Why your speedo isnot accurate.

[DrFfybes]

There has been some discussion of road v satnav v actual speeds, and the difference in variation.

When the rules were introduced, speedos (rather than Speedos (TM)) were mechanically driven, with an output from the driveshaft (or front wheel on a motorcycle) driving a cable spinning a magnet inside a metal cup retained by a spring, which moves because of Eddy Currants. The needle is on the cup. Higher speed, magnet spins faster, cup gets more force and turns further.

Obviously there are vast margins for error and tolerances - the spring retaining the cup is the main one as the strength affects the reading.

Then there is rolling diameter. Modern cars tend to have big wheels, but on an old mini with a 10 inch wheel the external diameter of the tyre is 48cm, modern cars with big wheels but smaller tyres tend towards 60cm. As tyres come with 6-8mm of tread new, and wear down to 1.6mm then three is a potential 12mm difference in diameter over the life of the tyre, so 2% change in circumference. So as the tyre wears down, the speedo starts to over-read.

On more modern cars the speedo is electronic - a sensor counts the turns and translates this to a digital or electronic display on the dashboard - the needle driven by a stepper motor. Thus on a modern car there is no real need for the speedo not to be accurate with a new tyre. [We will ignore the folly of a certain Italian manufacturer to reset this motor at "power on" such that if the rubber needle rest (situated at 12 mph) falls off then the needle drops too far and 'zero' is in the wrong place.]

As speedos are not permitted to under-read, they were always made with some tolerance to over-read, simplest is to mount the dial wrong by a fixed amount (say 3 mph) but also to make the drive mechanism slightly slack. I gather this is still current practice.

So, on a modern car you are likely to get a fixed over-read compared to the satnav, PLUS a small margin as your tyre wears down. This seems to be the case on our BMW which is pretty consistent at 3mph over the satnav. On an older car the over-read is often higher at higher speeds as the spring retaining the Eddy Current cup is slightly under-spec giving a percentage difference. The Toyota is about 3 mph over up to 30mph, but reads 5mph over at 70.

This, of course, assumes your satnav is correct. Out Tom Tom can differ from the Samsung Galaxy be a couple of mph unless at a steady speed n an open road. Both are 4 years or so old, so more modern ones might use more sensors to increase accuracy.

This means someone driving an indicated 65mph in the middle lane of a motorway could actually be holding up a caravan driving at a true 60 mph, giving them nowhere to legally pass.

Paul

Paul

[richlist]

Very interesting.

But if I want to drive at 65mph and that's what my speedometer is indicating that's what I'm gonna do regardless of holding up any caravans.

[Arborbridge]

Very interesting.

But if I want to drive at 65mph and that's what my speedometer is indicating that's what I'm gonna do regardless of holding up any caravans.

Hopefully, though, not in the middle lane

[Arborbridge]

I've compared my speedo to the many speed indicators one sees by the roadside in towns. They tell me that my speed in the car reads 1 MPH over the speed indicated by those cameras, which is wisely, on the "safe" side. Never seems to vary from district to district. That's near enough for me.


Arb.

[nmdhqbc]

This, of course, assumes your satnav is correct.

I always wonder if the Sat Nav's are clever enough (or have enough info) to take into account the gradient. If they only have GPS coordinates to work on it would under estimate speeds when you go up or down hill. It being the longest side of a right angled triangle and all.

[swill453]

This, of course, assumes your satnav is correct.

I always wonder if the Sat Nav's are clever enough (or have enough info) to take into account the gradient. If they only have GPS coordinates to work on it would under estimate speeds when you go up or down hill. It being the longest side of a right angled triangle and all.

GPS knows about height as well, so as long as the algorithm that calculates the speed takes it into account it should be correct.

Scott.

[nmdhqbc]

GPS knows about height as well, so as long as the algorithm that calculates the speed takes it into account it should be correct.

I didn't know that, thanks. I just always though of GPS as the coordinates I get from google maps sometimes. I guess there's more to it.

[PhaseThree]

Unfortunately the GPS height information is around 3x worse than horizontal information, so you can't get a accurate fix within 10-20 meters. This combined with the inherent problems with GPS signal reflection etc means that a good indication of speed can only be obtained on a flat open road.

The old fashioned alternative is to use the roadside marker pegs set every 100m, over 10km or so you can get a good average speed indication

[Arborbridge]

This, of course, assumes your satnav is correct.

I always wonder if the Sat Nav's are clever enough (or have enough info) to take into account the gradient. If they only have GPS coordinates to work on it would under estimate speeds when you go up or down hill. It being the longest side of a right angled triangle and all.

GPS knows about height as well, so as long as the algorithm that calculates the speed takes it into account it should be correct.

Scott.

I'm not clear that heights are accurate enough with GPS - not unless a phone is much cleverer than a marine GPS unit. The GPS signal can also jump around by a considerable distance (as you'll find out if you move slowly or are stationary), so how good the speed result is depends ultimately on how good the smoothing and error trapping algorthim is. With GPS for walking using OS maps, I'd expect the height information to come from contours at the known position. This allows the app to draw a profile cross section of your walk with information on ascent and descent. I doubt that this uses only GPS derived height.

Arb.

[Arborbridge]

Unfortunately the GPS height information is around 3x worse than horizontal information, so you can't get a accurate fix within 10-20 meters.

Sorry, I wrote mine before seeing this. Nice to have some confirmation of my suspicion, though.

[jackdaww]

This, of course, assumes your satnav is correct.

I always wonder if the Sat Nav's are clever enough (or have enough info) to take into account the gradient. If they only have GPS coordinates to work on it would under estimate speeds when you go up or down hill. It being the longest side of a right angled triangle and all.

GPS knows about height as well, so as long as the algorithm that calculates the speed takes it into account it should be correct.

Scott.

=========================

does the satnav get the elevation data from the satellite , or from the map.?

[swill453]

does the satnav get the elevation data from the satellite , or from the map.?

Well it can get it from either, so your question would have to be more specific. I wouldn't know the answer though

I use the Viewranger app for hillwalking, and it shows me a graph of map height vs. GPS height, and lets me choose either for the track summary. They follow each other pretty well on a macro level, but not completely the same.

GPS height on the top of mountains seems to match the "official" heights to within a couple of metres. Which since the device isn't at ground level, and may be in my pocket while I'm standing on top of the summit cairn, isn't too bad.

Scott.

[UncleIan]

I'm not clear that heights are accurate enough with GPS - not unless a phone is much cleverer than a marine GPS unit.

Am I being dim (or a smartarse about to get schooled), or would marine GPS units not really care about height? Height being sea level?

[Arborbridge]

I'm not clear that heights are accurate enough with GPS - not unless a phone is much cleverer than a marine GPS unit.

Am I being dim (or a smartarse about to get schooled), or would marine GPS units not really care about height? Height being sea level?

They may not care about it, but you can view what the GPS "thinks" it is on one particular page. That gives me an idea of the accuracy - I can't quote numbers off hand, but I think sailing at 30 feet isn't unusual!

Arb.

[Arborbridge]

does the satnav get the elevation data from the satellite , or from the map.?

Well it can get it from either, so your question would have to be more specific. I wouldn't know the answer though

I use the Viewranger app for hillwalking, and it shows me a graph of map height vs. GPS height, and lets me choose either for the track summary. They follow each other pretty well on a macro level, but not completely the same.

GPS height on the top of mountains seems to match the "official" heights to within a couple of metres. Which since the device isn't at ground level, and may be in my pocket while I'm standing on top of the summit cairn, isn't too bad.

Scott.

I hadn't realised there was a choice. I have viewranger, but I can't get on with it so well as the OS equivalent.

[AsleepInYorkshire]

SAT NAV VS. SPEEDOMETER: WHICH SHOULD I TRUST?

Garmin

AiYn'U

[didds]

Very interesting.

But if I want to drive at 65mph and that's what my speedometer is indicating that's what I'm gonna do regardless of holding up any caravans.

Hopefully, though, not in the middle lane

but maybe if the inside lane is full of trucks doing 50mph...

didds

[Lootman]

Very interesting.

But if I want to drive at 65mph and that's what my speedometer is indicating that's what I'm gonna do regardless of holding up any caravans.

Hopefully, though, not in the middle lane

but maybe if the inside lane is full of trucks doing 50mph...

Yes, as I recall this entire issue arose because someone claimed that the most relaxing way to drive was in the left lane. And I don't think that is the case at all. It's full of trucks, as you say, with resultant loss of visibility, plus potentially smoke, fumes, spray and so on. Plus you constantly have traffic merging from your left.

The middle lane is more relaxing. It's not a problem when the left lane is a convoy of trucks, as noted. But it is when the road is open and quiet, when nobody would need to "hog" the centre lane anyway.

On crowded motorways then it is most relaxing to minimise your number of lane changes, even if that means accepting that your speed may vary as a result because of prevailing traffic.

[swill453]

The middle lane is more relaxing. It's not a problem when the left lane is a convoy of trucks, as noted. But it is when the road is open and quiet, when nobody would need to "hog" the centre lane anyway.

The above is merely paraphrasing exactly what the Highway Code says, i.e. drive on the left unless overtaking slower traffic.

On crowded motorways then it is most relaxing to minimise your number of lane changes, even if that means accepting that your speed may vary as a result because of prevailing traffic.

Each to their own, not everybody's primary aim is to be relaxed when driving though.

Scott.

[Arborbridge]

Very interesting.

But if I want to drive at 65mph and that's what my speedometer is indicating that's what I'm gonna do regardless of holding up any caravans.

Hopefully, though, not in the middle lane

but maybe if the inside lane is full of trucks doing 50mph...

didds

Yes, of course. I made the assumption that the inside lane was empty. Silly of me