Colder weather has zapped my mileage to 44mpg

Which is another way of saying "prove there is a problem".

Prove there isn't - you've got a lot of numbers to refute.

 

I found only two 2010 A3 TDI users with MPG information for me, but, of course, since the car was released recently, they don't know what the drop is until summer, but comparing cold against miserably cold, there is not much difference… the fact that the A3 TDI was released in the winter give you a hint that Audi was not really concerned about poor FE in the winter...

EPA for the 2010 A3 TDI is 40 MPG... the 2 folks who reported their mileage to me are getting around 44 MPG with 100% hwy driving even in colder temperatures...

 

I mentioned 65 since I quickly noticed WI in your info and it says 65 mph for that state. It seems to the case for IL as well but not IN.

Just because you see a high number in the table doesn't mean that it's allowed all over the state, let alone even a majority. It says 70 mph for passenger vehicles on rural interstates in CA and WA. Guess what? The speed limit is NOT 70 mph anywhere I've seen in the SF Bay Area (other than I-5) nor anywhere in the Seattle area.

I wish I had some statistics that showed what % by mile each state had that either qualifies as a rural interstate or qualifies for its highest speed limit. For CA (from driving in the SF Bay Area and LA), I'd suspect it's <10% of highways, by distance.

 

I agree, but 90% of my driving is on hwy with speed limits of at least 65, but more often 70. The average speed of my trips is 66-67. I'm not a car racer despite of being from a country which has made great car racers and my official residence being Indianapolis. I drive safely and defensively…

 

I think it is time for a poll: Which car do you prefer, all else being equal:

Car 'A', that outperforms EPA and returns 45 mpg average; or
Car 'B' that underperforms EPA and returns 50 mpg average ?

 

Well, car B does not exist, since 50 MPG is EPA for it...

Maybe you meant?

Car 'A', that outperforms EPA and returns 45 mpg average; or
Car 'B' that underperforms EPA and returns 45 mpg average



or if there was really nothing else but fuel efficiency, we all should be praising about the Chevy Volt right?

I don't think so...

 

We just got back from a round trip to Nashvile, 234 miles (374 km), 54-57F to Nashville and 57-48F to HSV, average 54F (12C). We tested the highway cruise segments along I-65 between the Alabama border (look for Ardmore, AL) and Nashville, TN. The highway segment totals:

	Column 1	Column 2	Column 3	Column 4	Column 5	Column 6	Column 7
0	leg	Hwy distance(km)	Hwy speed(km/h)*	Ind. MPG	Act. MPG**	Gal(L)	time
1	HSV->Nash	67.6 mi (108.16)	73.2 mph (117.12 km/h)	50.5 MPG	48.1 MPG	1.4 gal (5.30L)	0:55:24
2	Nash->HSV	61.8 mi (98.88)	71.8 mph (114.88 km/h)	50.3 MPG	47.9 MPG	1.3 gal (4.92L)	0:51:40

* - GPS measured, cruise control except for local traffic adjustments
** - corrected using 5% factor

So if we take both legs and sum the distance, time, and fuel burn, we get:

	Column 1	Column 2	Column 3	Column 4	Column 5
0	Hwy distance(km)	time	Gal(L)	Avg. Speed(km/h)	MPG (L/100km)
1	129.4 mi (207.4)	1.784 hr	2.7 gal (10.46 L)	72.5 mph (116 km/h)	47.9 MPG (5.04 L/100 km)

* - sum of Hwy legs

So taking a quick look at my earlier benchmark:

So when my tired eyes draw a line up at 72 mph, I read out ~48 MPG on the left axis ... close enough to 47.9 MPG. Curious that the EPA rates this vehicle as 48 MPG on Hwy. Still, some vehicle changes:

• 84F (29C) vs. 54F (12C) - denser air (closer to Std. day)
• Shell 89, E10 vs. Shell 87, E10 - higher energy fuel
• <600 mi vs. >8,000 mi - vehicle miles
• 35 psi vs. 45 psi - tire pressure
• OEM transaxle oil vs. 1st transaxle oil change
• OEM alignment vs. front toe corrected

Now we haven't fitted a lower air intake block. Also, three wheel geometries are OEM, rear wheel toe and camber, and right front wheel camber. Finally, we're still running OEM tires.

Bob Wilson

Source data:

​

 

Very nice Bob... considering the air temperature during your tests compared to mine, I'm pretty convinced the drop in my MPG is just normal for the Prius... i.e. nothing wrong with the car...

 

How accurate are these mpg numbers?
I bet the TDI also 10% over estimate same as VW's.

Ken@Japan

 

This is the temperature relationship I use:

	Column 1	Column 2	Column 3
0	Temp	MPG	% vs. 60F
1	30F	47.9	90.0%
2	40F	50.3	94.5%
3	50F	52.5	98.7%
4	60F	53.2	100.0%
5	70F	54.2	101.9%
6	80F	55.2	103.8%

* - 60F is standard day temperature using data from NHW11 Prius using lower air inlet block at 50F and below.

It looks like a 10% drop in mileage as temperature reaches freezing so we're looking at:

48.0 MPG @60F -> 43.2 MPG @30F
​

When you get your diesel, I hope you'll take some time to benchmark the "mph vs MPG" data ... especially the stock vehicle from the dealer. This provides a benchmark to evaluate different tuning techniques. My typical benchmark is to average opposite direction runs of 10 mile mile (16 km) distance after the vehicle reaches a cruise control set speed. Just make sure it is dry and document the date.

GOOD LUCK!
Bob Wilson

 

Great data, Bob! Definitely looks like a steeper drop from 40-30 than from 70-60.

 

Bob,
Thank you for collecting data.
However, I think the data has a lot of deviation.
I believe it mislead us to show such precious three significant digits under the wide deviation.

Ken@Japan

 

Hi Ken,

The data comes from the old GreenHybrid mileage database but I agree about the problem of 'precision:'

It is a hard problem because the raw data was so noisy. So I used a Gaussian filter(*) to smooth the data and hopefully extract a usable, smooth trendline. The extra precision bits were my tired eyes reading the Y-axis values.

There is a risk that the true values could be different and I would recommend anyone using my temperature data to take it with some skepticism if nothing else because it comes from a 1.5L NHW11 Prius. Yet in carefully instrumented and monitored, long distance, cold weather trips, the lower range seemed to follow the air density curve. For example,

Yahoo! Groups

Analysis of temperature effects on mileage is complex:

• air density - the only well behaved function
• tire resistance - very likely non-linear
• transaxle temperature - known non-linear lubrication effects
• ICE temperature - if not at 85C, likely non-linear

I needed something to 'guesstimate' (guess + estimate) a usable, temperature correction factor. This is based upon my vehicle data and could be subject to observer error as well as coarse metric (tank records ~2 weeks.) But I agree, the percentages read from the "y-axis" of a trendline chart are not precise ... it is all that I have.

Bob Wilson

* I have often found that a Gaussian filter applied to noisy data does a good job of reducing the noise in the data and allowing underlying signals to be extracted. It is type of weighted, over-sampling that in graphics is called "sharpening" and is based upon the probability of a data point being from a signal having some uncertainty of being at any one point.

** My thanks to ken1784 who provided the air density values at "Prius Technical Stuff."

 

Hi

I'm new to the site and from the Uk.

I have an old Japanese import Xreg 2000 toyota prius for just over 2 years now and it has been really good to me. It has about 72000mls on the clock. Regarding sub-zero temps, as you may know over here in the Uk in the last month we've had the coldest winter since the 1980s and alot of snow. When the temperature hits below 0 degrees and I try starting the engine, the master light (car with an exclamation point) comes on and a warning sign bleeps continuously, then the engine would automatically shut off. It stayed like that for about week while the weather was bad. Then as soon as the temperature went above 0 degrees, the car started normally again and has been running the same as before in the last 2 weeeks as if nothing is wrong.
I've been trying to find a local mechanic who can diagnose the problem but with it being an import most don't have the equipment and the main dealer is really snobby over here. Someone mentioned it could be a temperature sensor? If so where is it located? Would appreciate it if anyone has any idea what could be the problem?

Btw our other car is a new Honda IMA hybrid. I had to use that for work while mine was out

 

Well... here is the opinion of someone with experince in both cars...the 2010 A3 TDI and the Prius (gen 2 I think)...

TDIClub Forums - View Single Post - A3 vs Jetta

 

Hmm... so, if you're going to cherry pick... The link takes you to someone that has had the A3 for two days, and is getting better mileage than his Prius? Without posting a single point of MPG at all?

And there's a post above that where one of the A3 users is getting 50 MPG at 60 MPH. I do not know dude, but I can get 60-63 MPG at 60 MPH. This is better than a Prius how? I'd be truly curious to know what the user gets when speeds are bumped up to 65-75 like you travel? I bet it falls off at least 6 MPG, if not more.

Look, if you're going to argue that you'd rather change up for driving experience, I can see it. But that's not the perception you're showing to many here.

 

I think there's plenty of evidence to show that diesels don't take as much of a mileage hit in cold temps as the Prius, but also plenty of evidence that year-round average mpg doesn't keep up. So if the seasonal drop on the Prius exacerbates your seasonal affective disorder, get a diesel.

But if you'd like to talk about what might be causing the drop-off in mileage on the Prius in cold temps, in the hope of maybe finding one or two things that could be partially mitigated, please continue to post in this thread.

 

Yeah, the air density info is pretty surprising - I didn't realize it changed that much (12% increase from 70F down to 15F)... Does that factor in humidity differences? (dry air is more dense than humid air - and cold air is dry air)

That may be enough by itself (or combined with rolling resistance) to explain the steepening drop in lower temps at highway speeds, where aerodynamics are a bigger factor. This would be even more true at higher hwy speeds, where we already have data showing a steeper drop-off.

In other words, if in warm temps a 70mph speed is near the edge of the 220g/kWh "sweet spot", then in cold temps a 70mph speed would be well into (and near the edge of) the 230g/kWh "sweet spot". And a 75mph speed in really cold air (15F) may be outside of even the 230g/kWh sweet spot entirely...

Non-hybrids (which have lower coefficients of drag) would be less affected because they are evenly wasteful at a broader range of speeds.

 

If you were a pilot this wouldn't surprise you at all. It's truly amazing how sluggish an airplane can be on a hot, humid day.

Tom

 

Hum… You may have failed to get the idea. What I was trying to show is that TDIs does not appear to have the same level of MPG drop as the Prius HSD, I did not and will not intend to compare overall Prius vs. TDI mileage. Am I a bit clearer now perhaps?