Influence of window opening vs. A/C use on fuel economy

Page 1 of 2  
It has become popular lately to claim that turning on A/C uses less fuel than opening windows. On a recent round trip to Kansas I used my handy Scan Gauge to check milage under different operating conditions.
(What else is there to do driving across Missouri and Kansas?)
I checked the calculated fuel economy over approximate ten mile segments on each tank of fuel. I used the cruise control and the calibrated Scan Gauge mph measurements. Usually I could go the whole ten miles without touching the gas or brake. I tried to be as consistent as possible.
For open windows, I started out rolling them all the way down but after a couple segments of that, I decided that no one could stand the tornado effect for long distances so I tried various partial open positions which improved ventilation without being punishing. Generally this was the rears open 4 inches and the fronts either closed or open 3 inches.
The vehicle, unfortunately, was not typical for most drivers: 1998 Odyssey 4 cylinder with a Thule car-top cargo box. Newer A/C systems and less drag-challenged vehicles may yield different results, but here goes...
Test 1 8/31/06 76 mph I35 Southbound Hilly 22.8 mpg overall (measured)
A/C        Windows        # Segments    Ave. MPG     Relative MPG Off        Closed            4        19.83        100 Off        Full Open            2        19.45         98 Off        Part Open            2        18.90         95 On        Closed            4        18.24         92
Test 2 9/4/06 75 mph I35 Northbound Hilly 19.7 mpg overall (measured)
A/C        Windows        # Segments    Ave. MPG     Relative MPG Off        Closed            5        21.62        100 Off        Part Open            3        21.27         98 On        Closed            1        20.90         97
Test 3 9/5/06 75 mph I35 Northbound Hilly 23.4 mpg overall (measured)
A/C        Windows        # Segments    Ave. MPG     Relative MPG Off        Closed            3        20.43        100 Off        Part Open            2        20.65        101 On        Closed            3        19.27         94
Test 4 9/5/06 73 mph I55 Northbound Flat 21.6 mpg overall (measured)
A/C        Windows        # Segments    Ave. MPG     Relative MPG Off        Closed            6        22.22        100 Off        Part Open            5        21.68         98 On        Closed            5        20.82         94
Overall, weighted by number of segments relative to windows up, A/C off, fuel efficiency was:
Windows open        98% A/C on            94%
In this test, turning on the A/C cost three times as much fuel as opening the windows. There was no indication that it made any difference whether the windows were wide open, rears down 4 inches or front and rear both down 3 - 4 inches, but there was limited testing of this, and the effect was small in any event.
As noted before the vehicle may not be typical, but these were the results. One further point of interest; shortly after I started one segment I came upon a speed reduced work zone. I didn't use the segment in the above calculations but I noted that average speed was 63 mph and average fuel consumption was 26.4 mpg. It appears that the difference between going 60 and going 75 was about 4 mpg.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

The Mythbusters did a controlled experiment, with the result being windows down = a lot better mileage. http://cartalk.com/board/showthreaded.php?Cat=0&Number41160&page=6&vc=1
The link above leads to CarTalk's forums. The post notes that their computer models said the mileage should be practically the same. The real experiment showed otherwise.
-- R Flowers
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Windows down messes up my wife's hair.
A few miles per gallon is cheap for my peace of mind.
Just my .02
Spdloader
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Big haired wife?
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
wrote:

Let me guess - you're not married?
-- R Flowers
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

I am. I am fortunate to be married to a woman who doesn't worry about her hair.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Nope. Just particular.
Spdloader
wrote:

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Spdloader wrote:

Yeah, you have to factor in the cost of a trip to the hairdresser too!
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On Wed, 06 Sep 2006 19:41:47 GMT, snipped-for-privacy@mindspring.com (Gordon McGrew) wrote:

Interesting information, but when it's 105 when I leave work, I'll use my AC and just pay the difference. Better than being covered in sweat and having all the dirt stuck to me.
Average Hi temp for August in my area0 degrees. Thanks, but I'll just keep the windows up!!!
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On Wed, 06 Sep 2006 19:41:47 GMT, snipped-for-privacy@mindspring.com (Gordon McGrew) wrote:

You sir are a gentleman and a scholar.
However, the numbers are small enough, say 2%, that on a $50 tank of gas we're talking $1.00. And you sure can't hear the radio with the windows open more than a crack at speed.
I do guess the window numbers would be more significant for an Accord without the antlers and big slab sides, might get it up to, oh, who knows, 5%?!? But presumably the A/C numbers would be the same, which would validate the claim which your experiment did not, but would still be just about $1.00/tank in the other direction.
OTOH, windows would impose less overhead at slow speeds, say in stop and go, ... oh, where will it end?!
J.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
JXStern wrote:

I'm with you on this one - yeah, there's a MEASURABLE difference, but is it really that NOTICEABLE to the average driver? And is the difference worth the comfort?
It's kinda like the amusement I get from seeing people lining up or going out of their way to go to a gas station that has prices a few cents less per liter - with my 50 liter tank, a two-cent difference means a savings of a whole $1.00, IF I have to fill up from empty. It's just not worth $1.00 of my time to sit and wait in line or to go out of my way.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
wrote:

The difference would be about $2 per tank for this vehicle or about fifty cents an hour. Well worth it when needed - which it really wasn't on this trip.
If this test were done in an Accord or Civic or even the Ody without the roof box, the cost per hour might be similar but larger as a percentage of total fuel cost.

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Gordon McGrew wrote:

Fair'nuff... my Accord gives me usually 500-550km on a tank (80% city driving), which works out to around 28mpg (alright for an '87 with almost 420,000km), so it really isn't worth the difference for me.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
JXStern wrote:

It is good information, but the true differences are difficult to assess, as many segments are averaged together, so one cannot calculate the standard deviations for each group. It does seem that on 9/7 there was less head wind.....
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
wrote:

Better check the date. Tests were run on 8/31, 9/4 and (mostly) 9/5.
The behavior of the Scan Gauge is a little flaky in regards to measuring fuel consumption. I recalibrate it at every fuel stop and it can vary by 10% or more. Since the calibration changed between test sessions, you can't compare the absolute milage figures between sessions. Also, gradually changing terrain, elevation, wind and different test speeds complicate the situation. That is why I looked at relative fuel economy between the sessions. Below is the individual session data in case you or someone else wants to do more sophisticated analysis. Just keep the above in mind.
Test 1 8/31 76 mph
A/C        Window    m/g        miles on        closed    18.2        10.0 off        closed    19.5        10.0 off        wide opn    20.2        10.0 off        closed    21.2        10.0 on        closed    17.4        10.0 off        closed    18.9        10.0 off        wide opn    18.7        10.0 on        closed    18.8        10.5 off        closed    19.7        10.0 off        F3 R3*    19.3        10.0 on        closed    19.1        10.0 off        R6        18.5        10.0
* F3 R3 = front window down 3", Rear window down 3"
Test 2 9/4 75 mph
A/C        Window    m/g        miles off        closed    21.0        10.0 off        R4        20.0        10.0 off        closed    22.5         8.7 on        closed    20.9        10.0 off        closed    22.4        11.4 off        F2 R4    22.4        10.0 off        F2 R4    21.4        11.4 off        closed    21.7        10.0 off        closed    21.0        10.4
Test 3 9/5 75 mph
A/C        Window    m/g        miles off        closed    19.6        10.0 off        R4        20.5        10.0 on        closed    18.5         7.8 off        closed    20.0        10.0 on        closed    19.2        10.2 off        R4        20.8        10.1 off        closed    21.7        10.0 on        closed    20.1        10.0
Test 4 9/5 73 mph flatland
A/C        Window    m/g        miles off        closed    21.6        10.0 off        F3 R4    21.1        10.2 on        closed    21.2        10.0 off        closed    22.5        10.0 off        F2 R4    22.4        10.9 on        closed    20.6        11.6 off        closed    21.4        10.0 off        F3 R4    21.7        10.0 on        closed    20.6        10.0 off        closed    22.8        13.7 off        F3 R4    21.4        10.1 on        closed    20.3        10.5 off        closed    21.6        12.0 off        F3 R4    21.8        11.0 on        closed    21.4        10.0 off        closed    23.4        13.0
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Gordon McGrew wrote:

My bad on the dates. 7.2% loss between air on and air off with windows closed, with almost identical standard deviations with decent sample sizes.
It looks like with windows partially open is just about the same as them closed (.4% difference probably insignificant due to standard test error) until you get to the 6" opening in the rear (though there is only 1 sample), which is just about the same as with them wide open (again, small sample size). 6" to wide open the same as with the air on and closed.
Probably could use more "open window" numbers just to verify, but it makes sense if you look at some simple HVAC calculations for orifice flow....
A 14" diameter that has about 154 sq/in of area (semi close guess to a car window, I haven't actually measured one) with a round edge discharge co-efficient has a flow rate of 31K CFM at 1 PSI, 70k CFM at 10 psi and almost 100K CFM at 20 psi (sharp edge drops numbers by 40%). I'm not sure about the effects of the angle of the flow on the opening and I haven't measured the air pressure entering a car through an open window moving at 70 mph, but I'm sure it would be quite high (and then multiple the CFM x 4!) That is a lot of drag......
I'll take the 7.2% loss without the wind howling in my ears!
Thanks for the clarifications. Good real world application.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Ditto.
What Gordon found appears to be consistent with some study (or a summary of a study) I read not too long ago. Tom and Ray of "Car Talk" may have cited it at their web site, if memory serves. Namely, even when moving at highway speeds, not using the A/C and leaving windows open for some cooling is best.
Plenty of studies on the net on this. E.g.
--
Several sources claim that closing windows and using a car's
air conditioner will provide better fuel economy at freeway
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Great job. Please test next without the luggage on top and replace 4 cylinder with 6.

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Great Post, it's something I always wanted to do but didn't. Question, during A/C winows closed, was the Vent on Re-Circulate of Fresh Air Position Thanks
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On 6 Sep 2006 17:32:02 -0700, snipped-for-privacy@hotmail.com wrote:

Fresh Air.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Related Threads

    Motorsforum.com is a website by car enthusiasts for car enthusiasts. It is not affiliated with any of the car or spare part manufacturers or car dealers discussed here. All logos and trade names are the property of their respective owners.