Some graphs of temperature vs. MPG
 

I started trip logs for my drive to and from school, and I'm bored right now, so I made some graphs of the data. I record the date, distance traveled, EOC miles on that trip, MPG for the trip, and the temperature. The distance, EOC, and MPG are measured by my MPGuino, and the temperature I look up at Welcome to Weather Underground : Weather Underground.

It's obvious that temperature makes a pretty big difference in MPG, but I thought I would make some graphs of it anyway, it's pretty interesting how the MPG follows the temperature:

First, the MPG vs. Temp and EOC vs. Temp on a log scale axis for my trips to school:
http://ecomodder.com/forum/attachmen...1&d=1237388832

You can see a pretty good correlation between MPG and temperature here, but it's better if I take the log of the temperature, and then graph it:
http://ecomodder.com/forum/attachmen...1&d=1237388832

And the same thing for trips from school:
http://ecomodder.com/forum/attachmen...1&d=1237388832

And with the log of the temperature...
http://ecomodder.com/forum/attachmen...1&d=1237388832

The EOC figures can't really be directly related to the temperature, because my EOC practices are more determined by other people on the road. If there are people behind me on one of the roads I travel, I usually won't EOC as much, because there are no passing zones, and I feel bad making everyone else go 35 in a 55. I think it just worked out that on nicer days, there were fewer people behind me ;)

It's only a small time frame, so the correlation isn't perfect, but I thought it was still interesting to see.

The question is how much does engine warm up factor in?

Well, for all the trips going to school, I start the car, and spend about 1 minute clearing off the windows, as there's been frost on them every day. For the trips from school, I start the car, and leave, as the windows have been clear every day.

Next year, I'll try to get a block heater, and compare the results. I'm sure that they will be much higher, as engine warmup is probably the main factor for poor mileage in the cold. There's not much point in getting a block heater now though, as winter here is pretty much over.

Once the engine is warmed up though, the coasting mileage is still worse at low temperatures than it is at high temperatures. Obviously, there are other factors, like wind, but on colder days, my mileage at 55MPH is definitely worse than it is on warmer days.

Here is some of my collected data. Daily commutes for 2008 and 2009 so far. No AC in the summer.

http://mcguckin.us/files/car/mpg_temp_2009_03.png

Cool, I was just about to ask for a plot of temperature versus mpg, rather than the two plotted against the date. The data are extremely noisy as I expected. McTimson, would you post a similar plot of your data?

Also, Pale, is that a polynomial fit? I would expect roughly linear behaviour, with discontinuities at the temps where you change your warm-up sequence, your ECU changes to a winter regime, and where you turn on your A/C.

Very cool data, guys!

I'll see if I can get a scatter plot in a little bit. I don't think it will look as nice though, I don't have nearly as many data points.

Alright, here's the MPG vs. temperature graph for my data:
http://ecomodder.com/forum/attachmen...1&d=1237424941

It's not as linear as I would expect, but I think that with more data points, it would become smoother. Again, this is only from like 2 weeks worth of recording, so it's not perfect, but it shows the obvious relationship between temperature and MPG.

I also attached the excel file if anyone wants to play around with the graphs.

There is a definite leveling off of any gains past about 80 degrees. I don't use AC, so that's not the reason. I'm guessing 80-90 is the optimum intake temperature for my car.

I have very little data at the cold end, as you can see, so that's somewhat suspect. Most of this winter's points were on the high end there, while last winter had the low points. I think early results skewed the data. That, or just unfamiliarity with winter driving for mileage.

Pale,
Wow, lots of data points - what a nice change. However, don't you thinkk the distribution of those data points indicate that huge variables other than temperature are effecting your mpg? I'm not sure it is possible to pull the temperature component out like you have done. Your r* value of .16 tends to bear this out by indicating a very low correlation between the x and y axes. (For those who may not know, an r* (r-squared) value of 1 = perfect correlation, while 0 = no correlation).

For example, at 71F your data shows a high of 80 mpg and a low of 35 mpg (and everything in between). That's a difference of 45 mpg! Clearly some unknown variables are having a very large effect on your mpg even when temperature is constant. Therefore, the smoothed regression line calculated from data with an r* of 0.16 has to be taken with a huge grain of salt. A problem with Excel charts is that they often imply stronger correlation than is actually there because Excel will draw a nice regression line even when one may not really be appropriate.

Another problem in accepting that this line accurately represents temperature's effect on mpg is that seasonal variables independent of temperature are not controlled for. For example, is some of your winter commute in the dark when the days are shorter? Traffic generally moves more slowly in the dark than in daylight and would reduce your mpgs. Or does the lower winter Sun shine in driver's eyes, slowing traffic? Neither of these confounding variables would have anything to do with ambient temperature per se, yet would produce an apparent correlation when T and mpg are plotted against each other on a graph. This is the bugaboo that catches so many researchers - their study actually measures an unrelated correlation rather than the cause and effect they were thinking about.

Of course there are other variables involved. This is real-world data, not test conditions. The higher probability of frost at 28F, for example, is included here. There's a huge human factor (me) that skews things. Wind is a big one, and rain too. Rigorous statistical analysis this isn't. I have data collected on these other factors, as well. This chart is simply a data-dump, not counting for any of the other factors.

Those points at 71F, for example. The 80 mpg one was with a warmed-up engine from earlier trips, and with a 15 mph tailwind. The 35 mpg was a 1.0 mile trip - should probably not be included in the data.

I debated whether to include the trend line. Clearly that was a mistake. This discussion is detracting from the value of the data itself.

Pale, I'm not saying putting a trend line in was a mistake, just pointing out how easy it is to give the line too much weight as a descriptor of a temperature/mpg relationship. I didn't mean my post to be a gotcha kind of post - just a discussion of your data that I thought might be interesting. Sorry if it came across as anything other than musings about interpreting data.

Interesting plots, guys.

Andrew: re-label your chart with "month" instead of "temp". Then it's unassailable. :)

One only has to isolate a single variable pulled out experimentally to see the significant effect of temp on efficiency. This is a coastdown test of rolling resistance at various temps:

http://www.recumbents.com/MARS/media...0graph%201.jpg

Source: MARS projects - Crr vs Temperature

The data is for a bicycle/trike, but applies generally to auto tires as well. Anyone who pays close attention to efficiency knows they have to adjust their glide points on familiar routes as ambient temperatures swing, even with tire pressure and road conditions held constant.

;) I can play that game too.

A certain stretch on my commute. Hit the 50 mph sign at exactly 50 mph, and coast until the spot where the pavement changes. Record the final speed. This was last spring, but I stopped bothering to record this data when it showed to be so very linear.

http://mcguckin.us/files/car/coastdown_temp.png

It would be even closer, but a couple of those low points showed me that my tires were a few psi low. Once I fixed that, the data got back in line.

Nice! Mind if I use that graph at some point in a MetroMPG post?

I'm such a nerd for the data that seeing all the pretty graphs almost makes me wish I was a commuter, so I could play too. :P

Go ahead and use it. I'm happy to provide data.

I've been figuring that the low temperature must affect coasting as well, but never took measurements of it. Thanks for the graph, PaleMelanesian, that's a pretty big difference in speeds. Just another reason for summer to get here soon :)

It does not get more straight forward than this on temp vs mpg

http://i6.photobucket.com/albums/y23...003EchoMPG.jpg

This comes from 3 years, 185 tanks, for a 2003 Prius:
http://hiwaay.net/~bzwilson/prius/pri_mpg_temp_110.jpg
Due to the humidity, when temperatures reach 85F, I may use the A/C. It is a driver safety issue.

I've recorded some data from trip from Huntsville, AL, to Columbia, SC, and back. A high pressure system came to Dixie and the temperatures ranged from 15F to 39F and back to 18F when I got home. Because I left at 4:00 AM and got back a midnight, there very little wind as a drove across the high pressure system. My highway mileage tracked the air density.

Bob Wilson

I would love somebody in the warmer climes posting their data, so that effect of further higher temps (> 80F) without the pesky AC factor coming in can be seen. Would FE still be climbing with temp? Would it wind down at some point? Through what mechanism?...

Look at my data above. That is WITHOUT AC usage.