Pre-certified Bluetooth module for mass production

[angust_uk]

Hi all

I'm working on a design which will be built in significant quantities (>10k) and which requires a Bluetooth interface for low bandwidth control data between the embedded device and a custom app (using Bluetooth SPP). There are many popular choices such as the common HC-05, but these do not seem to appear to be pre-certified.

Has anyone used pre-certified (in the US and EU) modules in their own design which come from a reliable source and are reasonably priced in quantity (<$2 is the target)?


Thanks!

[tszaboo]

Do you need CE or FCC or RED or MIL-STD or X or Y or Z certification?

[angust_uk]

Since it's a consumer device to be sold in the US and EU, FCC and CE certification is needed. Obviously RED is now applicable but I'm not entirely sure whether this affects modular approval rather than the whole device.

[Pinkus]

Mmmh- why?
You will have to certify your complete solution anyway. No matter if you are using anything pre-certified or not - it won't help you in any way (but peace of mind... though only temporarily).
So ... avoid the hassle with looking for already pre-certified stuff. Itwill not help you in any way, as at then end, only your final device matters. And believe me, it can fail easily even if build using only pre-certified modules.

[voltsandjolts]

Of course, but using pre-certified modules reduces the risk of EMC problems at a later stage.

[angust_uk]

Mmmh- why?
You will have to certify your complete solution anyway. No matter if you are using anything pre-certified or not - it won't help you in any way (but peace of mind... though only temporarily).
So ... avoid the hassle with looking for already pre-certified stuff. Itwill not help you in any way, as at then end, only your final device matters. And believe me, it can fail easily even if build using only pre-certified modules.

Not entirely - using a pre-certified module significantly reduces the cost of EMC testing and removes the need to carry out Bluetooth SIG's tests which are quite costly! Of course the device will have to be tests as a whole, but as stated by voltsandjolts the risks of failure are less if the module is pre-tested.

Anyway, back to the original subject - has anyone used anything that might fit the bill?

[mjs]

Is Bluetooth classic and SPP a must ? I moved to BLE some years ago and not coming back to the old world. There's some learning curve, but for my purposes BLE has been perfect except for one or two data streaming cases.

If BLE works for you, check out Cypress, Rigado and Silabs offering through a bigger distributor and you might hit your target price.

[Psi]

Tip for young players with regard to pre-certified Bluetooth modules.

There are two types of certification for bluetooth modules. I forget the exact terms used.

One kind is fully certified. You can use it and don't have to do any RF tests on your product. You just grab the cert documents from the manufacture and they will apply to your product using the module.
This kind of module will have a metal shielding can over the RF section on the module.
(NOTE: having a metal can doesn't mean that it is definitely certified)

The other kind has no metal can and will only be 'partially' certified.
You will need to pay to RF test and certify your product as a whole if you choose this kind of module.

It's easy to pick a module, read that it is 'certified' and assume you are all good.
Then later on find out it was only a limited cert and you will need to get RF tests done.

[ogden]

Not entirely - using a pre-certified module significantly reduces the cost of EMC testing and removes the need to carry out Bluetooth SIG's tests which are quite costly!

Exactly. It shall be noted that EMC tests like FCC Part 15 is not enough for BT radio. It needs all the RF tests to apply for FCC ID (in US). FCC part 15C certification (example) is totally different can of worms you wish you do not open. As far as I remember, module manufacturers suggest that designing and certifying own RF part is worth when volume exceeds >50k AND this considering that you have RF engineers and needed equipment for that. If not, then 3rd party RF design can be even more expensive requiring even bigger volume.

It's easy to pick a module, read that it is 'certified' and assume you are all good.

Very good point. First you never pick module without on-board antenna. For US you make sure that module received FCC ID, it is searchable in FCC database and there are test reports for module in question. Yes, manufacturers sometimes cheat by testing only one of many module versions. For EU you ask manufacturer to produce report from accredited lab and check that module passed:

EN 300328 - radio equipment testing of data transmission equipment operating in the 2.4 GHz ISM Band
EN 301489 - electromagnetic compatibility
EN 62479 - compliance to human exposure to electromagnetic fields

[angust_uk]

Many thanks for all the advice - appreciated! I think I've got enough to go on now.


Cheers!

[tszaboo]

Ah, yeah, and CE + CE != CE in theory.
In practice, if you use a pre-certified module, and you keep the power levels down, you could skip the entire RF testing (for CE, not FCC). It really depends on the equipment, and doing the testing anyway might be cheaper.
RED might even be skipped, it is applicable to higher power transmitters.
My point is, first step you need to identify which standards are applicable, needs testing, etc.
Write down on a paper, do a planning, get a signature from your boss, that these are the only tests that need to happen and only stuff I need to comply with. Do this as the first step of the project.

[Psi]

you never pick module without on-board antenna.

Yes, thanks. I forgot to mention that.

It will never have full certification without one though.

[ChristopherN]

Hi,

I would recommend using BLE as well, Classic support is not as good as it once was. That said, it will be very hard to get current modules for $2. I just got quotes for one project with 100k/year and we're still at around 3,50€ for a fully certified module with BT5 and some processing power from western manufacturers.

Two main ways for cost saving can be:

• Moving the application to the module, most modules have plenty of processing power and memory.
• Use something from Espressif (ESP8266 /ESP32).

If you want to use the Bluetooth logo you have to register with the SIG in some cases, there are rumors that certifying modules with 4.2 will get more expensive over time.

BR
Christopher

[SiliconWizard]

Mmmh- why?
You will have to certify your complete solution anyway. No matter if you are using anything pre-certified or not - it won't help you in any way (but peace of mind... though only temporarily).
So ... avoid the hassle with looking for already pre-certified stuff. Itwill not help you in any way, as at then end, only your final device matters. And believe me, it can fail easily even if build using only pre-certified modules.

Not entirely - using a pre-certified module significantly reduces the cost of EMC testing and removes the need to carry out Bluetooth SIG's tests which are quite costly! Of course the device will have to be tests as a whole, but as stated by voltsandjolts the risks of failure are less if the module is pre-tested.

Agree with this. Just don't expect it to be a picnic though, using a pre-certified module will indeed make things a bit easier, but the RED directive still is a PITA. I suggest thoroughly reading the part that explains the different use cases (type of device, RF modules vs directly integrated RF, modules that are permanently connected to the device, or modules that can be put/removed by the user - yes it makes a difference -  etc.)