MPG increase with temperature: 1.4% for every 10ºF

Diesl

Veteran Member
Joined
Dec 3, 2012
Location
Chicago
TDI
'78 Golf Diesel (long gone); 2012 Jetta Sportwagen TDI w/ DSG
My fuel efficiency went up from 35.5 mpg in February to 37.5 mpg in August, for the same commute. Here is the graph of five-tank average mpg vs average temperature. Until about March I extracted the weekly averages from the 'weatherspark' history plot, then I got lazy and just took the monthly average temperature. For the last few weeks in August I computed the average temperature for each tank from the daily observed averages.

The outliers (high points around 37 mpg/30-40 ºF) are from December 2012.
Using the five tank average gets rid of fillup variations. Maybe I should use daytime temperature averages.

The overall trend is a 1.4% increase in fuel efficiency per ten degrees F temperature increase, a 5.5% effect between winter and summer.
 

MikeMars

Veteran Member
Joined
Mar 6, 2008
Location
UK
TDI
Vento 1.9 TDi (retired), A4 1.9 TDi (rear end collision), VW Passat 1.9 TDi (retired), Audi A2 1.4 TDi
Yes, I see something similar (haven't got the exact figures with me). Basically, for me, MPG improves as the temperature increases (... until the climate control needs to be turned on).
 

Diesl

Veteran Member
Joined
Dec 3, 2012
Location
Chicago
TDI
'78 Golf Diesel (long gone); 2012 Jetta Sportwagen TDI w/ DSG
AC on/off doesn't seem to make much of a difference for me. I have been driving with the AC on (recycling on, 10-11 o'clock temperature setting) for July and August. Air recycling on vs off might make a big difference in how hard the AC has to run.
 

Diesl

Veteran Member
Joined
Dec 3, 2012
Location
Chicago
TDI
'78 Golf Diesel (long gone); 2012 Jetta Sportwagen TDI w/ DSG
MPG as function of air temperature, speed etc.

The pattern doesn't fit winter diesel. It's a steady increase, not a sudden step:


It could be though some sensor drifting. [I guess I'll find out when it gets cooler again, which should be soon:]
Update 2016-12-31: avg. speed added

Added 2019-6-28: mileage vs date

I think these two plots show the best correlation; the annual speed maxima and minima seem to roughly match the mileage variation.

Overall my MPG is going down ;-(


(old) Update:
I had never looked at how much the drag loss actually changes due to air density change with temperature. Guess what: going from 30°F to 70°F lowers air density by 8%! Density enters into the drag power linearly. I do about 70% of my driving on the highway, where air drag dominates. 0.7*8%~5.5%! That almost fits too well....
Update Sep 5: Added a graph that shows five tank average mpg versus five tank average temperature, and hopefully updates automatically from the spreadsheet. (The graph in the first post won't update, and uses the average daily temperature for the middle tank.) - 2019-6-28: now deleted, since it shows no correlation when all years are plotted on top of each other; for individual seasons, see plots below.

Update Nov. 8: it is getting cooler, and the diesel milage is dropping. Temperature vs time chart added.
Dec 16: use http://forecast.weather.gov/product...RD&product=CF6&format=CI&version=1&glossary=0 average for days actually driven.
Dec 21: Correcting some of the earlier temperature averages; e.g. Nov 30 - Dec 11, 2012 went from 48°F to 41°F.
Update 2014-2-28:
Disentangling the last plot by season:
First the falling temperatures of winter 2012/2013:

Maybe the initial rise (from right to left) is the running in of the engine??

Then the rising temperatures of spring and summer 2013:

The falling temperatures of fall and winter 2013/2014:

The drop in MPG follows a higher MPG curve than the preceding rise during spring and summer.
Spring/Summer 2014:

Fall/Winter 2014/2015:


Summer 2015:


Update 2016-12-31: I added average speed. Unfortunately there are two competing factors at work: the ever increasing highway speeds in the Chicago area, and a change in this car's mix of highway vs. city driving. It looks like city driving is winning, and dragging my average down.
 
Last edited:

Ultrasonic

Well-known member
Joined
Jun 10, 2012
Location
UK
TDI
Skoda Octavia 1.9TDi
Interesting data, thanks for sharing.

As you've concluded, I believe that for longer runs it is the change in average air density that is the dominant factor. (For short runs, the prolonged engine warm up time in cold weather will have a bigger impact.)

Any idea what happened in December and January though? Was your driving pattern different perhaps?

Edit: This is a post from the Ecomodder forum that attempts to calculate the relative effects of air resisitance and the use of winter diesel, which may be of interest:

http://ecomodder.com/forum/showthread.php/winter-diesel-fuel-19705.html#post273835
 
Last edited:

03_01_TDI

Banned
Joined
Dec 10, 2003
Location
Denmark
TDI
Na
Tire air pressure changes with air temps. This and quicker warm up time would account for most of the gains.

Now for a gasser engine a hot air intake can really aid in mpg.
 

Ultrasonic

Well-known member
Joined
Jun 10, 2012
Location
UK
TDI
Skoda Octavia 1.9TDi
Tire air pressure changes with air temps. This and quicker warm up time would account for most of the gains.
Tyre pressure changes would only have a small effect, and the only way the effect you describe would be a factor at all is if Diesl didn't regularly check and adjust their tyre pressures. I could be wrong, but my bet is that someone who logs mpg data like this probably checks tyres pressures now and then ;).
 

03_01_TDI

Banned
Joined
Dec 10, 2003
Location
Denmark
TDI
Na
The rule of thumb is for every 10° Fahrenheit change in air temperature, your tire's inflation pressure will change by about 1 psi (up with higher temperatures and down with lower).

Plus the summer sun warms up the tire and road surface.

In general for every 1 psi in tire pressure you can raise your gas mileage by .4 percent. Depending on tire type and sidewall flex.
 
Last edited:

powerfool

Veteran Member
Joined
May 25, 2012
Location
Indianapolis Metro Area
TDI
None anymore
The rule of thumb is for every 10° Fahrenheit change in air temperature, your tire's inflation pressure will change by about 1 psi (up with higher temperatures and down with lower).

Plus the summer sun warms up the tire and road surface.

In general for every 1 psi in tire pressure you can raise your gas mileage by .4 percent. Depending on tire type and sidewall flex.
That's all great and is good to know, but it can be a non-factor if he checks his air pressure and adjusts his tire inflation accordingly, as was previously mentioned. (EDIT: and usage or not of LRR tires).
 
Last edited:

piotrsko

Veteran Member
Joined
Aug 11, 2013
Location
Reno Nv
TDI
2013 Golf, 2000 F-250 (7.3)
Air density (density altitude) is a big factor to drag resistance. Flying calculations for take off change like this, too.
 

powerfool

Veteran Member
Joined
May 25, 2012
Location
Indianapolis Metro Area
TDI
None anymore
Air density (density altitude) is a big factor to drag resistance. Flying calculations for take off change like this, too.
It certainly does. Think of air resistance like running the gauntlet in football. The spacing of the gauntlet is the air density. Your speed is just the same.

So, let's say running at 12 mph, you come into contact with 3 opponents per second. You have to overcome all of that force to continue your forward motion. If you double your speed (24 mph), you will come into contact with 6 opponents per seconds, which requires double the force to overcome and continue your forward motion. This is where lowering your speed comes into play... every bit of air you displace requires force and you "accumulate" for more friction. This is just like impedance in electricity... if you increase the speed of the electricity, it increases the resistance... and it is observed by the increase in heat generated... or just like water pressure... if you double the length of a hose, you reduce the exiting water pressure.

So, air density works exactly the same way. Let's say you double the density of the gauntlet. Now, at 12 mph, you come into contact with 6 opponents per second, which is the same amount of force required to overcome half the density at double the velocity. So, with more dense air, you can overcome the increased drag by slowing down (or by drafting). If the roads are icy, it would also behoove you to slow down. Other than that, you could improve your aerodynamics.
 

Ultrasonic

Well-known member
Joined
Jun 10, 2012
Location
UK
TDI
Skoda Octavia 1.9TDi
Plus the summer sun warms up the tire and road surface.
You have a point there. In that I think the peak pressures obtained in the tyres whilst actually driving will likely be highest in summer when it is sunny. That's certainly my obvservation based on a few measurements. That is, the difference between the 'cold' pressure that I set and the actual in-use pressure is likely to be largest in summer.

At one time I wondered if this accounted for the higher mpg I got on long runs in hot weather, and that if so I could match this all of the time by simply running with much higher tyre pressures :cool:. But when I looked at what the pressure differences actually were I realised that the rolling resistance difference was nothing like big enough. I then had a mini Eureka moment when it dawned on me that changes in air density with temperature could be the dominant factor. As posted above, when you look at the numbers it becomes apparent that this is the case.
 
Last edited:

Diesl

Veteran Member
Joined
Dec 3, 2012
Location
Chicago
TDI
'78 Golf Diesel (long gone); 2012 Jetta Sportwagen TDI w/ DSG
I do check my tire pressure. :)

I have a cheap pressure gauge I keep in the car. I corrected the tire pressure four or five times since last November.

If the tire pressure slowly crept up with ambient temperature the effect would go in the right direction, but as somebody already pointed out the rolling resistance is most likely only a small part of the overall drag. (How to check: There are posts detailing the 'butt dynamometer' method, and how to extract drag from a coast down. It should be possible to fit air drag and other losses separately. Maybe somebody with a VAG-COM has already done that?)
 

jettawreck

Top Post Dawg
Joined
Aug 2, 2004
Location
Northern Minnesota-55744
TDI
2001 Jetta and 2003 Jetta
If all these air density and temp equations really worked in real world driving like they supposedly do, I would never break 50 mpg in the -0*F winter temps.
There are other factors. Wind/speed is a huge factor. Lots less drag/friction in warm temps, much faster normal coolant temp, for sure, but colder fuel and intake air temps let the ECU run the injection timing more advanced. My best mileage is almost always in the cooler temps of spring and fall when it stays above freezing for the low.
If you would take your current base fuel mileage and factor in your temp equation, air density/drag, etc., it looks to me that would bring it down to less than 30 mpg in sub-freezing temps (and that's not even cold yet). That would be horrible mileage, even for a 2012 CR TDI. I don't think the change will be the same as the temps continue to drop off.
 
Last edited:

03_01_TDI

Banned
Joined
Dec 10, 2003
Location
Denmark
TDI
Na
If all these air density and temp equations really worked in real world driving like they supposedly do, I would never break 50 mpg in the -0*F winter temps.
There are other factors. Lots less drag/friction in warm temps for sure, but colder fuel and intake air temps let the ECU run the injection timing more advanced.
If you would take your current base fuel mileage and factor in your temp equation, air density/drag, etc., it looks to me that would bring it down to less than 30 mpg in sub-freezing temps (and that's not even cold yet). That would be horrible mileage, even for a 2012 CR TDI.

Exactly. Plus we would have seen speed decreases in Motorsport racing and top speed records due to air density and temp equations.
 

ZootsAlures

Veteran Member
Joined
Jul 6, 2013
Location
Windsor, CT
TDI
2012Passat SE TDI
3D Spacial Engineering at its finest!

We need more cow-bell!

Great thread-


"Just my opinion, I could be wrong" Dennis Miller
 

Diesl

Veteran Member
Joined
Dec 3, 2012
Location
Chicago
TDI
'78 Golf Diesel (long gone); 2012 Jetta Sportwagen TDI w/ DSG
[...]
If you would take your current base fuel mileage and factor in your temp equation, air density/drag, etc., it looks to me that would bring it down to less than 30 mpg in sub-freezing temps (and that's not even cold yet). [...]
Factual diesel mileage (measured mpg) does not change if you look at it or analyze it differently.

My minimum, observed in February, was around 35.5 mpg. Average temperature was about 30 ºF.
 

jettawreck

Top Post Dawg
Joined
Aug 2, 2004
Location
Northern Minnesota-55744
TDI
2001 Jetta and 2003 Jetta
Factual diesel mileage (measured mpg) does not change if you look at it or analyze it differently.

My minimum, observed in February, was around 35.5 mpg. Average temperature was about 30 ºF.
Guess I missed something there....
Anyways, there are so many things that affect mileage and engine efficiency that to expect a certain gain/loss just due to temp increase/decrease factors may be a good way to anticipate seasonal mileage.
My February temps averaged around 0F and still made 50+ mpg in 2009/2010/2011. (50/52.1/51)
Dec and Jan are the coldest months (-20*F not uncommon) and the average drops just a bit under 49 mpg then. Starting out cold and driving in sub-zero*F temps (not to mention snow/slop on the hiways) has a way of destroying the mpg average for the year. I'm realy surprised how little the mileage actually drops off. From a high of 56ish mpg to a low of 48 is better than I would expect considering the environmental factors. When its -20*F the car doesn't even coast downhill w/o rapid slowdown. Lots of frictional loss from heavy/thick grease in axles/CV joints, etc. Tires are very stiff in the cold. Amazing.

The old '01 MkIV ALH is my cold weather car. It seems to really like the cold air.
 
Last edited:

Diesl

Veteran Member
Joined
Dec 3, 2012
Location
Chicago
TDI
'78 Golf Diesel (long gone); 2012 Jetta Sportwagen TDI w/ DSG
Jettawreck, I'm trying to understand my car's fuel mileage variation. I'm not saying yours should be the same.

Interesting point about the rolling resistance changing with temperature; according to this post here (quoting the EPA...), it changes 3.3% for every 10ºF, or faster than air drag! But it's not clear (to me) whether that includes tire pressure change due to temperature change, or whether tire pressure is kept constant. Also, I would need to measure my tire temperature. Maybe it doesn't change all that much with air temperature?

Anyways, just taking the numbers at face value, if rolling resistance changes -3.3%/10ºF, and air drag -2%/10ºF, then how do I explain my observed -1.4%/10ºF?

PS: These guys have a fuel efficiency calculator!
 

Diesl

Veteran Member
Joined
Dec 3, 2012
Location
Chicago
TDI
'78 Golf Diesel (long gone); 2012 Jetta Sportwagen TDI w/ DSG
Good point. A few percent shift could be just some sensor drifting. I think when I first noticed the drift/change/whatever I made a list; tire wear was on it, driver getting older, very slow purge of winter diesel, I forgot what else.

I enjoy the car a lot, the good fuel efficiency is just the icing on the cake, so I'm not worried about a few percent, but this steady drift just seems peculiar. Come winter we'll know more. With some imagination the mpg vs time plot is already leveling out. :) Which reminds me: gotta update Fuelly and truedelta! The plot here should already have updated by itself, since I enter the data into the Google Drive sheet right at the pump.
 

jettawreck

Top Post Dawg
Joined
Aug 2, 2004
Location
Northern Minnesota-55744
TDI
2001 Jetta and 2003 Jetta
Jettawreck, I'm trying to understand my car's fuel mileage variation. I'm not saying yours should be the same.

Interesting point about the rolling resistance changing with temperature; according to this post here (quoting the EPA...), it changes 3.3% for every 10ºF, or faster than air drag! But it's not clear (to me) whether that includes tire pressure change due to temperature change, or whether tire pressure is kept constant. Also, I would need to measure my tire temperature. Maybe it doesn't change all that much with air temperature?

Anyways, just taking the numbers at face value, if rolling resistance changes -3.3%/10ºF, and air drag -2%/10ºF, then how do I explain my observed -1.4%/10ºF?

PS: These guys have a fuel efficiency calculator!

I don't really understand it either, really. It seems that with all the losses colder temps bring the mpg drop should be worse.
Tires in sub-zero temps are horribly stiff and the same with axle joints, tranny fluids, etc. Factor in cold engine performance during warm-up, and the other enviromental effects I'm shocked mileage doesn't drop much more. All the alleged "losses" don't add up to the result for me. Even the first few miles aren't as bad as they seem they should be.
A couple things that help mileage in the cold are variables in the equation, I think. From monitoring Scan Gauge readings the colder ambient temps obviously keep intake temps and fuel temps lower. ECU advances injection timing further when these (and other pieces of the puzzle that must make up the timing "map") are cooler.
I run injection timing pretty advanced both static timing and adaptation in the '03 which is mostly the "summer" car. When running in the cold, I was uncomfortable with how high it advanced, so the '01 (cold weather and towing unit) doesn't get the same set-up. I have it advanced some, just not as much. It seems to work.
All of our results will vary, as they say. Different cars/engines, environments, driving routine/distance, etc all make for a whole lot of variables. Shorter trips/in town driving makes for much lower numbers than longer drives especially in older temps. Most all my data is from my 40 mile work commutes.
I need to update my Fuelly data to get current on the '03. I quit tracking the '01 since using it mostly in the cold and for towing use. It obviously would make it quite inaccurate for it's normal average.
 

Ultrasonic

Well-known member
Joined
Jun 10, 2012
Location
UK
TDI
Skoda Octavia 1.9TDi
Exactly. Plus we would have seen speed decreases in Motorsport racing and top speed records due to air density and temp equations.
Land speed record attempts are deliberately carried out in warm conditions specifically because of the air density/aerodynamic drag problem.

Edit: see e.g. 40s into this clip (not the best example but one I happened to just come across): http://www.youtube.com/watch?v=LSFX9vrwJf8
 
Last edited:

Ultrasonic

Well-known member
Joined
Jun 10, 2012
Location
UK
TDI
Skoda Octavia 1.9TDi
I don't really understand it either, really. It seems that with all the losses colder temps bring the mpg drop should be worse.
I'm not saying this is necessarily the case for you, but one thing that may mask the difference is AC/CC use in the summer.

I definitely do see an appreciable seasonal variation in mpg (see my Fuelly link).
 
Last edited:

Ultrasonic

Well-known member
Joined
Jun 10, 2012
Location
UK
TDI
Skoda Octavia 1.9TDi
If you would take your current base fuel mileage and factor in your temp equation, air density/drag, etc., it looks to me that would bring it down to less than 30 mpg in sub-freezing temps (and that's not even cold yet).
Just reading through your post again I realise I'm not sure how you got to the 30 mpg figure? That would be a 21% drop in mpg which is rather more than has been suggested on this thread?

(My Fuelly data does show getting on for a 20% winter/summer difference but this is certainly not all due to aerodynamic drag. My daily commute is only 9 miles each way, so the longer engine warm up time in winter is a significant factor for these trips. It is on long highway drives where I attribute reduced fuel economy in winter to typically higher air density resulting in greater aerodynamic drag. The highest mpgs I've ever achieved have always been during long highway drives on hot summer days.)
 
Last edited:

jettawreck

Top Post Dawg
Joined
Aug 2, 2004
Location
Northern Minnesota-55744
TDI
2001 Jetta and 2003 Jetta
Just reading through your post again I realise I'm not sure how you got to the 30 mpg figure? That would be a 21% drop in mpg which is rather more than has been suggested on this thread?

(My Fuelly data does show getting on for a 20% winter/summer difference but this is certainly not all due to aerodynamic drag. My daily commute is only 9 miles each way, so the longer engine warm up time in winter is a significant factor for these trips. It is on long highway drives where I attribute reduced fuel economy in winter to typically higher air density resulting in greater aerodynamic drag. The highest mpgs I've ever achieved have always been during long highway drives on hot summer days.)
Apparently I did a pretty poor job of reading the first post.
I don't consider air density in the colder temps a real factor. Too much other stuff that has a bigger effect (in my opinion only). Fact of science it is, but in daily driving it's less of a factor than a breath of headwind from what I can log.
All I can go on is my years of fuel mileage data. I can pull a few 56 mpg/900 mile tanks spring thru fall, but average just under 53. Winter temps drop it down to 46-50 commuting depending on how many days are cold thru out the tank duration.
We don't get a lot of hot weather here and I don't hardly ever do any long drives. Not much AC or CC use. Mostly modetate speeds on two lane rural roads. Sort of boring, perhaps, but pretty traffic/stress free.
Sorry to jack the OPs thread, got a bit carried away-again.
 

Ultrasonic

Well-known member
Joined
Jun 10, 2012
Location
UK
TDI
Skoda Octavia 1.9TDi
All I can go on is my years of fuel mileage data. I can pull a few 56 mpg/900 mile tanks spring thru fall, but average just under 53. Winter temps drop it down to 46-50 commuting depending on how many days are cold thru out the tank duration.
That sort of variation is in line with what others are talking about I'd have thought?

We don't get a lot of hot weather here and I don't hardly ever do any long drives. Not much AC or CC use. Mostly modetate speeds on two lane rural roads.
The air resistance effect will be most noticeable for long highway drives, and more so at higher speeds, so your driving pattern may be a factor here.

Nobody is claiming the air density factor is the only one. The fact that there is more wind, rain and snow in winter is definitely significant too! But if I compare a long highway drive on windless sunny days in winter and summer, I get better mpg in summer (something like 54 USmpg vs 50 USmpg @ 60 mph), and this I put down mostly to the air density effect.
 
Last edited:

MikeMars

Veteran Member
Joined
Mar 6, 2008
Location
UK
TDI
Vento 1.9 TDi (retired), A4 1.9 TDi (rear end collision), VW Passat 1.9 TDi (retired), Audi A2 1.4 TDi
... All I can go on is my years of fuel mileage data. ...
Whereas with my tank data, I see a strong correlation with temperature (just like the OP), which is consistent with the physics of the situation. I guess (as usual) it is a case of 'your mileage may vary'.
 
Top