Edge 510 elevation readings issue

Yost

I programmed a ride into Plotaroute which showed around 2000ft gain, but when I did the ride, my Edge 510 claimed 2759ft. I programmed the same route into Ride With GPS and this also showed 2000ft. So why is the Garmin so far out? I believe the Edge 510 uses the barometric altimeter to calculate elevation, so does it mean that something is wrong with this? If so, is it fixable, or is it possible that both of the sites I used to plot routes are showing the wrong figures and the Edge 510 is actually correct? Cheers.

matudavey

If there's changeable weather it can affect the pressure quite considerably.
I've seen a ride that looked like someone climbed a couple of hundred metres when they restarted after a lunch stop!

meanredspider

Yup - you just need a pressure front to pass through. Did you ride a loop and was the altitude the same at beginning and end? Have you tried elevation correction on whatever data tool you are using (Garmin Connect etc)? Altitude is always a bit of an estimate but with the device and the route tools.

Yost

I did a 50 mile loop, same start/finish point. The weather seemed pretty consistent throughout. I didn't correct anything on Garmin Connect as wasn't sure whether it needed correcting or not. From what you're saying, it sounds like the Garmin may have been reading incorrectly so I should try correcting for this ride?

Yost

Just enabled elevation corrections and it brought the elevation down to about 1600ft! I've also tried this for some other rides, and on each occasion it is reducing the elevation figure by at least 30%, even those rides where my Garmin elevation reading had matched my figure from the pre-plotted route. Therefore, not sure that using the correction feature is helping either!

matudavey

the correction probably uses a terrain height database and will have to interpolate between the nearest points, so will always have inaccuracies.

Another issue can be the sampling frequency and measurement precision.
if the measurement fluctuates up or down a few metres, then a flat road can quickly add up a lot of climbing, as the reading drops a metre or two and then increases a metre or two.

This should be less of an issue with barometric altitude but can definitely affect GPS altitude. If you were using this measurement to get some sort of average, then the error would cancel to approximately the right value (which is why there may not be a mismatch between start and end elevation).

However, the climbing counter is only using readings one side of the real value, so will be skewed by the lack of precision.


You can set the altitude at certain locations, like home or the top of a hill, then garmin will recalibrate to that altitude when you're at that location.

First.Aspect

matudavey wrote:

If there's changeable weather it can affect the pressure quite considerably.
I've seen a ride that looked like someone climbed a couple of hundred metres when they restarted after a lunch stop!

This is rare. Rare like in 15 years of using such devices the most I've seen ascent/descent differ is about 5%. Usually less than 1%.

This OP's question is a bit like "how long is the coastline?" Answer is, "as long as you want", because its fractal.

Garmin is just counting pretty much every undulation - resolution is +/- 1m. Websites use a series of spot heights connected linearly. They give different answers.

Some websites also ignore variations below a certain threshold.

meanredspider

First Aspect wrote:

matudavey wrote:

If there's changeable weather it can affect the pressure quite considerably.
I've seen a ride that looked like someone climbed a couple of hundred metres when they restarted after a lunch stop!

This is rare. Rare like in 15 years of using such devices the most I've seen ascent/descent differ is about 5%. Usually less than 1%.

Depends where you live, too, I'd imagine. I don't suppose it happens very often in California but the North Atlantic will see it far more. I've seen it half a dozen times riding year-round in the Highlands. More likely, though, it's a combination of errors because it's a massive difference. Do a group ride over a large elevation change and compare data afterwards and you'll see that Garmin measurement alone is full of error.

sloppyschleckonds

Something odd going on with Garmin atm, i was doing a climb today which i know is only 25%, but garmin was showing 40%!

Lookyhere

When i had similar issues with grade and total ascent, Garmin basically said there are limits with the technology, altimeter and GPS all effect the ascent and gradient shown.
elevation correction seems the only way to sort it.