You are hereForums / Preparation / Geek Gear (GPS/Comps/Lights/etc.) / Garmin 500 Barometric Altitude Accuracy?
Garmin 500 Barometric Altitude Accuracy?
I use a Garmin 500 to log my rides and have noticed today that I am getting some dubious altitude readings.
Today I got a 150m spike, up and back down again, in the first 200m of the ride log. Now I know I am good but even I couldn't ride that. I also did Bobbin Head Fire Trail down to the top of the steps, this was logged as being at -15m. I have gone back and checked past rides and last week it logged the same spot at -20m. The head unit had a software upgrade between these two readings
I have elevation corrections turned off as I believe the 500 has barometric altitude and that is the garmin recommended setting. If I turn it on then my 500m vertical ride goes to well over 1000m which I don't believe is correct.
Any suggestions as to what I am doing wrong or how I can fine tune this?
Thanks
- Login to post comments
- Bookmark & share
There is an inherent problem that has been well discussed on many web sites with Barometric devices (see links below). Just riding on an indoor trainer with garmin running one can get an increase or descrease in altitude due to changes in temperature.
These devices provide a indicator of height (and distance) and are not accurate to the Nth decimal place. I've recorded altitude and distance from a couple of sources and always found differences between them. Check out the differences on something like the various readings on Garmin Connect site for the Cape to Cape - there can be 3 or 4 hundred metres difference in height (and kms for distance)
Average difference between Garmin and Google Earth is 1:1.06, that equates for every 1000 metres registered by Garmin device, Google says it is 1060 mtrs - as long as you go down to below 4km above the ground in G-Earth. Any higher and you get a larger error.
Best bet is to make sure you have calibrated your Garmin as per its manual and do this regularly - and for every location you typically start a ride from, find out its elevation and program it in.
http://forums.groundspeak.com/GC/index.php?showt...
http://forums.groundspeak.com/GC/index.php?showt...
http://www.dcrainmaker.com/2010/05/understanding...
PS: not claiming Google Earth is any better - just it is reasonably consistent.
Garmin's patent- http://www.patents.com/us-6970795.html
To stop it saying that you are below sea level you need to set an elevation point. You can do this by going into Menu -> GPS -> Set Elevation. Before setting the point you will need to find out how high above sea level is where your want to set the point.
You should set one of these where you normally start a ride. When you press start recording on your garmin it will pick up the elevation point if you are within 100m or so of it.
Thanks guys.
I should be ok on distance travelled as I have put in wheel circumference and use a speed sensor. Surely it is programmed for that to take precedence over the GPS readings? As for the circumference I rolled out three or four full revs, did this 3 times and then averaged everything out, with me on the bike, so that at least should be reasonable accurate.
Switch it on some time before you ride, they seem to take time to stabilize. Also make sure the little breather holes don't become blocked, a drop of water is sufficient to block them.
If you set distance by roll out of the bike alone then it will be innacurate. Assuming you had the tyres at your normal pressure, when you add your weight in full kit + water etc. (75kg+?) you will see the tyre bulge, reducing your rolling circumference. Set without the weight it will have a higher level of innacuracy. I set my 500 on auto and let it sort itself out
. But it doesn't matter how you set them up there are inbuilt errors so accuracy for height and distance isn't spot on.
Just have to accept a little innacuracy as part of the system but it is usually accurate enough for practical purposes.
The gps distance is averaged over 5k so it works out how many revolutions in 5k and then uses that. Given they are +- a few metres that makes the GPS much more accurate than using circumference. The software does seem to defer to the sensor if GPS reception is dodgy.