It's quite easy to check whether the figures shown on your trip computer are of much value.
Immediately before you fill up add the number of miles remaining (as indicated by the trip computer) to the number of miles you've travelled since the last time you filled up. This will give you the range of the tank based on your recent journeys. Immediately after you've filled up see how many miles the trip computer says you have left. This will also give you the range of the tank based on your recent journeys. The range shown on my trip computer after I've filled up is consistently about 60 miles less than the range calculated before I've filled up.
I accept that this doesn't take into account exactly how full the tank is after each fill, but the consistent difference of mine (equivalent to about 7 litres) satisfies me that I have come to the right conclusion ~ which is that my trip computer is just a toy. And it was just the same on my previous (similar model) car.
The other thing I've noticed is that although the average fuel consumption display is capable of reading to 0.1 mpg, when the reading changes it invariably changes by 0.4 or 0.5 mpg. For example, it will show 38.2 mpg, 38.7 mpg or 39.2 mpg but never anything between.
|
|
Could the 60 miles difference not be the computers default margin for when the low fuel light comes on? I believe most systems are set up allowing for those who chance their arm everytime and don't fill up until the gauge shows empty and the low fuel light has been on for some time!
|
Could the 60 miles difference not be the computers default margin for when the low fuel light comes on?
I can't see why. I rarely let my tank drop below half full.
|
I can't see why. I rarely let my tank drop below half full.
Lol - I'd be filling my tank every 3 days if I took this approach! ;-)
|
|
|
>> Could the 60 miles difference not be the computers default margin for when the >> low fuel light comes on?
When my fuel light comes on the range is near to 0. But it's got at least 7-8 litres remaining. So this theory holds for some cars.
|
on my car when the low fuel light comes on when i seem to have 1litre remaining on a 42 litre tank !
I suspect that there is actually more than that as when the car was almost completely empty I filled it up and the 42 litre tank took 45 litres.
Has anyone else noticed that more litres are needed to fill up a tank than what the tank capacity is ?
|
"Has anyone else noticed that more litres are needed to fill up a tank than what the tank capacity is ? "
Bound to be the case, as there is no allowance for the capacity of the pipe to the tank - so if you brim the tank there's another 5l quite easily over and above the tank itself.
|
|
|
|
|
|
For example it will show 38.2 mpg 38.7 mpg or 39.2 mpg but never anything between.
That sounds like a BMW. Most other makes I've driven go up and down in .1 increments.
A vauxhall hire car I had switched to litres/hour when stationary which I thought was rather neat.
I've always regarded fuel consumption displays as optimistic toys. They're not under regulation like speedos, and what manufacturer wants someone complaining about "only 36 mpg" when they could just as easily set the display to let the person think they're getting 39.2 - the .2 makes it sound so much more accurate than it really is!
Edit: my latest car doesn't have a trip computer - even though they're toys I miss having it - it was useful for working out the best compromise between speed and economy, especially when towing.
Edited by Marlot on 27/04/2009 at 09:43
|
>>and what manufacturer wants someone complaining about "only 36 mpg" when they could just aseasily set the display to let the person think they're getting 39.2 - the .2 makes it sound so much more accurate than it really is!
Couldn't agree more, and isn't it incredible how many people take this reading as gospel?
The only car I've owned with a reasonably accurate computer was our old Renault Scenic II. I don't think it was ever more than 1 mpg "out" based on brim to brim calculations. The VW setup in the Golf is about 10% optimistic.
|
Notoriously innacurate in my experience. I don't even bother looking at mine. I have the display set to show me outside temperature instead which is much more interesting to me.
What I do calculate is pence per mile. Simply divide the cost of each refill by the miles done since the last one. Provided at current prices it comes in at around 12p a mile or less whatever car I've been using I'm happy enough.
|
I have the display set to show me outside temperature instead which is much more interesting to me.
But not very accurate either in my experience
|
But not very accurate either in my experience
The first car I had with an outside temperature reading was my old 306. With the car parked in direct sunlight, 50°C was not an uncommon reading on the stupid thing until the car had been moving for some time.
|
The first car I had with an outside temperature reading was my old 306. With the car parked in direct sunlight, 50°C was not an uncommon reading on the stupid thing until the car had been moving for some time.
I thnk they can only give anything like an accurate record of the air temperature when the car has been moving for a fair while, ohterwise they will be affected by the temperature of the vehicle and the road surface
|
|
|
|
That sounds like a BMW.
I should be so lucky as to be able to afford something so expensive! It's a Mk I Focus.
|
Early indication is the computer on the CC3 gives about the correct average fuel consumption, or maybe under-reads by 1-2 mpg.
|
|
|
|
|
The range shown on my trip computer after I've filled up is consistently about 60 miles less than the range calculated before I've filled up.
These rolling averages update using the latest 'current consumption' (or instantaneous consumption) then multiply that by the amount of fuel remaining to calculate range . If you've slowed down, manoeuvred & stopped, then started & moved off again, that will affect (i.e. raise current consumption) - if you multiply that by the number of gallons/litres you've filled up with it's not surprising there's a difference! 50-70 miles (for a full tank) sounds about right to me - and is not, in itself, an indication of inaccuracy.
Edited by woodbines on 27/04/2009 at 10:23
|
>>which is that my trip computer is just a toy.
Just like everything else on a car's dashboard. Gimmicks for the easily led - see Oilrag's snake oil thread for more.
|
>>Just like everything else on a car's dashboard. Gimmicks for the easily led
Surprising comment NC - being an engineering type, I assume you understand the idea of something being made to fullfill a functional requirement - no more, no less.
If a guage,dial or other indicator of function, were engineered to your demanding standards I doubt many of us would able to afford a car at all. These aren't space shuttles or nuclear subs in which the merest deviation or inaccuracy is life or mission threatening - they're handy aids for car drivers to get an overview of function/speed/status etc.
I take it you look at speed indicators, rev counters, oil/charge indicators, fault checkers etc. occasionally? If the former then welcome to the world of the easily led!
|
>>being an engineering type, I assume you understand the idea of something being made to fullfill a functional requirement - no more, no less.
Very much so. They are fit for purpose; just not the purpose that they purport to be for, and for which the car buying giullible think they are.
So, yes, a temperature gauge which normally sits in the middle of its sweep and suddenly begins to read a lot higher merits some action, but, the numbers behind the needle are largely meaningless, and the function could be much better, and more cheaply, acheived with ECU monitoring and a light/buzzer combination.
I'm happy to watch the speedo, because I know that it's innacuracy is mandated to be either accurate or safe, and so, I use it to protect my licence. The water and oil pressure gauge, I watch for changes, rather than reading anything absolute.
The fuel gauge gives me a rough idea when to pull in and fill up - I certainly don't trust it to tell me when the tank is empty with any accuracy, and I never let it get anywhere near there.
All bar the speedo could, for me, be happily replaced with lamps and/or buzzers - which I would argue provide a more distinct warning, and do not rely upon driver vigilance - so, from an engineering point of view, cheaper *and* better.
|
|
I think this is all a bit harsh on the trip computer. Jazz has only a basic model and consistently over-reads 4-5 mpg at the 55mpg level, which is not bad.
|
|
|
So yes a temperature gauge which normally sits in the middle of its sweep and suddenly begins to read a lot higher merits some action but the numbers behind the needle are largely meaningless and the function could be much better and more cheaply acheived with ECU monitoring and a light/buzzer combination.
My car has a "proper" coolant temperature gauge (ostensibly) calibrated in degrees C. This was actually quite useful to diagnose a failing thermostat - instead of the needle being pinned to "90 degC" after warming up, it was sitting about 5-10 degress lower.
Of course, an oil temperature gauge would be even more useful for indicating when the engine is properly warmed up, but they aren't common these days...
|
|
|
|
|
These rolling averages update using the latest 'current consumption' (or instantaneous consumption) then multiply that by the amount of fuel remaining to calculate range . If you've slowed down manoeuvred & stopped then started & moved off again that will affect (i.e. raise current consumption) - if you multiply that by the number of gallons/litres you've filled up with it's not surprising there's a difference! 50-70 miles (for a full tank) sounds about right to me - and is not in itself an indication of inaccuracy.
I take both readings when I'm stationary at the pump. I take the first reading a fraction of a second before I switch off the ignition and I take the second reading after I've switched on the ignition again but before I've started the engine. Between one reading and the next the car hasn't moved.
|
|
Im with NC, unless you know exactly what it is sensing, and how the consumption gadget does its calculation it is only a rough estimation.
Edited by Old Navy on 27/04/2009 at 13:18
|
|
My consumption check is easy, 100 miles per quarter tank for the top three quarters, then I fill up at 300 miles / quarter full. Accurate enough for me, and to spot a problem. I dont bother with the electronic trickery.
|
The trip computer on my 2007 Astra is 1.0 to 0.5 mpg pessimistic over a whole tank compared to manual brim to brim calculations. I have been pleasantly surprised at its accuracy!
The response time for updating the display when instantaneous consumption is being readout is very fast compared to the system on my old Omega. I put this down to a more advanced electrical system i.e CANBus.
No complaints at all.
BTW 1.9 150 CDTi averages 37 mpg urban (hilly conditions) and 52/53 motorway/country driving.
|
The reason trip computers appear to be accurate to 0.1mpg but only display a selection of MPGs (e.g. 37.7, 39.2, 40.9, 51.4) is that they do the calculations in metric (litres per 100 km) and then convert into MPG.
The trip computer in the last two cars I've had (both Citroen) has been reasonably accurate, in that the figures correlate to the number of litres put in at each fill-up and the number of miles driven. Of course the odometer could be wrong....
|
Presumably mpg calculations depend on conversions from number of wheel rotations to distance travelled (unless the system uses GPS). That must vary depending on tyre radius (which depends on tyre pressure, temperature, loading, ...), the friction(/grip) between tyre and road...
What sort of variation does this result in?
|
|
|
The reason trip computers appear to be accurate to 0.1mpg but only display a selection of MPGs (e.g. 37.7 39.2 40.9 51.4) is that they do the calculations in metric (litres per 100 km) and then convert into MPG.
A classic case of having more precision than accuracy, I think!
|
|
|
|
|
These rolling averages update using the latest 'current consumption' (or instantaneous consumption) then multiply that by the amount of fuel remaining to calculate range .
I think there lies the rub. It's the amount of fuel remaining that the car can't accurately measure.
|
I think there lies the rub. It's the amount of fuel remaining that the car can't accurately measure.
I liked the Renault computer. When the range indicator got to about 60 miles, it would just beep at you and show two dashes. You could almost sense the Gallic shrug and "pffft" as it did so. ;-)
"Stupeed rosbif eeediot. Hurry up and buy some fuelle and stop doing zis to me." ;-)
|
|
|
|
|
