Battery usage stats

[t.parker]

I think it may be worth posting some battery usage stats (including information on how we use the watch) so that can build a picture of how long Sense 2 may last on a single charge. The way we use the watch particularly matters as it will affect battery life. Here's an example of my recent usage:

Battery High	100.00%
Battery Low	25.00%
Hours	42
cEDA	On
SpO2	Off
GPS Usage (hours)	04:10:00
Notes	Never Always-On Screen wake on gesture only during day, minimum brightness

I think most of my battery usage is GPS. 75% is gone in less than two days. Not using any fancy features (there are really none on Sense 2 anyway). The power consumption of cEDA is puzzling me as sometimes I see the battery down more, other times less and no different from not using cEDA at all. Unlikely, I will get 3 full days with this usage pattern.

I will continue sharing this data. With the next charge, I will stop using GPS and keep cEDA on.

[mike-MKvip]

Interesting, thanks for the info. Where did you find that stat?

[t.parker]

I just monitor my usage, the battery percentage you can see by swiping down. Simply, I manually log every day the battery state in spreadsheet. I include readings pre- and post- activities, etc. This is my current usage (real-world data). As you can see, I do use GPS a lot (running) and currently testing cEDA. This evening I will put it on charge and start another battery power cycle at midnight. Will exclude GPS though so that will tell how much using cEDA costs.

I would like other users to share their battery usage including how they use their watch. It may come handy for anyone looking for info on how long Sense 2 lives on single charge and find the answer from somebody who uses the watch similar way (to he honest, such thread would be helpful for every model).

[mike-MKvip]

Ah, OK, thanks.

[Guy_]

For comparison. On a Versa 4 getting about 7.5 days. At 36% after 5 days.

No GPS or Snore but have SpO2, timers, frequent screen use for testing, notifications including 1 min display timeout at times.

Battery is holding up better than expected and SpO2 has worked every night, though only the one value, no range is given.

[rbittman]

My observation with the Sense 2 is that battery usage is exactly the same as it was on my Versa 3. I have these settings:

cEDA: on

Brightness: Max

Always On: OFF

Notifications: just Text and Phone

Screen Wake: both gesture and button

GPS: doesn't connect, so not used yet

These are the same parameters I used on the Versa 3. After 1 day's usage, battery percentage is 84%, every morning. So, it's using 16% per day. At this rate, maximum battery life would be just over 6 days. However, I tend to charge it once a day, so it never gets that far. With the Versa 3, I backpacked for 4 days, and it did not need charging in that time.

[Killroy-TM]

My observation with the Sense 2 is that battery usage is actually excellent compared to my Versa 2. I have these settings:

cEDA: ON
Brightness: Normal
Always On: OFF
Notifications: eMail, Fitbit, Text, WhatsApp and Phone
Screen Wake: both gesture and button (Timeout 8sec)
Sleep Mode Schedule: from 11pm to 9am.
SpO2: ON
GPS: OFF

[moothee]

I got 4+ days (+- 96 hours) and 20% left.

cEDA on (without notification), no GPS, not much tracking, almost all notifications, AOD off, SpO2 on, screen wake gesture, multiple alarms.

Now testing without cEDA (anyway I am not sure it is useful :D). 

[t.parker]

Now, from a full charge, 12 hours drained 7% of the battery (including night time). If it would continue like that it would give me 6-7 days. Same conditions (cEDA on and no SpO2). Probably, if I cease using GPS I can squeeze 5-6 days out of it. cEDA doesn't seem to be a massive battery drainer (which is good news) but after a few days, I find it rather useless. The GPS on the other hand seems to drain the battery really quick but I'm not sure if it's more or less than Sense. I have some logs from the original Sense (a week before I got Sense 2):

Original Sense:

Battery High	100.00%
Battery Low	25.00%
Hours	93
cEDA	N/A
SpO2	Off
GPS Usage (hours)	07:01:00
Notes	Same usage pattern as for Sense 2

This is 3 hours more on GPS and the same battery level (100 to 25) within 93 hours. It's almost 4 full days (eventually, on this charge there was 1 more hour of GPS and by the evening battery went down to 0% which gave me 8 hours of GPS and a total of 122 hours (5 days).

Looking at battery drain with/without cEDA there isn't really much difference whether the sensor is in use (although there may be if it detects a lot more responses, I don't know that) so I think the majority of power consumption is done by the GPS on Sense 2 and that seems to be worse than on the original Sense. There are two sources telling how long Sense 2 lives on the GPS: the spec says 5 hours, manual says 12. I believe it's neither but probably 6-7 hours. Yet to be tested (Ideally side by side with original Sense in the same conditions with turned off cEDA on Sense 2).

Am I the only one using GPS? 😀

[rbittman]

Not the only one using GPS! I got it to connect yesterday and used it for a 5 mile walk. It mapped my route very accurately. It used more battery life, but probably about what I remember happening with the Versa 3. I won’t be using it that much because I have limited route options and it’s pointless to map the same route over and over. But I’m glad it works well. 

[t.parker]

@rbittman the distance without GPS is way off for me. On today's run of 8km, it's overestimated by 700m. The distance calculation uses GPS to correct the data coming from the accelerometer (at least I hope so, this is what all the watches do). I haven't had such a distance discrepancy when running with GPS (but I may be wrong). Also, without GPS the elevation data is gone (which is strange because it uses an altimeter to estimate elevation, not GPS, not sure why Fitbit stops tracking the elevation). Users who upload their run/bike/hike/walk activities to Strava (or other services) don't have much choice but to use GPS.

My run today without GPS ate only 2% of the battery so probably I could get several days on a single charge (and that would meet the claims of 6+ days). But it means using so few features that as well I could revive my old Charge 2 (and use connected GPS) 🤣

[rbittman]

How are you checking distance? Are you running on a track?

my Sense 2 stated distance was 4.99 miles. The Versa 3 I had before stated this route was also 5mi, give or take. So the devices are in agreement, but to check them, I’d need to ride my bike using the cyclometer or do some other independent check. 

You may be interested in the YouTube channel of the Quantified Scientist. He tests various trackers and compares them. 

His research indicates that the Fitbits are rather mediocre at HR accuracy (which you’ve confirmed). They’re pretty good at GPS and step counting (though not perfect). They’re really good (used to be best, prior to the release of the newest Apple Watches) at sleep stage tracking. I wanted to focus on sleep which is why I went with fitbit. When I was a competitive athlete I used a Garmin for better GPS and Body Battery (line cEDA, but better).

[t.parker]

@rbittman I know Quantified Scientist but unfortunately, it's not QS who does my running and cycling but me 🙂 I test with multiple GPS devices on known routes and also on a measured running track. I also use running POD. When cycling, the distance is measured by Fitbit using GPS only (and it comes very close to the bike speed/cadence sensor data, no issues here, at least on Sense, haven't cycled yet with Sense 2). Fitbit has problems with distance on hills (no 3D-distance support, doesn't take an elevation into account), and big problems with a higher cadence as legs can move faster than arms which distorts distance calculation and steps recognition (arms not in sync with legs).

Steps are not that relevant to me as Fitbit ignores cadence. On my Sunday run, it undercounted the distance slightly (not too much, 100m - 120m on 21km route) and undercounted steps by 550 (compared with POD which is on my foot). In today's run, it was undercounted by 140 steps. Nothing significant. 

Now, why the distance without GPS is so off? It's because it's hardly possible to measure distance just with the use of an accelerometer without having any other data to correct it. The accelerometer can measure distance only in a very short period of time and if not corrected it will start drifting. That's why Fitbit has a setting for stride length. But stride length for non-GPS tracking is kind of useless. Here's why:

Stride length changing during run

The chart above shows how my stride was changing during the last run. So without GPS, there is nothing to correct accelerometer drift and Fitbit must rely on the fixed stride length and try to figure out the distance based on that (it doesn't do stride x steps, that wouldn't work, it's more complicated). Hence, for some people such distance will come close to what they expect, for others, this will not work. This is one of the reasons why I need GPS 🙂

[t.parker]

With no GPS at all I got 6+ days:

Battery High	100.00%
Battery Low	13.00%
Hours	162 (6d18hr)
cEDA	On
SpO2	Off
GPS Usage (hours)	0
Notes	Exercises tracked but without GPS, Always-On off

cEDA probably takes away some of the battery life and I could probably get 7-8 days from a 13% charge but that's without GPS. For people who don't care about GPS data (so tracking calories, active minutes, and roughly distance with no GPS) it should be good enough. For me, that is pretty much disqualifying as without GPS the elevation gain isn't tracked and the distance is off (my stride is not calibrated by the GPS data). Yet, I need to compare how long the GPS really lasts on Sense 2. So far, a 6+ days promise has been delivered including (a rather useless) cEDA sensor turned on.