Author Topic: Selling in-house assembled vs. manufactured product  (Read 1546 times)

0 Members and 1 Guest are viewing this topic.

Offline bntlTopic starter

  • Newbie
  • Posts: 2
  • Country: ca
Selling in-house assembled vs. manufactured product
« on: January 14, 2025, 06:22:50 am »
Hi everyone,

Quite new here, but I have visited this forum a lot! I have read some articles both in this category and under Technical, but had a question regarding selling electronics (consumer products).

Basically, we're starting a crowdfunding soon and had some question (team of three, based in Canada).

We're working on a smart wearable (thin bracelet; strap, smart device - pcb with a few sensors + ble module (nrf52) enclosed in alloy and epoxy - see through - similar to fitbit).

If we get enough backers, but not a lot, I was wondering if it would be "legal" to assemble and package the product completely in-house and ship it to the backers.
Our goal is to get enough funding to also get FCC certified, and if we make enough, have the product assembled with a manufacturer.

I have already found a manufacturer, and can also completely assemble it in-house (we have over 20 bracelet being used at the moment (friends, family, relatives that were happy to test it and use it)).

If done in-house:
- we'll be using JLCPCB (and/or PCBWay).
- either assembled by JLCPCB/PCBWay, or self-assembly, with parts from DigiKey/LCSC.

For self assembled pcb, we have a 3d printer converted to PnP - not too fast but good enough.

The "final package" consists of:
- the bracelet
- charger
- usb-c for the charger

Now, this is our first time doing a crowdfunding. We have sold quite a bit of "dev kits" in the past, but they were never finished products.

Please do let me know if we're going in the wrong direction - hopefully not!

Thanks,
Bent
« Last Edit: January 14, 2025, 06:38:27 am by bntl »
 

Online Kean

  • Supporter
  • ****
  • Posts: 2544
  • Country: au
  • Embedded systems & IT consultant
    • Kean Electronics
Re: Selling in-house assembled vs. manufactured product
« Reply #1 on: January 14, 2025, 07:16:01 am »
That all sounds fine except for one thing... as you will be selling a finished product, FCC is not optional if you are shipping to customers in the USA (or some other locations who accept FCC reports).

Also, if you are shipping into other countries, you likely need to get equivalent testing done - e.g. CE for Europe, ISED/IC for Canada, RCM for AU/NZ, etc.

And being a crowd funded project doesn't really make any difference, nor would your location or business type, or the contract manufacturer.

You could kind of get away with it in the past when selling a dev kit, but not officially - and not to EU any more.  Germany being particularly strict at import.  That isn't to say people don't do it, but it isn't legal and you will very likely encounter delivery issues with at least some end users, and maybe worse.

If you are using a pre-certified nRF52 module (as opposed to just the silicon) then you should be able to do the testing relatively cheaply, and each additional certification can sometimes piggy back off the prior testing.  Or at least it generally gets easier and the process becomes more familiar - especially if you do CE first!

BTW, do you really need to supply a charger if it is USB-C?  Most people already have more chargers than they will ever need, plus leaving it would will save you money on compliance tests and shipping.  Just make sure you test the product with a range of off the shelf chargers.
 
The following users thanked this post: thm_w, mskeete

Offline bntlTopic starter

  • Newbie
  • Posts: 2
  • Country: ca
Re: Selling in-house assembled vs. manufactured product
« Reply #2 on: January 14, 2025, 01:25:52 pm »
Hi Kean,

Thanks for the informative reply, that was very helpful!

I was aware FCC, CE and ISED, but not RCM for AU/NZ. I am assuming a lab can test for multiple certifications.
I think before even starting the crowdfunding, we should get more information related to these certifications from either a lab or original sources (gov sites).

So if I understand correctly, as long as we're able to get certified, manufacturing these in-house shouldn't be an issue.

We attempted to design the pcb with a pre-certified module, however, we couldn't find something that fits our design: due to the shields and square design (most prominent). Our current pcb is approx. 11mm in width.

We did thought about not shipping a USB-C pre-packaged, I see that can influence the tests, we may as well not ship one - we'll reiterate on this for sure!

Thanks again!
 

Online Kean

  • Supporter
  • ****
  • Posts: 2544
  • Country: au
  • Embedded systems & IT consultant
    • Kean Electronics
Re: Selling in-house assembled vs. manufactured product
« Reply #3 on: January 14, 2025, 03:45:04 pm »
OK, be aware that if you are not using a pre-certified module then you will need to go through an extra set of radio testing.  I've not done it myself, but I understand it will add quite a substantial cost to the certification process (several times more than the usual emissions/immunity tests).  I suspect Nordic can help with some low-level radio control firmware needed for the testing.  Espressif have something similar for their ESP32 chips.

Yes, a lab can help with these tests - but not all labs are accredited for all tests, so they sometime outsource or just shrug their shoulders.  BTW, I think CE is officially called RED now.  Like C-Tick is now called RCM, and IC is now called ISED, etc.

Another thing is if you are using Bluetooth technologies, you probably need to join the Bluetooth SIG as a (free) member and then pay for certification of your product to use the BT patents & trademarks.  This is something I have done (membership), and my company has been allocated a unique identifier for when I was working with Bluetooth Mesh.  See https://www.bluetooth.com/develop-with-bluetooth/join/
It is interesting to scroll through the assigned company identifiers, particularly noting companies you may know of who only joined relatively recently... https://bitbucket.org/bluetooth-SIG/public/src/main/assigned_numbers/

Also, I just thought of something else.  As this is a wearable, you may have to do SAR testing.  Another significant cost if it is needed.  And you really wouldn't expect that is as there are plenty of other nRF52 based wearable designs, and any mobile phone transmits at much higher power but yet must have passed.  I don't know if you can use that reasoning, so you may need to have the tests done as proof of meeting the requirements.  Again, I've managed to avoid projects needing this (other than some concept prototypes that didn't go further).

Here are some references - but I suggest you find a local lab/consultant to help you:
https://www.rfexposurelab.com/sar-requirements-for-wearable-technology/
https://www.ul.com/services/wearable-technology-testing-and-certification
https://www.tuvsud.com/en-us/industries/consumer-products-and-retail/wearable-devices
https://www.reddit.com/r/rfelectronics/comments/1ednpqy/esp32_wearable_sar_test/

Once you work your way through that minefield, I don't think there is any real restrictions on who does your manufacturing - after all you aren't making medical/automotive/aerospace products???  Up to you if you want to use someone with ISO9000 or similar for QC processes.  But you should also note that the certification is only valid as long as the finished product doesn't vary in any "significant" way from what was certified.  If you make revisions (to schematic, PCB, BOM, even the housing design) then you need to analyse the potential effects to determine if the change is significant and requires re-testing.  Fun stuff.

Maybe check out the updates from the SlimeVR project as they talked about a bunch of these topics, and they may be willing to chat with you more if you reach out to them.  I can't recall if they did SAR.
https://www.crowdsupply.com/slimevr/slimevr-full-body-tracker/updates/

In the end, this is all about risk management.  It helps with paperwork problems with shipping (one of the earliest and most visible issues you could encounter), and provides protection from potential government enforcement (or private action) if you have done something horrendously wrong in your design or paperwork.

I've likely forgotten something else... and maybe even misstated something unintentionally.  The above is only worth what you paid me for it (nothing)... so please do you own research and get professional advice.
 
The following users thanked this post: mskeete


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf