Speedometer Accuracy

[teagueAMX]

Originally Posted by willysp

They use GPS:

The system that we use to record-braking data is the same we use for acceleration testing, our GPS-based VBOXIII by RaceLogic. This system samples our location 100 times a second in three dimensions using anything from six to 12 satellites for triangulation, and therefore can calculate distance, speed and acceleration in virtually every direction (though the X and Y coordinates are most accurate).

http://www.roadandtrack.com/article....rticle_id=8155

Doesn't this photo from the R&T article you cited look like the guy is in a e89 Z4?

GPS can be terrible accurate now days, and lot more so than our GPS systems, regardless of brand, would cause us to think. They always say accuracy within 30 feet, but that article proves they have much more potential.

[willysp]

Originally Posted by teagueAMX

Doesn't this photo from the R&T article you cited look like the guy is in a e89 Z4?

GPS can be terrible accurate now days, and lot more so than our GPS systems, regardless of brand, would cause us to think. They always say accuracy within 30 feet, but that article proves they have much more potential.

Yes - certainly looks like an E89! Didn't notice that.

With high end gear (and it's fair to assume that R&T has it), and augmentation like WAAS (Wide Area Augmentation System - developed by the FAA and DOD) accuracy in an open environment in good weather (e.g. a track on testing days) can be within a couple meters.

[teagueAMX]

Originally Posted by willysp

Yes - certainly looks like an E89! Didn't notice that.

With high end gear (and it's fair to assume that R&T has it), and augmentation like WAAS (Wide Area Augmentation System - developed by the FAA and DOD) accuracy in an open environment in good weather (e.g. a track on testing days) can be within a couple meters.

Actually I hear it can be closer than a couple of meters. A friend owns an excavation company and they use gps gear extensively within 10ths of a foot in 3d. Land surveyors use gps equipment extensively for established property markers, boundary lines, topographical data, etc. 6 feet +/- won't get it when it comes to setting a property line.

I find this amazing considering that only a few years ago the military allowed public use of gps but only on a much lower resolution. GPS tek has exploded in the last couple of years, so I'm figuring they must now have much higher resolution satellites or technology, and they've released their old ones for public use. They just don't let on about it.

[willysp]

Originally Posted by teagueAMX

Actually I hear it can be closer than a couple of meters. A friend owns an excavation company and they use gps gear extensively within 10ths of a foot in 3d. Land surveyors use gps equipment extensively for established property markers, boundary lines, topographical data, etc. 6 feet +/- won't get it when it comes to setting a property line.

I find this amazing considering that only a few years ago the military allowed public use of gps but only on a much lower resolution. GPS tek has exploded in the last couple of years, so I'm figuring they must now have much higher resolution satellites or technology, and they've released their old ones for public use. They just don't let on about it.

Yes, I've also heard that surveyors can be accurate to within centimeters. Don't know what type of gear it requires, nor whether it works on moving objects.

But, as a geek, I find tech like this to be fascinating.

An interesting anecdote is that the military turned on SA (selective availability, to reduce civilian accurary) in the Gulf War, but turned it off circa 2000 because so many civilians and soldiers were buying commercial gear for their friends and loved ones to use in the war. And SA has stayed turned off. And I'm sure you right - there is something the military has today that we don't. And I'm all for that - I like the idea of the military being able to used an unmanned drone to drop a surprise down the small chimney of an unfriendly and thereby reducing the risk of collateral damage. And that last comment of mine has what to do with Z4's?

[teagueAMX]

Originally Posted by willysp

And that last comment of mine has what to do with Z4's?

[eranidoobie]

anyone know yet if there is a way to disable the odometer reading temporarily to compensate for that 8-10% increase in read milage?

[Dr Stig 2]

Originally Posted by willysp

Yes, I've also heard that surveyors can be accurate to within centimeters. Don't know what type of gear it requires, nor whether it works on moving objects.

But, as a geek, I find tech like this to be fascinating.

The most accurate systems of GPS use "differential GPS"

GPS signals over hundreds of kilometres from space to earth can only position you accuratley to about 2 metres, and in bad conditions or with poor coverage, as bad as 10-30 metres.

If a GPS system has a nearby known fixed point that can be compared to a moving GPS location (eg a GPS base station and a roving GPS receiver) then the differential of the positions can be determined to a fraction of a centimetre, if they are using the same satellites to locate themselves at each location.

This is what the surveyors use.

In a moving car, the differential is done using the same GPS receiver. They dont compare two different locations at the same time, they compare two locations a tenth of a second apart, or two sequential signals from the same receiver.

While the absolute location is not any more accurate, the changes in its location using the same signals are as accurate as the two signals used by the surveyor's equipment.
Because the amount of movement is accurate, and the time unit is accurate between measurements (timed by an atomic clock on the satellites), the speed and direction are incredibly accurate as well. (just not your absolute location at any one time - which is alwasy out by up to 2-3 metres in the horizontal plane, or more if you have poor satellite coverage or lock)

You sometimes see a glitch in your GPS speed, this is when it changes from one satellite to another and your differential in measurements jumps from one accurate reading set to another accurate reading set - it can add in a few meters to your distance covered in a tenth of a second (which is how often the satellites send out a signal) so it can slow you down or speed you up just for a few readings on the display (many GPS unita average their speed calculations to try and minimise this change in signal accuracy).

Make sense?

The VBox units cant actually take readings faster than the satellite put them out, they interpolate the data to get more "accurate" estimates of timing & speed based on the data the read.
This means if they read 99mph at 10.1 seconds and 101mph at 10.2 seconds, they will interpolate the sample to tell you they measured 100mph at 10.15 seconds, even though they dont actually take a reading from the satellites at 10.15 seconds.

[SANguru]

this has been long discussed on every board. If you look at the data on the DME, analog speedo is no way tied to the odometer. Speed on all 0-60 test are also done on the VBOX and not looking at the speedo.

[Mr. ///M3 RD]

Originally Posted by teagueAMX

Two things come to mind:

1. What about 8-10% affect on warranty?

2. Computer adjustments for tire sizes?

This ^^^ agree 100%

[z4z4z4]

http://www.**************/search.php?searchid=94185

The BMW Coding Forum should have threads on adjusting speedometers. I know they have activated digital speed display for our cars, which may prove more accurate than analog. You'll have to register to pass through their pearly gated forum. Lot of good drama and mystery reading on there.

[tot]

On my car the digital speedo is spot on, within 1km/h when compared to GPS speed.

[Memphis1]

i don't know if anyone on here is young enough to remember the up,up, down, down, b,b, a, a codes we used for nintendo... but basically doing that there's a way to hack the drive into opening a master menu and you can choose to set the speed digitally on the little area where you normally see the miles to empty.

[35iDriver]

A 2 year old thread resurrected!!!

[timwhayden]

When coding your digital speedo on ************** you are given 2 choices: a digital readout that reflects the analog needle (as some have said 3% or higher off) or a digital readout that is actual speed!

[timwhayden]

The asterisks in my post were originally for the BMW coding forum but would not print the actual address.