Teardown Of Sonos And Amazon Smart Speakers Reveals Interesting Engineering Details

Taking things apart is always fun, and this What Cracking Open a Sonos One Tells Us About the Sonos IPO”>excellent writeup of a teardown of a Sonos and Amazon smart speaker by [Ben Einstein] shows what you can learn. [Ben] is a Venture Capitalist and engineer, so much of his write up focuses on what the devices say about how the company spends money. There are plenty of things to learn for hackers, though: he details how the Sonos One uses a PCI daughterboard for wireless communications, while the Amazon Echo has a programmable radio on the main board.

The advantage of the Sonos approach is that you can use the same daughterboard on multiple models: as long as each model has the PCI Express slot you can just plug the daughterboard into any of them. The Amazon approach is more expensive, but has one huge advantage: the radio can be reprogrammed to support other wireless standards remotely. This, [Ben] points out, shows how the two companies have a different approach: Sonos is trying to save money by making modules that can be used on different models, while Amazon has spent more on design and components to create a device that can handle current and new standards.

This is yet another clue that Amazon is thinking about Echo as a gateway to the home rather than a speaker with some new tricks….It is always tricky to estimate BOM cost without diligently researching each custom part and purchased component, but my suspicion is that despite the 25% lower price tag, the Echo Plus is about 15–20% more expensive than the more premium Sonos One.

[Ben] doesn’t think that Sonos is going to do well in the long run with Amazon as a competitor. I don’t know enough about this market to judge his argument on this, but it is well worth reading his article to see the neat engineering tricks and choices that both manufacturers have used. Plus, it’s always fun to see gadgets in pieces.

[Via Techcrunch]

24 thoughts on “Teardown Of Sonos And Amazon Smart Speakers Reveals Interesting Engineering Details

  1. Amazing technology like the long deprecated, 25 year old SMB1 protocol which was replaced with SMB2 in 2001, yet Sonos can’t seem to figure out how to use that instead.

    1. It’s far more likely not. VC’s don’t need to teardown the products they ‘support’. They can require delivery of a bom, code, and schematics.

      Bored evening, market research, pure fun to strestch out the kinks in a few neurons. All far more probable.

  2. From article:

    “The power supply (electronically dirty, analog circuitry) is on the same board as the communications, MCU, and audio output (digital, clean circuitry).”

    Sorry, but dirty analogue circuitry??? I stopped reading from there,

    1. I read the whole part and it does sound slightly like a hit piece.
      Putting all the electrically noisy parts on one board and putting the sensitive analog stuff away from it looks like a smart design choice. And the PSU seems to have good clearances, nichicon caps and in general a sane design.

      Amusingly on the Amazon Echo front the talk about “The EFR32MG12 has a configurable sub-GHz radio programmed for the Zigbee home network protocol, which is one of several future low-power, IoT home networking standards. This is yet another clue that Amazon is thinking about Echo as a gateway to the home, rather than a speaker with some new tricks.”
      Is false as the EFR32MG12P232F1024GM48 mcu from SiLabs only has a 2.4GHz, it’s not a dualband model.
      https://www.silabs.com/products/wireless/mesh-networking/efr32mg-mighty-gecko-zigbee-thread-soc/device.efr32mg12p232f1024gm48

      And in general Sonos is referred to as “Speaker Company” and “upstart” while Amazon gets referred to as “Innovative Tech Giant”.
      Which apparently uses needlessly expensive manufacturing processes.

      1. Your point about the radio not going to sub-GHz range is a little beside the point, it does support other standards which will be used in home networking.

        It’s curious to read that the comparison in between an ‘established’ voice controlled speaker and a ‘startup’ offering a voice controlled speaker. Sonos clearly already have a lot of expertise in connected devices. Their sound distribution among devices is far better than the Echo’s.

        1. Maybe a bit. Dunno if that particular model has licenses for other than BT.
          But yeah the hardware should do Thread, SigFox and ZigBee easily. And if that model cannot do it, some pin-compatible drop-in model will do.
          For massive volume massmarket stuff the Amazon model of not using modules and instead having everything on the pcb’s should offer cost savings with high enough volumes.
          But the Sonos pci-e card approach should offer faster time to market and less EMC testing.

          And yeah, I’d like to see a comparison between speaker a newer and older manufacturers as that’s more of a level playing field.

  3. Quote: This is yet another clue that Amazon is thinking about Echo as a gateway to the home rather than a speaker with some new tricks…

    That and the reprogrammability is why Echo should worry people. It is a “gateway to the home” that could be used by some nasty people to do some nasty things.

    1. Ahh yes, a lot of people seem to think that these gateways are only one way gateways but that is only because they are only concerned with the one direction that provides them with stuff. It is not their fault though, this is how we have been trained to thin, only to see the immediate problem in-front of us and not to think of the grander implications of the issue at hand.

  4. Now I’m interested in how the holes are created on that speaker grille. There’s no way they’re all drilled–CNC or otherwise. My guesstimation is around 1300 holes!

  5. Stein’s thesis that he can tell a company’s “trajectory” from a hardware teardown may or may not be true but he does a very confusing job of supporting that thesis. The bulk of his post simply shows that the Sonos speaker is better designed than the Amazon Speaker. The only relevant points I could find were:
    1. Amazon doesn’t care much about the cost of producing the device – which shows that the hardware is a means to an end rather than the end itself. Which will make it more difficult for Sonos to compete.
    2. Sonos isn’t using their own voice control technology – so they will not benefit from the other streams of revenue. And in the future Amazon could simply withdraw access to the technology from
    Sonos or restrict it’s functionality.

    The mixing of hardware teardown and analysis of the Sonos IPO made a very strange combo.

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.