Hi @clabos 

Welcome to the Sonos Community!

Firstly, I wouldn’t pay too much attention to what Home Assistant tells you - the “idle” that we refer to is the fact that the digital amplifier inside the device has powered down after 3 minutes of silence. As far as I know, this information is not shared to any other devices, though it is recorded in the diagnostics.  All Home Assistant is telling you is that the device is online and not currently playing.

As for the recorded consumption rates that you’ve seen, and considering you’re recording higher values than expected on all your devices, I would test another, third-party product like a low-energy light bulb to verify the meter has been properly calibrated. If it does indeed still seem that the Sonos devices are consuming more than they should, I can only recommend you get in touch with our technical support team directly. You will probably need to supply some pictures of the meter readings, and some diagnostics submitted at the same time - you may want to prepare these beforehand to shorten the call.

I hope this helps.

Thank you Corry for your reply.

I did have a call with the Swedish Sonos support earlier last week but got dismissed by a seemingly junior support person who just repeated what is supposed to happen instead of listening to my observations to understand the issue.

I have now recorded two timelapse clips that show a 9-10W reading of a control lightbulb and 11-13W readings for both a Play1 and Play3 that are constant for 5 min (didn’t think it would make sense to record more because it won’t ever change).

Is there anyone in the central support team that can help me and look at this? If not, do you have any tips for how to get the attention of the local support team?

Thanks again.

/Claes

Edit: Submitted diagnostics too, ref 759498494.

I’m curious: Are any players Grouped? Are you using Line-In on the PLAY:5’s?

Hi @buzz 

Two of my 1:s are grouped into a stereo pair. One of my 5:s use line-in (TV audio). The rest of the players are stand-alone.

8 devices connect via wifi, 3 via ethernet cable.

Hi. This is interesting.  I think you should repeat the experiment with the line in removed from the Play:5.  All Sonos devices should have not been playing audio for 10 minutes (say)..

Incidentally, if three of your devices are wired the other devices will not connect over WiFi but over SonosNet (unless you have "disabled WiFi" on the wired devices).

I should have expressed this slightly differently. In SONOS speak “Rooms” can be Grouped, players can be “Bonded” into pairs or surround systems. The Bonded players then become a Room.

Thanks for feedback and ideas.

Regarding grouping (sorry, misunderstood) - yes, some players were grouped during my experiment. I sometimes group the dining areas and I noticed that the Play:3 was grouped with the kitchen Play:5 during my experiment. The Play:1 in my experiment was not grouped though.

I have now done additional testing (see below) and recorded videos but the conclusion is that regardless of what I do the Play:1 and Play:3 draws 13W constantly and the Play:5 18W. According to https://support.sonos.com/s/article/256?language=sv the Play:1=3.8W, Play:3=4.4W and Play:5=8W.

I am the only one observing this behaviour? 

Anyway, this is what I did:

Six Sonos speakers were connected (five unplugged) when I started the testing
a) Play:3, wifi
b) Play:1 (master), cable
c) Play:1, cable
d) Play:5, wifi (USED IN TEST 7)
e) Play:5, wifi, line-in
f) Play:1, wifi (USED FOR TESTS 1-5)

My test scenarios:
1) No grouped devices. No audio has played last 10 minutes. (21:10)
Result: Play:1 shows constant consumption of 13W for the duration of the recording (5 min)

2) Device (c) is using cable, now disabling wifi in the device settings. The iOS app crashed at the first attempt to disable wifi, but succeeded at next attempt. (21:23)
Result: Still 13W 

pWhile waiting for the next test I checked the meter itself by measuring an 11W light bulb, got 9-11W readings, so the meter in itself seems ok]

3) Removing the line-in cable for device (e) (21:38)
Result: Still 13W

4) The master (b) is connected by ethernet cable, now disabling wifi in the device settings. (21:50)
Result: Still 13W

5) Unplugging ALL devices except the test device and the master. (22:00)
Result: Still 13W, some flickering between 11-13W

6) With all devices except master still unplugged, a previously unplugged Play:3 device is measured. (22:07)
Result: 13W

7) With all devices except master still unplugged, a Play:5 device is measured. (22:14)
Result: 18W

I submitted diagnostics at 22:25, ref 499480291

Once all testing was done I had to restore the wifi-setting in the master, otherwise the other devices weren’t discovered when I plugged them back in. I don’t think I understand how the mesh Sonos wifi vs house wifi works…

All feedback and ideas are welcome. By the way, I haven’t heard back from the local support team, they said “we will contact you in case we get any new information” :/

Will try to contact them tomorrow.

Hi @clabos 

In the diagnostics, I can see your devices powering-down their amplifiers, so they are going into “idle” state as they should.

Your ethernet-wired devices will naturally use more power than the minimum possible as they are providing WiFi for the other devices and must respond to their communications, even when nothing happens - in essence, replying to “are you still there?” They are also receiving multicast packets intended for other devices on your network that they need to process to know to ignore - any activity on their incoming ports will increase CPU usage and thus the power drain.

There are also times when your wired Hall speaker is having to re-transmit packets due to transmission failures. Reducing wireless interference will likely be the answer to reducing this, and thus power consumption.

Grouping, as suggested above, may increase power drain due to the increase of traffic between devices, though I assume this would be marginal.

I think the main issue might be the meter you are using, however - I’m sure that the caveat on https://support.sonos.com/s/article/256 is there for a reason. The meter will not be reporting every measurement it makes, so is it reporting the average, the median, or the peak? The units might be peaking at higher power levels for very short intervals, and it could be the peak measurement that the meter reports. I recently watched a Linus Tech Tips video on YouTube where he’d spent a lot of money on a professional PC PSU tester - more than he spent on 2 cars - and, while I’m not saying that you’d need to do that to get an accurate reading, I’m sure professional-level devices cost more for a reason and while I have no idea what we use to validate our reported ratings, I’m reasonably sure it wouldn’t be available in consumer outlets (shopping outlets, not power outlets).

I hope this helps.

Thanks for the feedback and tips @Corry P.

It is hard to argue against this is due to a metering issue, I do use a consumer grade meter. But it would only be true if Sonos has designed the devices to make them impossible to measure by peaking the power exactly in the right moment (assuming the meter reports the peak consumption) every measuement cycle and for a very short amount of time (to not skew the actual consumption). It does not sound very likely to me, I would have expected the readings to fluctuate in that case. My readings have been stable througout my observations, for several minutes. I have measured several types of devices (light bulbs, low power led lights, TV, switches, servers etc etc) with this meter and all of those showed reasonable readings. Are Sonos devices fundamentally different compared to all of the other types of devices I have measured?

I am happy to be proven wrong by someone more knowledgeable than myself. It would also be very interesting to hear from other Sonos users if they get the advertised readings. Please post any readings and info about your meter/method here in this thread.

I am also happy to send one of my devices for testing, either to Sonos or to anyone interested/capable/kind enough to do proper testing.

clabos,

Try a modified measurement technique.

Rather than measuring only a SONOS unit, measure a SONOS unit in parallel with a stable power draw, such as an incandescent lamp. First you would measure the lamp, then subtract this from the reading when both are measured. SONOS uses a switch mode power supply that may be flustering the power meter. There may be a threshold for the required parallel lamp. A small lamp might not do the job, while a larger lamp might work better. By “small” I’m thinking in the 10W range. I’m thinking that you should use at least a 20W lamp. Obviously, a very large lamp will decrease your resolution with respect to the SONOS draw, but if this technique works around a peak/average waveform issue with respect to measuring SONOS draw, the technique will be a success overall.

Switch mode power supplies are very difficult to measure accurately. Which model power meter are you using? How does it handle power factor and harmonics?

Is there a significant reduction in standby power consumption for units that are ethernet wired and have their WiFi disabled? My Connect unit seem to run noticeably less hot in that mode, so that is likely to be the case? The next obvious question - where all Sonos units are ethernet wired, what is the advantage of leaving WiFi enabled if one does not need their WiFi for any non Sonos uses?

https://support.sonos.com/s/article/256

Please note that your Sonos product will consume approximately 0.3 W more with an Ethernet connection. With WiFi turned off, products can consume approximately 1 W less than typical figures.

None, assuming there are no wireless Sonos devices (including home theatre satellites) and no ancient CR100/200 devices. Android connection to SonosNet was withdrawn long ago. That said, Sonos Support will probably automatically request the radios be re-enabled if they’re contacted about any problem.

So the 5.6 watts of the Connect in stand by mode will drop to around 5 watts via the net reduction of .7 watts. I guess that isn't all of why mine seems to run less hot. But if there is no advantage in having the radio on, whatever is saved is saved and cooler running electronics can’t be bad.

And obviously Sonos kit is not to be compared to the typical modern amplifier that nowadays runs in about 0.5 watts in standby mode, a EU mandate I believe for kit sold in the EU. But it means that turning off the power to less frequently used Sonos kit may be financially better in the home on the direct power savings v shorter service life via power cycling give and take. For kit used daily, leaving it in standby may still be the cheaper thing to do. Unfortunately, there is no way to know for sure since the service life effect is not easily measured, assuming it exists in the case of Sonos kit.

With the radio off, perhaps the Connect has less to do in terms of shunting packets back and forth with other units.

Re wired vs wireless energy savings, your are probably ‘robbing Peter to pay Paul’, you save some energy by disabling wireless on Sonos device, but the ethernet will need to be plugged into a switch port or similar device, that will be using energy?

I have a smart meter that sends meter readings every 30 minutes to my energy supplier, so I can look at any 30 minute period. If I look during the night when the house is ‘idle’ I can see the average ‘idle’ energy consumed is pretty consistent with multiple devices still consuming energy, including Sonos devices. If you have Smart meter and ability to check readings taken every 30 minutes on the supplier website, you could turn off *all* Sonos devices one night, and compare hourly usage with them *all* turned on the next night, this would be a more accurate of actual usage for all your Sonos devices, you could then verify this with the published figures for total consumption of all your devices.

Also remember if you have a few devices giving off a little heat, you will be saving a small amount on your heating energy costs in the winter

TANSTAAFL

There are parts of the world that are more concerned about keeping the house cool! Where a class A amp, left on 24 hours as recommended in the audiophile world, would not help at all. My reason for leaving the WiFi off for all my wire connected units is just to reduce whatever temperature the Connect runs at and also to reduce chances for WiFi interference with other devices. The wire connection was done to assure very stable group play - not for allowing the Sonos radios to be off. That is just an associated benefit.

Hi
Thank you @buzz for your ideas about measurement approach. It seems this was very good advice!

I did two more tests by adding a light source to the measurement - firstly a 11W lamp and then a 40W lamp.

Since I don't have any data logger I made a 6 min video of each test and then manually noted the readings, second by second. Tedious but worthwhile it seems.

Test 1 - 11 W lamp + Play:1
Average power consumption = 21.06 W
Measured base load from lamp = 11 W
=> The Sonos added 11.06 W

Test 2 - 40 W lamp + Play:1
Average power consumption = 42.28 W
Measured base load from lamp = 38 W
=> The Sonos added 4.28 W

The “expected” line in the graphs below is the sum of the lamp + 3.8W for Sonos.

The documentation at https://support.sonos.com/s/article/256?language=en_US says the European version of Play:1 should draw 3.8 W. If I do the test again with a 60 W lamp I am probably going to get even closer to 3.8.

I consider my question solved now and I withdraw my claim.

Thank you all for contributing!

Thanks for the follow-up.