Maximum allowed output power ESB

[TinyRick]

Hi Nordic folks,

We are developing a product using nRF52832 with a Power Amplifier (RFX2401C) for extended range. For communication we are using ESB with some modifications to allow the use of a PA. The PA will add +22dBm to the output power.

We plan to release the product in Europe and USA.

Do anyone have a summary of what we will have to do to comply with the regulations in these countries? I am sure that someone has had the same kind of problem before here.

I have read a bit of the regulations for devices operating in the 2.4GHz band in Europe (EN 300 328), but as most of these documents this can be hard to read sometimes. Therefore we would like to know if someone on this forum has some tips.

[bjorn-spockeli]

Hi Rick,

As far as I know, you're allowed to transmit up-to +30dBm in the US as long as the bandwidth at 6dB is larger than 500kHz.

In Europe there are several regulations when transmitting above 10dBm, described in ETSI EN 300 328 Chapter 4.3.2. The most "problematic" part is 4.3.2.3 Power Spectral Density, which is limited to to 10dBm per MHz. Thus, with a 2MHz bandwidth the maximum output power of 13dBm, transmitting above this limit will require you to hop between frequencies (minimum of 15 channels)

In addition there is a requirement to "Listen Before Talk" (chapter 4.3.2.6.2.2), i.e. you need to check whether the channel is used or not before transmitting as well as requirements of the Medium Utilization, i.e. duty cycle.

Best regards

Bjørn

[AmbystomaLabs]

You should clarify this would only be true for non-bluetooth applications. So, it needs to be proprietary protocol. There is no bluetooth spec allowing more than 20dBm. And, at 20dBm bluetooth there are requirements for monitoring requests for power reduction.

As you pointed out even with proprietary protocol the radio would likely fail PPSD (peak power spectral density) for most regulatory bodies. Further, ETSI have recently made their SAR requirements more restrictive and a high power device would have problems with SAR.

And we haven't even begun discussing the problems of band edge compliance with a 52832 with nearly 20dB of gain on it.

TinyRick should investigate this thoroughly before proceeding.

[bjorn-spockeli]

@AmbystomaLabs: Yes, I should have stated that, I just reckoned it wasnt necessary since we're talking about ESB. Yes, I strongly recommend contacting a testing facility/lab regarding which parts of the regulation your design will have to comply to.

[TinyRick]

Thank you for the answers. I will contact a testing facility/lab.

Could you elaborate how you got the maximum of 13dBm with a 2MHz bandwidth?

Is it possible to listen before talk with the nRF52832? Won't the radio filter out all that does not match the configured addresses?

[AmbystomaLabs]

LBT refers to clear channel assessment and has nothing to do with protocol. The radio simply listens to see if there is RF energy on the channel before it tries to transmit. It has been part of 802.11 wifi since its beginning though not as common in other protocols. It works to mitigate lost packets from collisions (two devices transmitting at the same time). Likely etsi chose to write it in as a requirement because they are more of a standards oriented organization and write rules for the best possible user experience. They are very different from the fcc who really don't concern themselves with how a band is being utilized as long as you are not imparing licensed and government bands.