Re: About Non-Linear Speedometers Quote:
Originally Posted by zenren Difference of 1-2km/h on a GPS based speedo has multiple reasons: - GPS always computes speed based on time and straight line distance travelled between two consecutive points that it measures. If the roads are slightly curvy, you actually travelled a higher distance in that time than your displacement indicates and hence the car speedo indicates slightly higher speed.
- In most GPS devices, readings would have a slight time lag and hence what you see now would be your speed 1-2 seconds ago. You can check the speeds on the GPS speedo when you slow down - it'll take a second or so before starting to decrease.
|
Sorry, this is not entirely accurate.
Nearly all GPS units measure speed based on doppler phase shift effect. Even if it would measure time/distance (and some do) it would do that several times per second, so a curve road has no bearing on speed.
What is true is the fact that in order to get an accurate GPS speed reading you need to be traveling at a constant speed. So no acceleration or deceleration. At constant speed they are very accurate, much more accurate than your car's speedometer.
Whereas the absolute error in most car speedometer increases as the absolute speed increases the absolute error on the GPS remains (by and large) the same.
I have been thinking about these non linear speedometers. I have had a few cars that had those (VWs). I'm not yet convinced that they are more accurate. Because of the scale it appears as if you read the speed with great accuracy, but that doesn't mean that the actual reading itself is more accurate in that section of the scale than the "higher up" sections. Just because you have a more granular scale doesn't make the measurement itself more accurate, only the reading of the value.
It probably depends on how these non linear speedometers work. Is it driven by a motor, or is it in essence just a voltmeter? I.e. a coil in an electric field. The needle moves because of a different voltage applied.
Believe it or not, but I used to teach instrumentation, calibration and that sort of stuff nearly 30 years ago. It might come back to me. It probably doesn't really matter for the driver. For all intents and purposed you think you get a more of an accurate reading. As long as you don't break the speed limit nobody is the wiser as to what the real speed and or accuracy is. One of the reasons that typically car speedometer reads to high anyway all the time.
Truth be told, I never noticed that the speedometer was nonlinear until somebody pointed it out to me. So it appears, well at least on me, to work quite naturally.
Jeroen
Last edited by Jeroen : 20th March 2014 at 23:04.
|