John_V 

Pretty much correct. It doesn't matter if the device displays maps or not or if it uses a speed sensor or not, the lat and long and other needed data fields has to be parsed form the satellite data in order for the route to be captured. In the case of a Garmin, the ride data is stored in memory on the device and transferred to the Garmin Connect app which in turn creates an XML file of the data and sends it to the various websites (Garmin, Strava, etc) to be displayed. GPX, TCX, KML and FIT files are all modified XML files. FIT files are compressed XML files, sort of like a ZIP file on a computer. When using a speed sensor, the only processing that you are eliminating is parsing and calculating the byte array elements that have to do with speed and distance. It will use the sensor data over GPS since it can work with it, with or without a satellite signal. I don't see where having maps or not having maps, using or not using a speed sensor will do anything drastic in prolonging or exhausting battery life.

Using a cellphone for a cycling computer is no different when it comes to the processing of satellite data. The only advantage that a cellphone has is that there are several things that you can do, along with using third party devices, that will extend the battery usage for your ride. The problem with any battery is that after time and countless charges, they all tend to drain more rapidly when they reach a certain point. My almost 3 year old iPhone 5c battery will drain itself pretty quickly when it gets below 40%, regardless if it's being used as a cycling computer or not.

John_V 

Correct!

John_V 

I went and looked at some of the code we used in the project but couldn't find the text file containing simulated GPS data that was used to write our parsing routines without having to be connected to a GPS. I found this online and thought it might be helpful in understanding what is being read, calculated and displayed by the GPS unit. This is contents of a byte array converted to a comma delimited string array so that people can read it. This data is actually one continuous line broken down for easier reading and is only part of what is transmitted by the satellite.

$GPRMC,235316.000,A,4003.9040,N,10512.5792,W,0.09,144.75,141112,,*19
$GPGGA,235317.000,4003.9039,N,10512.5793,W,1,08,1.6,1577.9,M,-20.7,M,,0000*5F
$GPGSA,A,3,22,18,21,06,03,09,24,15,,,,,2.5,1.6,1.9*3E

The $ designates the beginning of an array element (called a GPS Sentence). The letters after the $ designates what data is in the array element. If you want to get lat and long you would read the second line. the 235317.000 is the time of day in 24 hr format (23:53:17), the 4003.9039,N is the long, the 10512.5793,W is the lat and if you want to get the altitude, it's the value 1577,9. So these are just three of the elements that get parsed, calculated and saved each second. And yes, programmers have a cheat sheet that breaks all this down for us so we know what the heck we're looking for.

njkayaker

You can disable the GPS receiver (on some of the units). That might save a little bit of power. If you are not on the map page, the device doesn't need to waste CPU cycles (power) drawing the map (it might be writing the map display directly to video memory). If you have sensors enabled, it doesn't need to waste CPU cycles seeing if there's data being transmitted. Not using navigation means fewer CPU cycles. The device certainly disables/enables features and features that are disabled very-likely save power (but probably not "huge" amounts). There doesn't seem any indication that the GPS data collection works differently with the speed sensor installed. That the data is written at 1 second intervals doesn't mean the unit isn't collecting it at a faster rate.

It doesn't seem that FIT files are XML.

Not doing stuff will (very likely) save some power. It likely won't be drastic but it might be noticeable (for some things).

JohnJ80

This is exactly the strategy that is adopted in almost any mobile device. Battery is a super precious system resource. Saving 1% on a 10 hour battery life gives back 6 minutes. Do that 5 times and you have an additional half hour. So if you increase the standby time of the GPS receiver - a big power consumer - you are going to get signficant benefit.

As mentioned, these batteriy chemistries have the characteristic of losing capacity over time for a variety of reasons including just plain old aging. Getting back a few percent here or there also helps to make up for that.

J.

Seattle Forrest

Everybody is forgetting or ignoring the fact that your Garmins work with sensors (speed, heart, temp, power, cadence for bike or foot) by powering a radio receiver to "hear" the signal.

Also for outdoor rides, Garmins record a trace, so the GPS is running whether you use it for speed or not.

Finally: "One good test is worth a thousand opinions."

JohnJ80

At 15mph, one covers about 22' every second. There is no need to maintain that level of linear accuracy on a GPS for a "breadcrumb" trace or for turn by turn directions. However, you would need to have updates every second in order for the speed to not have too much latency on the display. Therefore, I'd be very surprised if they are not dropping the GPS active time if there is a sensor available for speed and increasing it's active time when there is no speed sensor.

J.

Seattle Forrest

I say the speed sensor does nothing to improve battery life and probably reduces it very slightly. This is based on common sense and on understanding how everything works.

If anybody disagrees, put it to the test, and share your results.

I'd test it myself but I took my speed and cadence sensor off my bike when I got a pedal-based power meter (which also reports cadence, which has some value unlike speed). I've since given it away so I have no way to test this.

njkayaker

Not everybody. I pointed out that not enabling sensors would likely use (a bit) less power.

Having no sensors enabled on the head unit might/should reduce power consumption. Having multiple sensors might not use more power than having one (but there is likely more code being run with multiple sensors).

JohnJ80

What they are able to do and what they should do for battery conservation are two different things. I believe Garmin has their own GPS engine internal (designed their own RF and digital GPS chips). Almost any of the chips used in mobile applications have the ability to power down or go into standby with a rapid wake. Many are much more sophisticated where they can power down portions of the circuit internal to the same chip. I'd be shocked if Garmin hadn't done the same on their chip (it would be stupid not to).

But, there is no way to know. I'm sure they aren't going to say (I suppose I could ask them) because that's pretty trade secret stuff. That said, that's pretty much what most companies do (or are forced to do).

I don't think I'm understanding what you are trying to say. It would be the other way around. They increase their GPS standby time in the presence of an external speed sensor.

J.

JohnJ80

Convenient.

njkayaker

Who knows if Garmin designs their own chips.

The issue is whether they do something different with the sensor than without.

They record position data once per second with or without a sensor. The devices likely calculate speed using the data they are already acquiring for other reasons. Given that the speed is laggy (less responsive) without a sensor, there is no reasonable expectation that they would acquire position more frequently without a sensor. That is, the once-per-second data is good enough to calculate speed and there is no indication that the measured speed is better than whem there's no sensor.

The speed measurement is worse without a sensor.

You keep saying that Garmin increases power consumption to get a worse measurement. That's a silly trade-off. It makes no sense.

If they are recording data, the devices likely don't increase GPS standby time with or without a sensor.

njkayaker

Who knows if Garmin designs their own chips.

The issue is whether they do something different with the sensor than without.

The speed measurement is worse without a sensor.

You keep saying the devices increase power consumption to get a worse measurement. That's a silly trade-off. It makes no sense.

Without a sensor, the devices likely just calculate speed as a by-product of what they are doing anyway (recording data every second).

Which is saying the same thing as decreasing the standby time without a sensor.

They don't do that because they are recording data every second anyway (with or without a sensor).

The devices just use the data they require for that to calculate speed as a by-product. The speed from that is "good enough" and there's no indication that it's better than that without a sensor.

Seattle Forrest

Very, the pedals stay in place when I put the bike in the car and drive 200 miles.

Anyway, I don't have a horse in this race. If you think it's "convenient" that I'm unable to prove that some crack-headed theory on the internet is wrong, you should know I have better things to do with my time. But if you want anybody to take your guesses seriously, I suggest you bring proof and not analogies to how unrelated things work.

JohnJ80

If you don't have a dog in this fight then exactly why are you being so aggressive? Odd.

I'm in the semiconductor business in an engineering capacity and I have way more than a passing familiarity with how this is done from an engineering perspective. To give you further background on this would require NDAs - which is also convenient because it brings this to a close.

J.

RidingMatthew

Thanks for the responses
I moved the sensor to the front wheel and it has not pause/ resumed except for when I stop.
I have ridden under tree cover with the speed sensor and the device tells me it is locating satellites took half a second and then went back to displaying data.




I rode my old road bike with a cateye computer and the Garmin (no speed sensor) and the computer were pretty much the same. they started and stopped at about the same time. when I had a computer on my commuter it was not as precise as the Cateye or GPS. I have noticed that when I turned off the connection for the sensor it started and stopped more with my bike. In other words I could roll to a stop and when the speed sensor was engaged, I would be stopped for a second or two before it realized I had stopped. WHen I was rolling again, it was fairly quick to realize that.



Why would you disable the gps receiver on a garmin? that is how it draws maps etc. I noticed that to really extend battery life the display brightness is the biggest user. if you turn off the lap warning and it doesn't light up the screen. I have ridden most of the day with the Garmin 8:08:42 8 hours of being on running speed and cadence sensors as well. That is pretty good battery life and it still had more because I looked at our ride in camp after the ride.


calm down @Seattle Forrest and @JohnJ80 no need for people to hijack my thread.


I just wanted to know if it was silly to run a speed sensor on my commuter/ road bike.


I can see that running it on a mountain bike would allow the Garmin to know wheel size and count rotations. That would allow you to capture distance data more accurately especially with tight sometimes slow turns of mountain bike trails.

njkayaker

You wouldn't (generally). But if you aren't using the maps, you don't need the GPS (if you have the speed sensor). These things are often used without using the maps.

It wouldn't be silly. Lots of people use the speed sensor. If you are interested in "personal records" for speed, you might want the speed sensor. If you use your bike on a trainer, the speed sensor is nice to have. Cadence might be useful and, if you want cadence, you might as well get the combo cadence/speed sensor.

But you don't need the speed sensor (lots of people don't use them).

RidingMatthew

oh I have never thought of that I like it to draw maps when I upload to garmin or strava

01 CAt Man Do

Boy this thread sure turned into a pissing contest. I have nothing to add other than when I first got a smart phone I downloaded an app that supposedly would give me my speed via GPS when riding. Thinking that would be cool to have I tried it and quickly observed on my very first ride that there was a very large time lag on the speed display when it came to how fast I was actually riding. Not to mention that the app itself ate battery power like crazy. Took me all of ten minutes to determine that a device that relied on GPS only for speed was not going to be really in sync with my real time speed.

If I owned a Garmin that worked with wireless speed sensors I would certainly give it try. Sadly though I have little faith in the current wireless sensors otherwise I'd likely own one of the newer garmins by now. I only say that because I keep reading about all the complaints other people keep posting when it comes to how screwed up their wireless devices work. Nope, I'll hold on to my wired bike computer. Old tech but it's never let me down.

JohnJ80

I have had the wireless sensors on my bikes for the last year with no problems. I have been using the garmin ANT+ speed sensor on the rear hub and the Wahoo RPM cadence on the crank. If I did it over, I'd go the new Wahoo speed sensor on the hub just because it supports both BTLE and ANT+ plus because the RPM works so well.

Most people run the speed sensor on the front hub, but I didn't like how it looked there. If you ran it on the front hub, there is going to be a lot less stuff in the way for the radio transmission and it's hard to see that being a problem.

I absolutely love not having to deal with the stupid magnet alignment issues that come with the old style sensors. I haven't touched or adjusted any of the sensors on three bikes since I put them on. And that includes on my fatbike that took quite a bashing this last fall and winter.

So, Cat, I'd be perfectly comfortable recommending the current wireless sensors to you.

J.