2
\$\begingroup\$

Making a project using an ATMega328p and an RFM69W 433mhz radio module. I want to run the AVR at 3.3v to make interfacing with the RFM69W (which runs at 3.3v) possible without bidirectional logic level translation. My project has 8 buttons and a PS2-style analog stick. I'm considering using a switching boost converter to supply a steady 3.3v even as my coin cell batteries fade.

I know that the noise from switching regulators can negatively impact things such as radio transmission and reception. A linear regulator is obviously an option, but the quiescent current is typically much higher and the efficiency much lower for these.

Is the interference from these switching regulators noticeable? I don't need a lot of output current, and was planning on sticking to a smaller regulator - for example, the TPS61221 - http://www.ti.com/product/tps61221/technicaldocuments?HQS=TI-null-null-octopart-df-pf-null-wwe

\$\endgroup\$
1
  • \$\begingroup\$ This is a very broad question. There's thousands of switching boost converters, and how good they perform EMC-wise is as much about PCB layout as the circuit itself. It isn't ideal to supply radio from switch regulators, but often you don't have a choice. The single-most important aspect is to pick a part with a switching frequency which is not similar to the radio IF. \$\endgroup\$
    – Lundin
    Feb 27, 2018 at 8:38

2 Answers 2

Reset to default
3
\$\begingroup\$

Assuming you do a halfway decent PCB layout, following the layout recommendations in the DC-DC converter datasheet, and use a ground plane to avoid nasty loops, you should be fine. Make sure you follow all the decoupling recommendations for the radio module.

Also pay attention to the switching frequency of your DC-DC converter (fsw) as compared with the frequency f0 and bandwidth BW of your RF signal. You will most likely see noise spikes at f0 +/- fsw. If the BW of you signal is wide enough that those spike occur within your signal, you may have increased risk of interference.

Unfortunately, there's just no way to calculate these things. My experience is that you will probably be ok, but there's no way to say for sure other than trying it out.

Also - From the datasheet linked, section 3.1:

"...to ensure a correct operation of the built-in voltage regulators"

As implied by the above quote, it's possible that the radio module already uses linear regulators on the most sensitive parts of its circuitry, which would help you out a lot.

\$\endgroup\$
2
\$\begingroup\$

Is the interference from these switching regulators noticeable?

The radiated emissions of the switching regulator highly depend on the PCB design, circuit design and, if any, outer case/shield. For example, a long trace or a "loop" may act as a good antenna. If this "antenna" carries relatively high currents then it can radiate unwanted noise at a frequency of even 100 times the switching frequency.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.