Weather and fuel consumption

[Brian Ellis]

I've added the following update to my hybrid website:

Update 24 February 2008

I have found something that may or may not be rather astonishing. I also run a weather station and I find that the fuel consumption seems to depend also on atmospheric pressure. When it is about 1030 hPa, the consumption seems to be between 0.1 and 0.2 l/100 km lower than at 1000 hPa. In a way, this is logical in that more air (by weight) is swept into the cylinders at higher pressures, but would a 3% increase have such an effect? It is hardly supercharging. I can't be scientifically categorical as to cause and effect, but it seems from anecdotal observations that it may be. Of course, there are confounding factors, such as humidity. However, high pressure is associated more with low humidity. If humidity were to have an effect, I would have thought that high humidity would lower consumption due to conversion of hydrocarbons to methane and hydrogen, increasing the calorific value of the fuel, so I think that this can be discounted. I'll try and keep an eye on the correlation between weather and consumption.

[Elie]

I have noticed that my mileage decreases in the winter.

Also try to avoid ethanol where possible, ethanol increases fuel consumption therefor not a good alternative to gas.

Esso and Petro Canada add ethanol, but Shell doesn't. on my next fill up I will go to Shell.

[Brian Ellis]

It does not surprise me re ethanol. It has a lower calorific value per unit volume than petrol. As your injection system is volumetric, you receive less HP in your cylinders under a given set of conditions, therefore you press the loud pedal a little harder to compensate.

As for mileage in winter, there is possibly an optimum air temp. If you are well into the minuses, it will take a lot longer for your engine to heat up and the injection will be rich until it does. My consumption goes up in summer, because of aircon (rather necessary in 40++°C conditions!!!). As the aircon is electric, working off the high voltage battery (otherwise it would stop when the car stops at lights, as the engine cuts), it means I have less electric oomph available for driving the car and I have to use the ICE more..

[VJ]

In single propeller aircraft (which use a flat 4 cilinder boxer engine with large cilinders), the ceiling of the plane is actually caused in large part by the engine, which doesn't have enough power to make the plane climb further. This lack of power is caused by the lower pressure at high altitudes, so it does play. But as you say, the percentage is quite high...

Interesting find... but I think you ought to keep track of all factors (temperature, humidity, pressure, rainfall). I wonder if the consumption meter could also not be slightly off due to changing circumstances....?


Jörg

[MultimediaMan]

The power output difference/ fuel compsumption differences, outside of Temperature, Humidity and Relative Air Density, depends quite a bit on how "smart" your fuel injection system is.

I would imagine that on a hybrid, they keep this system as simple as possible:

Most Fuel Injection systems are closed loop meaning that nearly all of their parameters are pre-defined. To accomplish this Fuel injection systems today use a computer which follows a pre-determined spark timing/ injection "Map" which amounts to nothing more than Tables which map fuel delivery based on Engine RPM, Throttle Position, (Optionally) Airflow, the Oxygen Sensor, and (Optionally) the gear the Automatic Transmission is currently in.

If the airflow sensor is primitive (or nonexistant), it is very possible to run outside of the map's optimal parameters during Summer and Winter, which could lead to you running in "failsafe" modes which favor engine longevity rather than peak economy.

Building a really smart Fuel Injection system is not cheap: a completely open fuel injection system, which by definition, has a lot more sources of information from which to work (Including RAD) are quite a bit more fuel efficient (and generally provide better power as well). The downside of these systems is cost, and added maintenance. Sometimes getting Emissions Systems Longevity certification can be difficult for an Open Loop system. I can say from experience that troubleshooting drivability problems on an Open EFI system can be challenging indeed.

High-end cars have these advanced EFI systems: Hybrids, as a cost savings measure, probably run with something a lot simpler.

[Elie]

I agree with everyone here, I also think the way you drive has a big impact onfuel consumption.

Even in the summer, there are times that I can reach 700kms per 70 litre tank and sometimes when I drive a bit more aggressively I only reach 600 or 630 at best, and my BMW 528i 1998 has been rated pretty well on fuel consumption. So asside from the weather conditions, it's how aggressive you are on the gas pedal.

Regards,
Elie

[Brian Ellis]

Yes and no. With the hybrid, if I drive very aggressively, flooring the pedals, the consumption goes through the roof. However, it also increases significantly if I drive very softly, because the electric motor has little chance to kick in. This may also be a function of the continuously variable transmission, which is unlike anything with gears. If the battery is more than half-charged, I get max economy if I try to maintain a constant 3000-3500 rpm when accelerating or going up a steep hill, as this is the regime where the motor gives the most torque. If the battery is close to discharged, I drop the rpm. On the other hand, going downhill, it is often necessary to use the brake pedal as the engine has all its valves wide open and a slight pressure on the brake pedal charges the battery better than freewheeling; of course, the mechanical brakes don't come in like this. It is quite an art to achieve maximum economy.

[The PIT]

The power output difference/ fuel compsumption differences, outside of Temperature, Humidity and Relative Air Density, depends quite a bit on how "smart" your fuel injection system is.


And also how good these sensors are. I suspect they aren't very good in the first place as it would be a good place to start cost cutting. I'd be even surprised that they're actively measured in most cases.

I think what Multimediaman sums it up overall.

Best fuel consumption I ever recorded was 44 mpg in an XR2 mk1 with a rally cam in it on a nice summers day. Driving at 90 mph on the M4. Despite having a Renualt Clio now I haven't beaten that.

[Elie]

There were also studies conducted regarding the relationship between fuel consumption and speed. you say it's obvious the faster you go the more fuel you consume, that is correct. The study concluded, at 100kmh or 60mph your gas based vehicle consumes the least amount of fuel. This is one of the reasons why hiways speed limits are fixed at 100kmh or 60-65mph.

Thoughts?

[MultimediaMan]

There is a relationship to Speed, Drag and Fuel consumption: just ask any endurance racing team.

A lot of those studies were done decades ago: I'd like to see them done again with "slicker" cars aerodynamically-speaking. You can tailor the "sweet spot" a bit higher or lower with a conventional ICE.

With a Serial Hybrid, that "sweet spot" can be nearly anywhere, anytime.

[Brian Ellis]

I don't think there is a "sweet spot" in a hybrid because the engine revs do not necessarily bear much relationship with the speed, because of the CVT. If I'm doing a constant 80 km/h on a straight, flat, road, the engine revs can be anything from ~1500 rpm to 3000 rpm, depending on a) the electrical load and b) the state of battery charge. The instantaneous fuel consumption could be anything from about 2.5 to 7 l/100 km (the latter with a low battery charge, aircon, headlights, radio, etc. on). It gives priority to charging the battery, when it is really low, which is why the engine cuts at lights only when the charge is better than half (which is most of the time).

Yes, drag is an important factor and is not linear with speed, because at low speeds there is almost no turbulence (streamline flow), whereas the turbulence increases proportionally to the square of the velocity. The critical velocity of a clean glossy paint surface through air is >200 km/h, so it is only the turbulence due to the shape of the body that counts. The lower the speed, the lower the drag, obviously.

[Dr Mordrid]

At least here;

Winter weather = snow/ice = stop & go/dodge the idiots = lousy mileage.

Summer weather = clear roads & normal cruising speeds = better mileage.