Is there a problem with my Garmin 705?
#1
Thread Starter
Sucking Wheel at the back
Joined: Jul 2009
Posts: 779
Likes: 1
From: Bristol, VA
Bikes: Lynskey Helix Sport, Lynskey M290, Cervelo S3
Is there a problem with my Garmin 705?
I was just looking at my ride from yesterday and noticed that if I turn the elevation correction on, (Which uses professional survey data instead of the altimiter in the 705), That my climbing numbers are WAY different.
Here's the ride.
https://connect.garmin.com/activity/77661566
According to my 705 I had 3389 feet of climbing.
If I turn on the elevation correction, which I think would be more reliable, I had 4615 feet of climbing.
This is my second 705. The first one I got from Garmin the altimeter didn't work brand new out of the box. Wondering if they are not very accurate for this.
Which one is right?
Here's the ride.
https://connect.garmin.com/activity/77661566
According to my 705 I had 3389 feet of climbing.
If I turn on the elevation correction, which I think would be more reliable, I had 4615 feet of climbing.
This is my second 705. The first one I got from Garmin the altimeter didn't work brand new out of the box. Wondering if they are not very accurate for this.
Which one is right?
#4
Senior Member

Joined: Oct 2009
Posts: 183
Likes: 0
From: San Diego, CA
Bikes: '09 Fuji Team Pro
I thought the altimeter would be more accurate than survey data. Survey data typically records a point every 30ish feet and interpolates the points in between. On the page for your ride, down in the Device Info section, there is a '?' next to Elevation Corrections. The note says that devices that contain an altimeter generally contain accurate elevation data so they disable elevation corrections by default.
One thing that could effect elevation data is drafting. Since an altimeter uses atmospheric pressure to determine altitude, drafting will affect it slightly because there is less air pressure in the draft than there is off the front. The difference is usually like 50-100 feet, though, in my experience.
neneboricua
One thing that could effect elevation data is drafting. Since an altimeter uses atmospheric pressure to determine altitude, drafting will affect it slightly because there is less air pressure in the draft than there is off the front. The difference is usually like 50-100 feet, though, in my experience.
neneboricua
#5
Are you using the 1-per-second recording setting? The older 705s could run out of memory, but the newer versions don't seem to have that problem. I always leave the 1-per-second turned on.
Garmins can calculate accurate elevations when you pick "Average" when saving a location. It usually takes at least 30 seconds for the displayed elevation to settle down. I save the location this way for commonly used starting points. Then the start elevation will be accurate. Otherwise, it's an estimate based on the air pressure only.
In general, my 705 seems to be correct within 50 feet or so. On a Blue Ridge Parkway ride last fall, I would often be within 20 feet of the elevation posted at the overlooks.
I save my rides on the free "My Tourbook". It has an option to correct the elevation data. I always keep the actual data instead of correcting it.
Here's the 705 recorded elevations, with a red line overlay of the SRTM satellite survey elevation data. It's pretty accurate for most of the ride. Each horizontal line is 50 feet. The red line actually shows more short, steep sections that don't actually exist on the Parkway. Since the road is along the side of a steep mountain, even a small shift in location off the road would make a big change in the elevation.
Even the fast downhills, with the extra wind pressure, are accurate.

Zoomed:
Garmins can calculate accurate elevations when you pick "Average" when saving a location. It usually takes at least 30 seconds for the displayed elevation to settle down. I save the location this way for commonly used starting points. Then the start elevation will be accurate. Otherwise, it's an estimate based on the air pressure only.
In general, my 705 seems to be correct within 50 feet or so. On a Blue Ridge Parkway ride last fall, I would often be within 20 feet of the elevation posted at the overlooks.
I save my rides on the free "My Tourbook". It has an option to correct the elevation data. I always keep the actual data instead of correcting it.
Here's the 705 recorded elevations, with a red line overlay of the SRTM satellite survey elevation data. It's pretty accurate for most of the ride. Each horizontal line is 50 feet. The red line actually shows more short, steep sections that don't actually exist on the Parkway. Since the road is along the side of a steep mountain, even a small shift in location off the road would make a big change in the elevation.
Even the fast downhills, with the extra wind pressure, are accurate.

Zoomed:
Last edited by rm -rf; 04-07-11 at 11:24 AM.
#7
Senior Member
Joined: Mar 2005
Posts: 1,701
Likes: 1
From: Moraga, CA
Bikes: 2008 Cervelo RS, 2011 Scott CR1 Elite, 2014 Volagi Liscio
The ascent gained on my 705 is always about 10-11% less than what's calculated by Garmin Connect or Strava after I've uploaded the ride data. I have more confidence in the 705's altimeter based estimate since I've actually tested it versus USGS topo maps. I usually start and finish the ride at the same point so I also correct the ascent nos. based on the different altimeter readings before and after the ride.
#9
Descends like a rock
Joined: Oct 2010
Posts: 4,034
Likes: 16
From: Fort Worth, TX
Bikes: Scott Foil, Surly Pacer
#10
I think your GPS is correct, not Garmin Connect. Edit-- the ride has a 1500 foot climb at mile 15, and a 800 foot climb at mile 29. Add a half dozen 200 foot climbs, and the rough total is about 3500. So the 4600 number seems high.
That looks like a great route.
I downloaded your tcx track from Garmin Connect, and imported it to My Tourbook.
Here's the elevation graph with a red line elevation correction. Each of those small red "hills" adds up, 30,50,80 feet at a time. My Tourbook shows 3156 feet of climbing, counting 3 foot differences. Garmin Connect probably uses a different set of elevation and mapping data, so it doesn't compare exactly.

Here are the track points in the middle of the big climb. Each dot is a track point, the uphill ones are close together, of course. The dot color is speed, green is fast, blue slower. See how the track drifts off the road. This would be one of the red spikes on the downhill. Even though the uphill dots follow the road, the road's exact elevation at any spot is calculated from the nearest elevation data points, and can't be perfectly accurate.

Mile 15.5 to 16.5 is Hayter's Gap Rd between N Fork Rd and Brumley Gap Rd. It looks flat on Google Terrain View, but shows a phantom 100 foot hill on the red track adjustment.

L is the 15.5 mile mark, R is 16.5
That looks like a great route.
I downloaded your tcx track from Garmin Connect, and imported it to My Tourbook.
Here's the elevation graph with a red line elevation correction. Each of those small red "hills" adds up, 30,50,80 feet at a time. My Tourbook shows 3156 feet of climbing, counting 3 foot differences. Garmin Connect probably uses a different set of elevation and mapping data, so it doesn't compare exactly.

Here are the track points in the middle of the big climb. Each dot is a track point, the uphill ones are close together, of course. The dot color is speed, green is fast, blue slower. See how the track drifts off the road. This would be one of the red spikes on the downhill. Even though the uphill dots follow the road, the road's exact elevation at any spot is calculated from the nearest elevation data points, and can't be perfectly accurate.

Mile 15.5 to 16.5 is Hayter's Gap Rd between N Fork Rd and Brumley Gap Rd. It looks flat on Google Terrain View, but shows a phantom 100 foot hill on the red track adjustment.

L is the 15.5 mile mark, R is 16.5
Last edited by rm -rf; 04-08-11 at 09:51 AM.
Thread
Thread Starter
Forum
Replies
Last Post
hobkirk
Road Cycling
47
01-13-22 10:40 AM
TromboneAl
Electronics, Lighting, & Gadgets
5
06-03-13 10:53 AM
CBuff
Clydesdales/Athenas (200+ lb / 91+ kg)
22
08-08-12 02:17 PM







