RFM69 FCC Certification, Bandwidth, Frequency Hopping, etc

[mah115]

Hi everyone,
   I designed a device using the RFM69-915MHz Moteino M0, and I want to try and sell it, which means I'll need to get certified for FCC compliance.  I had a chat with the engineer at the test facility, and he said I'd need to know the bandwidth, and for narrow-band systems I may need to implement a frequency hopping scheme, etc.
   I'm not very familiar with RF engineering, but my understanding is that we're basically trying to limit power in each specific channel, i.e. spreading the power over a wide bandwidth is better than concentrating it in a narrow one.

I had a few questions:
1. How do I calculate the bandwidth of radio, and are there any knobs I can turn to adjust this?
2. Measuring the DIO0 interrupt pin with an oscilloscope (I'm using the older version of the RFM69 library that uses the physical pin instead of reading registers to indicate send complete), it takes about 0.8mS to send my packet, would I be able to be within the -1.25dB single channel limit if I dial back the transmit power a bit and limit my packet frequency?  Or does duty cycle not work that way?
3. Would it be easier to be within limits if I use the RMF95 radio due to its spread spectrum modulation scheme?

[Felix]

I'm not an RF engineer or expert either and I'm not intimately familiar with all the regions and limits and specific requirements, they vary by region and your brain will spin just looking at this list.
I will provide my .02 just here in case anything is useful.

my understanding is that we're basically trying to limit power in each specific channel, i.e. spreading the power over a wide bandwidth is better than concentrating it in a narrow one.

For compliance, yes.

1. How do I calculate the bandwidth of radio, and are there any knobs I can turn to adjust this?

Sorry for the lame answer here - such details are found in the datasheet, and the "knobs" are the registers. The library makes it a little easier by providing some function changing and other than that just access to writing to registers but you will have to experiment.

2. Measuring the DIO0 interrupt pin with an oscilloscope (I'm using the older version of the RFM69 library that uses the physical pin instead of reading registers to indicate send complete), it takes about 0.8mS to send my packet, would I be able to be within the -1.25dB single channel limit if I dial back the transmit power a bit and limit my packet frequency?  Or does duty cycle not work that way?

Sorry I don't know an answer here.

3. Would it be easier to be within limits if I use the RMF95 radio due to its spread spectrum modulation scheme?

Likely yes, take a look at TheThingsNetwork site for such guidelines - even LoRa cannot just transmit away freely, they have some pretty specific limits but they might be more geared towards EU where this network is dominant right now.