Motoring costs

How much does it really cost to run a car?

Sure you can look up this on the internet (like you're doing now) and get some numbers.  It's a fair bet though that these numbers will simply reference a source and not explain where they came from.  Over the past 6 years I've been recording the costs of running my car - absolutely everything: fuel use, mileage, maintenance etc., etc.  You can now see how much it actually costs to run a car.  Obviously when using these figures you need to compare like with like, for example, depreciation will vary considerably with the age and type of car, as will fuel consumption.

Car

1994 Peugot 306 sedan SRDT, 1.9 turbo diesel.

Bought in March 2001 for £3000 with 101000 miles on the clock.  Sold in March 2007 for £300 with 182000 miles on it.

Data recording

Everything that was spent on the car is included in the costs.  To some extent the maintenance costs are reduced as I did most of the basic servicing myself (a basic service is an oil and filter change every 6000 miles).  Major services and repairs (no crash repairs) were done at local independent garages.  Consumables such as tyres are included in the maintenance costs.

Year by year depreciation costs are estimated as the only two definitive figures are what I actually paid for the car and what I sold it for.  The intervening figures are estimates based on selling prices at the time.

Fuel costs are what I paid at the time to fill up.  Mostly in the UK although there were several trips to Europe.  You can see the average diesel cost over time from the data.

No allowance has been made for inflation in these costs.  i.e. "£1 in 2001" was the £1 I paid out, it has not been increased by inflation rates to 2007 price levels.  For information, inflation in the UK has been at about 3% over this period.

Fuel Consumption

This was recorded simply by noting the mileage and fuel required to fill the tank up.  Its not the most accurate method of obtaining a single reading as its difficult to fill the tank to the exact same level each time (full).  However, you can see the fill-fill variations on the fuel consumption charts and if you average these out then the fuel consumption is very accurate.

Costs

  2001 2002 2003 2004 2005 2006 overall
miles 16799 13600 11689 15720 11089 11022 80848
depreciation (£) 800 700 500 300 300 100 2700
tax & insurance (£) 510 389 379 379 377 377 2411
maintenance (£) 515 749 1315 800 435 419 4233
fuel (£) 1294 1052 913 1297 918 1030 6504
Total cost (£) 3118 2890 3107 2776 2031 1927 15851
pence per mile 18.57 19.77 26.58 17.66 18.32 19.61 19.61
The average cost breakdown per year is

 

average over 6 years (per year)

miles 13474
depreciation (£) 450
tax & insurance (£) 402
maintenance (£) 706
fuel (£) 1084
Total cost (£) 2642
pence per mile 18.57

Fuel Consumption

Here is the fuel consumption (miles per gallon) of the car versus miles.
The big dip at 109k miles was caused by fitting a roof rack, loading the car with climbers and climbing gear and driving down the French motorways at 85 mph to the Alps and back. 
This is the cost of fuel (diesel) over the years.  It is the actual cost that I paid at the time (pence per litre).  The big dips are from buying fuel in France and Switzerland.

CO2 emissions

So how do you work out your CO2 emissions from fuel consumption?
Firstly you need some basic data on fuels*:
  diesel petrol
density (kg/litre) 0.84 0.72 - 0.76
% carbon by mass 86.5 85.0 - 88.5
Calorific value (MJ/kg) 42.9 41.9 - 44.0
The average fuel consumption of the car was 45.8 miles per gallon
 = 19.47 km/litre
Now, to get CO2 emission we need to convert km/litre into km/kg.  For this you need the fuel density and it differs for petrol and diesel.  My car was a diesel, so 19.47 km/litre = 16.35 km/kg or 0.061 kg/km.
So, each km used 61g of diesel.
But diesel is 86.5% carbon so each km releases 52.9g of carbon atoms.  For this calculation I assume that all of the carbon ends up as CO2 rather than the even more obnoxious pollutant, CO.  CO emissions are a tiny fraction of the CO2 emissions, typically <1 g/km.
The atomic weight of carbon is 12 and the atomic weight of oxygen is 16.  To burn that carbon to CO2 requires two oxygen atoms for every carbon atom.  So every CO2 molecule is only 27% carbon. 
CO2 emissions is therefore 194 g/km.
i.e. 45.8 mpg for a diesel car = 194 g/km of CO2 emissions.
So, for every 10000 miles driven (16100 km), my car emitted 3.1 tonnes of CO2.
Based on my average mileage, (13.5k miles per year), my car's CO2 emissions were 4.2 tonnes per year.
IF my Peugot 306 was petrol rather than diesel then you need to take into account the different densities of petrol and diesel.  Petrol is 88% the density of diesel so for the same miles per gallon, a petrol engine's emissions would be 12% less than those of a diesel.  Note however that a diesel engine will do more miles per gallon than an equivalent petrol engine.
Miles per gallon CO2 (g/km) diesel CO2 (g/km) petrol
20 444 391
25 355 313
30 296 261
35 254 223
40 222 195
45 197 174
50 177 156
55 163 143
60 148 131
These figures are quite a lot higher than official emissions data for vehicles.  This is because official data are based on laboratory tests done on a rolling road and whilst not too bad for comparing car A with car B bear little resemblance to real world fuel consumption.  Things that will increase real world fuel consumption over "official figures" are: weight carried in car (driver, passengers, fuel, luggage, accessories), electrical load (lights, wipers, heater, radio, etc, etc), tyre size and tread, tyre pressures, weather conditions (any wind will increase fuel consumption as the vector component of a cross wind increases drag).  Driving style also has an enormous effect on fuel consumption as does the type of use.

Based on your car's fuel consumption, you can work out your own carbon footprint from the table below.

Miles per gallon CO2 (tonnes/10 000 miles) diesel CO2 (tonnes/10 000 miles) petrol
20 7.15 6.30
25 5.72 5.04
30 4.77 4.20
35 4.09 3.59
40 3.57 3.14
45 3.17 2.80
50 2.85 2.51
55 2.62 2.30
60 2.38 2.11
 
* source - Kemp's Engineers Yearbook, 1995
Home page