I would think it has to be. How else would it know to “wake up” when the app activates it?

People have complained about nit being able to turn off the bluetooth signal used for set up too……

One can hope that the device has an internal clock. This is not very difficult to imagine. For example Teufel has an option which is called Wifi standby  (On/Off) and is intended for this but unfortunately it does not support the Youtube music streaming service.

Regards

If a SONOS unit is wired to your network, this unit’s radio can be shut OFF, however, if this unit had been using its radio to wirelessly support other SONOS units, these units may not function. An automated approach is to power down the unit with a 3rd party timer.

Thank’s

It is a pitty that a timer is not embedded in to speaker itself for this kind of automation.

Regards

sl91

If you just want a bedside radio/clock/alarm, with no Bluetooth or wifi emissions, why not get a basic radio/clock/alarm? You’ll save a lot of money, too. 

Because I would like to have some top streamed music sound quality (Era100 for example)

Then I think, as @buzz says, a third party timer device is your solution. 

Hi @sl91 

Welcome to the Sonos Community!

The new Era models will in fact disable their WiFi radios when connected to Ethernet via the Sonos Combo Adaptor (unlike other Sonos devices connected to Ethernet). If you want to utilise WiFi, however, then a timer device would be needed to power-off the speaker while you are asleep, as @buzz suggested.

Note that powering a speaker off daily will likely shorten it’s life-span, in the long term (components will expand and contract with the changes in temperature) - just like any other electronic device.

I hope this helps.

@Corry P - please can you clarify this a bit more, as it came up in another thread recently. Many devices with electronic components are powered on and off daily, some with extreme temperatures, such as LED lights, an oven, a central heating boiler, etc. 

Is the ‘official’ line from Sonos, that Sonos devices should not be powered off, as this will shorten their life span?

I am sure that during the many man years of Sonos R&D, QA testing, quality electronic components, etc the Sonos devices must be capable of power cycled a couple of times a day, and still last a ‘decade’ or so?

You keep asking as if this is a question of quality control or testing, when it is simply reality.  A piece of electronics will go through a thermal cycle each time it is powered off/on, and that thermal cycle adds more wear and tear to electronics than leaving the electronics in a single thermal state.  This is simple physics at work and it happens whether the unit is built like a tank to withstand the thermal cycle or a thermal cycle could cripple the unit after a single reboot. 

Bottom line, no matter how well the unit is built, the thermal cycle happens, and it is more wear and tear on the unit than if you didn’t thermal cycle it.  This has no bearing on the actual length of the Mean Time Between Failure, except to say the MTBF will be lower than if one didn’t keep turning it on and off.

In 1977 I purchased the Quad combination 33-303 (pre-amp and amplifier). The documentation I received states that the units should always be on power and never to be powered down. The reason for this was to prevent any damaging of the circuitry, just like @Corry P and @jgatie wrote.

Since then I moved house 3 times and I can recall 1 power outing due to problems at the power plant. So since 1977 the system has been powered down 4 times. After 46 years of use on a daily basis I'm still using the system daily, the last couple of years with my Sonos Port. It hasn't let me down once, in my mind thanks to the guidelines of the Quad Company to never power it down.

Hi @craigski 

Thank you for coming back to me with this, as I did perhaps pick an unfortunate choice of words there. Rather than “will likely shorten”, I should have said “could conceivably shorten”.

All devices will heat up as they are powered on, and cool down as they are powered off. The resulting swings of temperature can, conceivably, induce stress on the internal components (or to their solder joints) as they expand and contract. On many individual units (not models), this may not have an appreciable effect - likely a majority of devices. On others, over time, it might. Considering that if such a fault were to occur it would likely be long after the manufacturer’s guarantee has expired, it’s better to minimise the amount of stress the components are subjected to from the start in order to maximise the possible time a device - any device - will last.

I think the examples of ovens and boilers are poor ones for fair comparison. My boiler has barely been turned off since it was installed (only when it stopped working for a week did I turn it off) several years ago, and the only part of an oven that is electronic is it’s clock - arguably a separate part, and certainly not the part that heats up! In addition, both devices are specifically designed to handle extreme temperature swings.

LEDs have an average life span - the longer they are on, the quicker they will fail (or fade). Although they are electronic - due to them being diodes - they are somewhat simpler than computers (Sonos devices are computers), so I again feel they are an unfair comparison.

The main (and official) purpose of leaving Sonos components on permanently, however, is so that they can respond to the app in a timely manner, or at all.

Of course, but to catch every single unit that will not do so before it leaves the factory would increase the cost of every unit made to the extent that they would no longer be as suitable for the home market, and what need does the commercial market have for WiFi speakers guaranteed to never break down? Such testing is generally left to industries such as medical, aviation, nuclear power and space - where it matters. Therefore, my advice to customers to leave the units powered on remains.

I hope this answers your concerns.

The reason I ask again, is I can’t find an official answer from Sonos. Of course I understand that power cycling will cause a change in temperature, and the thermo cycles will increase risk of failure due to different components/materials expanding/contracting at different rates. However, is it possible to design electronic devices that are more resilient to power cycles, shouldn’t fail within the designed lifespan. 

A car radio/entertainment system would be power cycled multiple times per day. It would possibly be subject to more extreme temperatures, as the car is outside. Yes it may fail earlier than if the radio was left on 24/7, but the car radio would be designed (and tested?) to be power cycled many times.

I would hope that Sonos do test power cycles on their speakers before releasing, hence my question about testing. Corry’s statement above is too vague/generic (for me).

edit: I think we posted at same time, I was referring to Corry previous post. I’m not suggesting each and every speaker is tested, more that the design testing process includes power cycling.

Without any inside knowledge, I would assume not as much care was given to rigorously testing the power cycling of something which is meant to be kept in an on state than something like a car radio, nor should there be.  I’m sure Sonos did enough to have a MTBF which is acceptable without pricing themselves out of the market.  And the fact remains, the manufacturer recommends keeping them on.  If you choose to ignore that recommendation, that is your choice.  

And how about  my 46 years of experience? Don't you think that's convincing?

It’s not very practical to test a design and then specify an exact number of power cycles before a unit will fail. For the sake of discussion lets say the design goal is 10,000 cycles -- on average. A comprehensive test would require cycling dozens of units for 10,000 cycles. This would require a lot of time -- delaying bringing this design to market. But, exactly how should this test be executed? If your ambient temperature is xx degrees and the test ambience was yy degrees, the test data does not fully apply to your situation. Plus, ambient temperature can vary wildly. For example a player installed in a cabinet will almost certainly be operating at a higher ambient than a unit sitting on a countertop. Higher ambient implies fewer cycles. Plus, some cycles might be short. On a short cycle the unit may never reach the test ambient, hot or cold and the overall reliability is tightly tied to the cycle depth.

Each part used in a design has a rating of xxx temperature cycles of yy degrees. Note that this is the temperature of that component operating in isolation, not considering any nearby components that may cause local heating. Given enough time, test resources, and designer skill, average product lifespan can be predicted. This design process is a monster project, justified only for huge volume or high value, critical  function products. Military, industrial, and medical products have been through additional testing and design considerations. While they are typically more reliable than consumer products, they will be several times more expensive for similar function and will likely not be on the leading edge of technology because of the additional time required for design and testing.

Don’t expect the marketing department to quote the expected number of power cycles. First, because the marketing department cannot predict your ambient situation, second because the failure predictions are only averages and there would be potential legal issues if the unit failed after only 9,999 cycles. Finally, conservative companies might specify only 5,000 cycles and be ridiculed by more aggressive competition.

The big picture is that there are multiple “failure clocks” ticking in each unit. Some clocks are simply time since the part was manufactured, others are tied to the depth of temperature cycles. It doesn’t seem fair, but some clocks run faster if the part is sitting in a drawer and not used. One particularly significant clock is the surge rating. As a power supply starts-up there is a momentary current surge for a couple components. This is usually their dominate failure clock and it is very significant. I once worked in a computer center where the mainframe computer was rated for a couple dozen power cycles. One would never deliberately power down this unit, but power grid outages would eventually cause the power supply to fail -- resulting in significant down time and a very expensive repair.

Conservative designs, such as beynym’s QUAD amplifier, run cooler and therefore expose internal parts to shallow temperature cycles resulting in a longer product life. This is a more expensive design. Here is a case where you get what you pay for.

A separate discussion is the build-operation-disposal environmental costs of a product.

Thank you all for responses, I do 100% get what you are saying, I do get the science. We can all find examples of devices that should not be power cycled (main frame computer), but also those that can be power cycled (Personal Computer with internal hard drive). 

Yes, the official line from Sonos its its ‘designed’ to be on all the time, as features such as instant ON, Voice control, Alarms, SonosNet will not work without power

Yes, power cycling any electronic device will increase the risk of the device failing.

I’m know I’m not going to get an answer, but I would hope that most Sonos devices should last a ‘decade’ even if they are power cycled once per day.

In general, product failure rate will be higher in the first few weeks (some on day one) than in the next decade or so, then failure rates will begin to creep up. There will be evolving failure signatures based on age. Many of the 2005 SONOS units are still operating. 

At one point SONOS claimed that over 90% of the units ever sold were operational. This was a misleading statement because the company has grown significantly. At this point I suspect that all of the units sold in 2005 could fail and have an insignificant impact on that percentage.

Maybe that claim was in 2015, and can be adjusted slightly to still be true today ‘90% of the units sold in last decade are operational’ ?