MPG vs Temperature - Some Data

[enerjazz]

After 15 months of ownership I have enough data to plot MPG vs Temperature and see some correlation. See the graph at the bottom of the page Paul's Prius Page http://enerjazz.com/prius/

Optimal temperature seems to be in the upper 60s (F). The fit curve shows the general relationship. Cold weather (below about 50 deg F avg) starts accelerating the mileage decrease. Warmer weather has a slightly gentler decrease.

More data will obviously help. I did throw out 3 outlier points from road trips with high headwinds (at various temperatures).

[DaveinOlyWA]

sorry have to dispute your conclusions. my data shows that the warmer it is, the better the mileage with the only variation being whether air is used or not. also you use average daily temperatures how do you calculate that??

i have to assume you are either extremely warm blooded or you are using weather channel data. that makes you conclusions extremely rough. looking at your data makes me think you are doing something at 70º to lower your mileage so i have to believe you dont turn on the air at 70º but more like 85º but the daily average after considering the overnight temps may lower the average daily temps.

i am tracking mileage by temps based on the car after being on the road 10 mins (i notice that it usually takes about that long to level off to a realistic level the actually time really depending on the speed of the car)

i have noticed a nearly 2-3 mpg difference from the early morning summer drives at 55-65º and the early evening commutes at 80-85º so my data directly opposes yours

[efusco]

I'm with Dave, I think your data is flawed, and thus your conclusions.

I show a very clear correlation of improved gas mileage with higher temps. Wayne Brown's THSII simulator data supports that as well. I do, at temps above 80, begin to see a slight decline in mileage if/when I turn on the AC, but when I drive without AC the improved fuel economy will continue well into the 90s. That said, a car sitting in the sun resulting in very high internal temps (and thus a hot HV battery) will show a reduced fuel economy.

[enerjazz]

The data is the data and it's not flawed. In fact it seems to support your statements. I use the weather data from the NWS climate records database (I also have a home weather station as a back up data source). It's the average of the high and low for the day. I then average those averages over the driving period for each tank.

Remember I only have 30 data points, so the line will take a better shape over time. I live in Texas guys - it's over 80 degrees in the morning in the summer and near 100 every day. The car is outdoors and despite the sunshield it's a dang oven after work. See my data at Interior Temp Data

I'm using A/C in the summer and I'm taking a 1-2 mpg hit which is what you might expect (and fits with Wayne's THSII data). My MPG by date graph on the same page will also indicate my best mileage is generally in the spring and fall when it's not too cold in the morning but not too hot in the afternoon.

When the average is in the upper 60's that's a typical 58 - 80 degree day. Warm enough in the morning but not too hot in the afternoon. That's where I'm seeing max MPG.

We don't generally get too cold here so you won't see the deeper cold weather drop off that others might see.

[Wayne]

I think as long as you specify the data was taken with A/C on (or at least that the cabin temperature was kept constant), that will help. I agree, you don't have a strict MPG-vs-temp chart, but you probably do have a 'real world' version of it. People don't run without A/C at the temps you describe, so this may reflect what they will see, regardless of the reason for seeing it. The hotter it gets the higher the A/C drag, and the lower the efficiency of the batteries.

Of course the real trick here is to try and keep all the other variables constant (wind, rain, route, speed, etc) while only the temperature changes. I know you probably didn't have much control over these other variables, which is why you are averaging so many data points.

Interesting chart.

[efusco]

Originally posted by enerjazz@Nov 19 2005, 01:36 PM
The data is the data and it's not flawed.  In fact it seems to support your statements.  I use the weather data from the NWS climate records database (I also have a home weather station as a back up data source).  It's the average of the high and low for the day.  I then average those averages over the driving period for each tank. 

Remember I only have 30 data points, so the line will take a better shape over time.  I live in Texas guys - it's over 80 degrees in the morning in the summer and near 100 every day.  The car is outdoors and despite the sunshield it's a dang oven after work.  See my data at Interior Temp Data

I'm using A/C in the summer and I'm taking a 1-2 mpg hit which is what you might expect (and fits with Wayne's THSII data).  My MPG by date graph on the same page will also indicate my best mileage is generally in the spring and fall when it's not too cold in the morning but not too hot in the afternoon.

When the average is in the upper 60's that's a typical 58 - 80 degree day.  Warm enough in the morning but not too hot in the afternoon.  That's where I'm seeing max MPG.

We don't generally get too cold here so you won't see the deeper cold weather drop off that others might see.

[snapback]158249[/snapback]

Ok, your data may not be flawed, but your interpretation of the results is flawed because of the multiple variables not taken into account. Average daily temp, by the way, does seem a flawed way to look at the data IMO. Your morning temp will be scewed low, your afternoon temp will be scewed high both having a greater negative impact on the results than the average temperture would have. IOW, if your morning temp was 40 and your afternoon temp was 80 your average temp would be 60. But if your car sat in the sun all day at 80 degrees and the HV battery was too hot that would have a negative impact as would the cool morning temp with the cold engine.

I just don't want anyone taking away your data and repeating in later threads "60 degrees is the best temp for mileage". Clearly that is a false conclusion to draw based upon your limited data.

[enerjazz]

Originally posted by Wayne@Nov 19 2005, 01:52 PM
The hotter it gets the higher the A/C drag, and the lower the efficiency of the batteries.

Yes, I can sometimes hear the batteries cry out in pain during the summer!

Evan, I don't see how you conclude my interpretation is flawed. My data is for the average temperature over the entire tank. If someone looks at it and thinks that it is for MPG at a specific temperature then their interpretation is flawed. Mine is spot on. I've added additional text to the graph to help with understanding.

[DaveinOlyWA]

ok Paul i DEFINITELY wouldnt test mileage without air conditioning either in your case...wow...

i actually have several hundred data points and was never tracking temperatures but the difference in mileage for round trips with neglible wind and elevation changes makes it obvious that temps do matter and until we get above normal ICE operating temps, i think the mileage will improve simply because the Prius gets to running temps faster which shortens to inefficient warmup cycle. i realized this mostly by accident because i worked 10-11 hour shifts all summer, i would leave in the morning for work and come home in early evening so i missed the hot part of the day.(that is why i have a lot of data with no air) sure the car was still warm but i even had the windows down and still improved my mileage

i just installed my canview from efier and its very much like it was when i first got my Prius. now i have a whole set of data sources and i most definitely intend to check out several different things.

bit off topic, but in just one hour of having it (first thing i HAD to do was show it off) and i already have seen that even the slightest brake pressure creates much more regen and brake lights dont use enough to register unless you push hard enough for the ABS pump to start running. (it uses about .1-.2 kw)

but i digress. i will be tracking (well attempting to anyways) how fast the ICE warms up at various driving conditions etc because of the warmup cycle points made above. cant wait

[EricGo]

Originally posted by efusco@Nov 19 2005, 02:22 PM
I'm with Dave, I think your data is flawed, and thus your conclusions.

I show a very clear correlation of improved gas mileage with higher temps.  Wayne Brown's THSII simulator data supports that as well.  I do, at temps above 80, begin to see a slight decline in mileage if/when I turn on the AC, but when I drive without AC the improved fuel economy will continue well into the 90s.  That said, a car sitting in the sun resulting in very high internal temps (and thus a hot HV battery) will show a reduced fuel economy.

[snapback]158205[/snapback]

It is hard to isolate outside temps from other variables that we know affect mileage, such as first warm-up of the day, changing routes, road conditions, wind, and different gasonline formulations.

For me the overwhelming difference I have noticed is the time it takes for the engine to warm up to it's optimal operating state. If I can drive for an hour at a time, my MPG stays in the 60's even in freezing weather. Shorter trips, and temps below 60 F leave me hoping for 55 MPG. When it is below 32F and the trip is short, 50 MPG looks darn good.

[tomdeimos]

Keep in mind there's more than one temperature to worry about.

Each has it's own effects! Assuming you start out in the morning and it was cold the night before, usually it is warmer when you leave.

So temperature of the battery may be closer to the night time low.
Engine may be closer to the current temperature, but with a boost from the thermos,
which will be affected by the night time low.

Then under way the engine warms up fastest, but the battery warms more slowly.
Max economy only happens when both are warm.

Also the catalytic converter has to be warm, and I've seen delays in the good economy operation after engine fully warm that may be due to that still warming up.

So far with the can vue I've seen rapid warm up of the engine due to my block heater!
So operating temperature of 170 is achieved at about 2 miles. This at 40 deg F ambient.
But the battery only goes up like 10 degrees on my whole commute of 12 miles, so never gets to normal temperatures. I see the effect of this in higher engine revs for all conditions as well as less battery mode.