior temperature) varied only 5 to 6 degrees between the black and white cars on average. So I conclude that the interior temperature only varied somewhere in the neighborhood of 5 to 6 degrees. The paint temperature, however, varied by about 55 degrees, which is ten times as much. So it is indeed true that black paint is much hotter than white paint, but the interior of the black car isn't that much warmer than the white car.
I bought an infrared thermometer, the DeltaTrak Thermotrace mini. It allows me to measure the temperature of any surface without contact to the surface. I just point the thermometer at the surface from about 6 inches away, and press the button. It reads the temperature out on the LCD screen.
I went to Toyota 101 in Redwood City, CA on 3/28/04. There were a dozen 2004 Camrys in the back lot, arranged in a nice 3x4 grid pattern and all facing the same direction. I made sure that they were all directly exposed to the sun, with no major shadows on any of them. The sun was relatively high in the clear sky. I took measurements between 1:10pm and 2:14pm. For each car I took seven measurements, standardizing on the locations so that it would be repeatable. To determine the temperature of the painted metal itself, I measured the Trunk top, Roof, and Hood. In order to indirectly measure the temperature inside the car, I took measurements of the Windshield, Left Front Window, Right Front Window, and the Back Window. I always measured the center of these areas, and held the sensor a fixed distance away and perpendicular to the surface.
Observant folks will notice that this picture is actually from a Corolla. Before I did the test whose results are reported here, I tried out my measurement technique at a few other dealers on Corollas, Infinitis, and Lexuses. The Camry test reported here had the best conditions in terms of number of cars facing the same direction and under the same lighting conditions.
The reason for measuring the windows instead of the interior of the car is simply because you have to open the car to measure the inside. Opening the car lets the air out, which would change the temperature of the car, and affect the results. The windows of a car are in direct contact with the interior, and so I theorize that differences in interior temperature would produce corresponding differences in glass temperature.
I did three rounds of measurements; first I measured all the cars once, then started over and measured them all a second, and then a third time.
The outside air temperature was approximately 80 degrees F. At 2:15pm just after measuring the cars, the pavement in the sun measured 111 degrees, shaded pavement was 72 degrees, a shaded white cinder block wall was 68 degrees, and a shaded metal wall was 73 degrees. So this is a typical warm but not unreasonable day.
While I was doing the testing, some of the cars got opened by the salespeople giving test drives. Any car that was opened was disqualified from the test, and excluded from the results that are posted here. Five of the twelve cars were disqualified because they were opened, so I'm only reporting results for seven cars.
The data was entered into Excel spreadsheets, and the three readings were averaged to produce the following table. The paint chips in the table are from the Toyota site for those colors (except for Crystal White on the Special Edition Camry, which isn't on their site).
|Car Color||Trunk||Back Window||Roof||Right Window||Windshield||Hood||Left Window|
Then for each car, the averages were calculated for:
The paint temp, which includes Trunk, Roof, and Hood.
The glass temp, which includes Windshield, Left and Right Windows, and Back Window.
The following table has those results, along with the factory paint color codes/names and interior colors. The interior color that I call Gray is called "Taupe" by Toyota.
|Average Paint||Average Glass||Paint Color||Interior||Toyota Color Code/Name||Sunroof||Model|
|110.6||109.5||Light Gray||Gray||1C8/Lunar Mist Metallic||No||XLE|
|132.7||110.9||Dark Red||Black||3Q3/Salsa Red||Yes||SEV6|
|87.6||106.4||Pearl White||Tan||Crystal White||Yes||LE Special Ed|
|131.7||108.3||Dark Gray||Gray||1E3/Phantom Gray Pearl||No||LE|
As you can see, the glass temperature ranges from 106.4 degrees F on the coolest white car to 114.0 on the hottest black car, a worst case difference of only 7.6 degrees. The difference between the average white glass temperature and the average black temperature is 113.1-107.5 = 5.6. It all depends on your tolerance for heat, but to me 5 to 6 degrees more for a black versus white car is not too bad. I don't think I'd be much happier if the car was 107.5 degrees for a few minutes after I return to the car rather than 113.1 degrees for those few minutes.
One concern in this type of testing is perhaps the glass temperature is not determined primarily by the internal temperature. Perhaps it's determined primarily by the ambient temperature. However I believe this to be unlikely, since the glass temperatures around 110 degrees are significantly higher than the ambient temperature of 80 degrees. The windows wouldn't remain that hot if they were getting their heat from just the ambient air.
Another concern with this type of testing is perhaps the glass temperature is determined by the sun heating the windows directly more than from the interior air heating them. The fact that the left window measured about 15-20 degrees lower than the right window (due to the sun coming more from the left) would support that concern. Car windows filter some percentage of ultraviolet rays out of the light that passes through, and this filtered light is likely converted to heat in the window. However, this concern is mitigated by the fact that for all windows the white cars are cooler than the black cars, and by about the same amount. The glass heating doesn't seem to be responsible for the differences in recorded temperatures between the cars. Something else to consider is that if 20% of the UV is filtered by the windows, then 80% hits the interior of the car, and most of that gets absorbed and converted to heat inside the car.
We can see that the painted surfaces of the car did indeed vary by a large amount between the white and black cars. The difference between the average black and average white cars is: 144.85- 90.25 = 54.6 degrees. This is about ten times what the glass variation was.
How do we explain the difference between the paint temperatures and the glass temperatures? On first glance it seems like if the outside of the car is much hotter, then the interior must be much hotter too. I think there are several effects at work here:
1. Convection. The hotter the paint is, the more the air tends to rise around it, carrying the heat away from the interior. Also, any wind will tend to move the paint heat away from the car interior.
2. Insulation. The paint is physically separated from the interior in the case of the hood and trunk. And in the case of the roof, the roof liner has insulation that keeps the heat away.
You can think of the oven in your kitchen. No matter how long you keep the oven on at 425 degrees, you don't have to worry about the kitchen itself getting up to 425 degrees; convection takes some of the air from the oven out the exhaust vent, and insulation in the oven keeps heat inside. The oven does raise the temperature of the kitchen, but nowhere near 425 degrees.
All this can explain why the paint doesn't affect the interior temperature as much, but then why does the interior temperature rise in the sun anyway? I believe that direct absorption of the sun is the main reason that the cabins get hot. The sun gets absorbed partially by the UV filter in the windows, and largely by the surfaces of the car interior. The sun that hits the dashboard or seats turns into heat, and because it's happening directly in the interior of the car, it's directly affecting the interior temperature. This explains why putting a silver sun deflector in the front windshield helps so much to keep the car cool; it keeps the sun from being absorbed in the interior of the car.
Comparison with public perception
Before doing these tests, I did a poll on the prius yahoo group, asking what people thought the difference in interior temperatures would be between a black and a white car.
Here are the poll results:
How many degrees hotter do you think the interior air of a black car will be in comparison to a white car. Assume both cars have been left closed an hour or more to stabilize with no reflectors or other window coverings, in direct sun at 1pm, ambient temperature 75 degrees outside. The interior of both cars is the same light neutral color. Temperatures in Fahrenheit.
So 5 people guessed too low, and 23 people guessed too high. This indicates to me that the public's perception is not in line with reality. People think black cars are more uncomfortably hot inside than they actually are.
This is interesting partially because because public perception makes market realities. Certain car colors are actually worth more than other car colors on the used car market; many pricing engines require you to enter the color in order to get a quote. If these price differentials are even partially based on false perceptions around sun warming, then you could save money buying a darker used car if you like that color.
This document was written 5/08/05. Comments? contact me.