How big of a deal is using a speed sensor on your real wheel with a Garmin, versus not using one?

I have to assume that the speed sensor is more accurate than GPS alone?

What's the difference between using one or not?

Use it if you got it.

GPS can be very accurate in sustained constant speed. But cyclists usually aren't doing that. Further using GPS or GLONASS for speed (depending on your model), uses much more battery for the same task....and is much less sensitive/accurate to accelerations.

Yes. But probably not by much (outdoors).

I don't use the speed sensor with my Edge 800 or Fenix 3. The two of them are remarkably close at the end of each ride, it's typical for them to differ by about 1/20 mile for every 10 miles traveled.

If you have a speed sensor, your Garmin will use it to determine speed and distance; if not it'll fall back to GPS. You don't have to do anything to make that happen.

I just called their support number.

The guy said speed sensor is only used on a trainer. Useless on an actual ride.

Interesting...

Garmin support is generally as useful as a one legged man in an arse kicking contest.

If you have a gps, the speed sensor is not required, but if you're riding where the GPS signal isn't reliable, you're better off with a speed sensor if you're interested in things like your max speed since a GPS glitch will often trigger some max speed reading that isn't reality.

^This.

The GPS alone will give you reliable numbers for the total distance, average speed, etc. To keep up with instantaneous numbers all the time -- your current speed, your max speed, etc. -- the speed sensor helps. Depending on conditions -- terrain, tree cover, or other interference -- the GPS may not give you readings every moment of your ride.

If you have the sensor you could prove that this isn't true by riding through a tunnel. If the speed sensor isn't used, your speed will drop to zero or null.

I used to use older hiking GPS units and this would happen occasionally. With newer units, it doesn't happen. They've gotten pretty good at filtering the most obvious noise out of the signal. The latest Garmin units seem to use accelerometers (and even the compass) to predict where you're about to be and ignore GPS points that are wildly out of line from the predictions.

I agree with DonBjr & Athens80.

My wife has a bike with a Cateye for indoor training, but she also has a Garmin Fenix 2 (watch style Garmin), so she rides with both Garmin and Cateye. She comments all the time about the variations between the two devices on the current speed. At the end of the ride, the statistics match up, but for real-time speed information, the wheel sensor is better. I don't know who needs their current speed to be 'that accurate'....

Yeah, I don't ride where there are tunnels to ride through, but I can see how that would be a problem.

It's not a big deal.

The speed sensor still works when you don't have a GPS signal.

The speed sensor reacts much quicker than the GPS for speed.

The primary measurement of GPS is location (and time). Speed is derived (in a simple way) and GPS speed becomes more accurate the faster you are moving (since the distance between points used to measure speed are much larger than the inherent "error" in GPS location measurements). Keep in mind that bicycles aren't "fast".

The wheel sensor should be more accurate for measuring slow speeds (but not many people would care!).

Where it might matter is if you are interested in an accurate maximum speed.

You probably won't be able to see your speed in a tunnel anyway!

You may be right with the newest garmin edge computers. My wife's 510 seems pretty good so far. However, on the same ride at the same place I did see the same max speed glitches on three different cellphones, and my own Suunto Ambit2 GPS watch. (In my case the Suunto movescount site shows my correct speed from my speed/cadence sensor, but strava ignores that data and just uses the GPS data and reports a false max speed.) This was a rails to trails ride under some tree cover.

I've got an old Magellan hiking GPS unit. One of the hikes I did last year showed a distance of many thousands of miles; what happened was exactly what you're talking about. Zooming in to look at the map, it shows me walking the trail, then for a moment I'm 50 miles away on a different trail, then I'm back on the one I was actually hiking. Same type of error many times, and the end result was nonsense.

I've been using the new Fenix 3 watch, and it has an entirely different GPS issue. It's never put my in the wrong county, but, instead, it cuts corners. It's like the thing can't accept or believe that I did a 90 degree turn, so it rounds it out some. It's still wrong, but it's an entirely different kind of wrong. Not that the system just had a glitch and I teleported to China for a split second, more like the software that asks "does this point make sense?" got it wrong. Here's what I'm talking about, I recorded these two GPS tracks together at the same time:





It's ironic and slightly disappointing that the brand new Fenix 3 isn't as good as the 5-year-old Edge 800. How does your Ambit 2 track?

What is funny?

That kind of wacko error was happening with Magellan's new Cyclo 505 unit last year.

I believe that. Their Explorist 710 will be my last Magellan, and probably my last hand-held hiking GPS.

I bet the issue is with acquisition time. The Edge acquires more often, resulting in less "rounded" corners, where as the Fenix acquires less, cutting corners. In a straight line they're probably both pretty accurate. Around curves... not so much.

With a speed sensor the auto pause / auto resume is triggered faster. And your computer will display 0.0 when stopped. I only have the speed sensor on two of my bikes and never miss it on the two that don't have it. The only reason I even have those sensors was from when I had a non-GPS ANT+ computer.

I move the magnet out of the way when using my road bike on rollers since the speed is so high its silly to count that distance for anything and all I'm interested in on those rides is HR, cadence and time.

I'm perfectly happy without a speed sensor. I have thought about getting a cadence sensor

Actually i found them extremely helpful and knowledgeable with their product.

I had a problem with my 510 and and after trying a few things they suggested it was corrected.

I've been using just the GPS for years with various units, but I'm thinking of putting the speed sensor on again. Over the long haul GPS is sufficient, but I do get some annoying behaviors. On a ride last weekend I had a max speed reading of 98MPH, which I think was a little bit low Also on many climbs I get constant auto-pausing even though I'm going well over my 2MPH auto-pause point.

If you don't care if it's accurate and only use the speed readout as a relative indicator during the ride and between rides than GPS speed is "good enough." GPS speed is calculated from measured distances and time intervals and then averaged so it has inherent hysteresis. Most cyclists in my group don't use sensors. But if you want accuracy (and consistent, instantaneous accuracy), then you need a sensor. Just depends on how you ride and what you need in any given situation.

I have cadence and speed sensors, both Garmin, paired to my Edge.

Have you tried setting it to "when stopped?" I get pretty good results this way.

I was thinking about this more this week. I can remember times on my rides where I was using my Garmin speed sensor, but the current speed would fluctuate in random places. A couple of stretches of road in particular. One is just wide open. The other is heavily covered with trees.

If my Edge was truly using my speed sensor during the ride, it wouldn't have those fluctuations since the sensor is active.

I'm wondering if the Edge defaults to GPS for speed unless it can't get a good signal for some reason. Meaning, I wonder if it makes a difference whether we are running the speed sensor at all when outside with clear access to the satellites.

You can test for yourself whether your Edge is using your speed sensor or GPS for speed and/or distance:

2. Go outside.
3. Start an activity.
4. Lift the bike.
5. Spin the back wheel.
6. Check speed/distance while the wheel is turning.

I think the Garmins always use the speed sensor instead of GPS if it's available. It's more accurate, and the unit wouldn't be able to tell immediately if the GPS signal wasn't good enough to use.

The older GSC-10 Garmin sensor is both speed and cadence. The speed sensor is faster to respond to stops than the GPS calculated speed. And I like having the cadence sensor. For the cadence, I use a strong neodymium magnet stuck to the end of my pedal spindle.


~~~~~

Here's an example of GPS inaccuracy near trees. It's interesting that one direction was accurate, but the other way drifted off the road. This wouldn't be enough to affect the speed calculation, but similar discrepancies at a slow speed during climbing in the trees would make a bigger difference.



Riding between 20 story buildings downtown shows way more drift--the signal from the satellites bounces off the buildings, affecting the distance calculations. Each dot is a GPS recording at one second intervals.