0
\$\begingroup\$

I am using a switch mode power supply RAC06-05SC to power up the microcontroller (ATMega16L) a wifi module (RN-131G) and a 5V relay. This power supply is already regulated (as stated in the datasheet) and it is regulated further down to 3.3V for the micro and the wifi module using a linear regulator.

I am told that switched mode power supplies generate a lot of noise, and I am trying to use this with a microcontroller which will use the ADC. My questions are given below: 1. Does a switched mode power supply affect radio frequencies of 2.4GHz? 2. Will the supply abruptly switch off the power? 3. Will it effect the functionality of UART? 4. Will it effect an external oscillator?

Or any thing else I should know?

EDIT

My relay schematic enter image description here

\$\endgroup\$
  • \$\begingroup\$ Um, what problems with the relay are you expecting your power supply to cause? \$\endgroup\$ – Nick Alexeev Jul 12 '12 at 4:53
  • \$\begingroup\$ The switch mode PS gives 5V output right and I am using a 5V relay. I thought that 5V from the supply might not be suitable for the relay as I am going to connect it directly to the it without regulation. I would be regulating the 3.3V for the micro and the wifi module but not 5V for the relay. So I was wondering if that is alright for the relay to work properly (i.e. without any regulation and with 100mV ripple from supply) \$\endgroup\$ – David Norman Jul 12 '12 at 5:25
3
\$\begingroup\$

In general you can assume that as long as the ripple is not too bad on a supply then it shouldn't cause an issue for functionality of an IC. In this case the datasheet gives a ripple of 100mV peak-to-peak, BUT you seem to imply that you're taking this voltage and further regulating it down. Passing a signal with a relatively small ripple and putting it through a linear regulator like a 7803 or an LDO will significantly reduce the ripple and cut back on switching noise passed through to the final circuit. So a ripple which is less than 5% of the expected voltage like this, being passed through a second line of regulation, should be fine for 95% of projects, even one using analog components like an ADC. Just make sure to have input and output filtering capacitors on the linear regulator, and not too much (2200μF is the limit for the power supply you gave) (maybe 47uF and 4.7uF (or 220uF and 22uF if you want it really smooth)).

Because the ripple is so small this supply would probably be fine for digital operations if you were using a 5V compatible chip and you weren't doing Analog work, but I digress.

So looking at the datasheet you provided, the operating frequency of the power supply is 132kHz. The chances that the noise from that is going to manage to produce harmonics in the 2.4GHz spectrum is pretty limited, especially at such small power outputs. Where you really have to start worrying about generated EMI is when you have isolating switching supplies, because these will pass power through the magnetic field, and the interface leaks EMI on most supplies I've used. Generally the higher-power the switching power supply the more noise it generates. I've had a 100Watt isolating switching power supply which put so much noise out I had 30V of noise on the frame of my robot, but that's way beyond the scope of this. I think you'll be fine with your usage!

Hope that helps!

\$\endgroup\$
  • \$\begingroup\$ that was very well explained thanks a lot for that. Another concern I have is my Relay. I am using a 5V relay and since the switch mode power supply gives output of 5V, I cannot regulate it to again give a 5V output. Will this supply work for my relay? I have added a relay schematic to my question to go with that. \$\endgroup\$ – David Norman Jul 12 '12 at 3:57
  • 1
    \$\begingroup\$ There should be absolutely no issues. Like I said, the only thing that will be affected by a ripple that small is going to be sensitive analog electronics. What you have there will be completely fine, you would need a really terrible supply to make that not work the way you were expecting it to! \$\endgroup\$ – Kit Scuzz Jul 12 '12 at 4:05
2
\$\begingroup\$

Regarding 1. A switching supply may or may not interfere with a radio module. I had a circuit where switching DC-DC converter was making a 900MHz module inoperable. If you see such problem, you can add more filtering in front of your radio module. You can also power the radio module from a separate linear regulator.

Regarding 3. It's not likely that a switching power supply would cause problems with UART.

\$\endgroup\$
  • \$\begingroup\$ it says in the datasheet "Output Ripple and Noise (20MHz BW)" what does this mean? \$\endgroup\$ – David Norman Jul 12 '12 at 3:41

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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