Hi all.

I'm currently working with a Murata SMPS in a project. http://power.murata.com/data/power/oki-78sr.pdf.

What I've found is my ADC values aren't as stable as the values I was getting with the last power supply I used. I also seem to have additional noise that the VR sensor part of my circuit is picking up. I did notice the old converter I was using had two chokes on it plus a few capacitors... which lead me to wondering if additional filtering may be the key. The stability seems to get worse when things start moving.

I'm thinking the circuit board may be partially to blame and I'm looking to rework the incoming power part of the circuit already as I have identified a few more places that may be causing problems. The circuit with the ADC on the other hand was not at all changed and bench testing shows it to be accurate and stable.

The datasheet of the converter lists the switching frequency at 500 kHz.

Is there a general set of choke and capacitor values that would work for variable frequencies on the input side?

Will a choke and capacitor help on the output?

 

Filtering a SMPS is somewhat trial and error.
I would try a choke and capacitor on the output with possibly a common-mode choke connected to both the output and ground between the supply and your circuit.
You might start with a few hundred microhenries of inductance and a few microfarads ceramic capacitor.
The common-mode choke should have as high an inductance as practical that will carry the power supply current.

 

In addition to the above, make sure that all your MCU's power pins are properly decoupled, and especially make sure your ADC's VREF inputs are very clean. Are you using an external voltage reference for your ADC, or are you using the internal Vdd/Vss connections for your reference? I've always gotten better results with an external VREF.

 

Filtering a SMPS is somewhat trial and error.
I would try a choke and capacitor on the output with possibly a common-mode choke connected to both the output and ground between the supply and your circuit.
You might start with a few hundred microhenries of inductance and a few microfarads ceramic capacitor.
The common-mode choke should have as high an inductance as practical that will carry the power supply current.

I'll have to admit I've never worked with chokes and capacitors other than what datasheets suggest for their products. If I understand the common mode thing it would be connecting one of the halves inline with the postive output of my converter and the other half inline with the ground right? Trial and error is getting to be not fun anymore, but it's been a good teacher.

I poked around on what's available on digikey... this one seems to be the highest inductance I can practicaly work with https://www.digikey.com/product-detail/en/kemet/SCF-03-650/399-10777-ND/4290843 if I understand everything so far correctly.

In addition to the above, make sure that all your MCU's power pins are properly decoupled, and especially make sure your ADC's VREF inputs are very clean. Are you using an external voltage reference for your ADC, or are you using the internal Vdd/Vss connections for your reference? I've always gotten better results with an external VREF.

The ADC I'm using is what is built into a pic24f32ka302. I'm actually using the built in voltage regulator for VREF (4.096v). I probably need to go back and look through the datasheet again for capacitor suggestions. Maybe the old converter worked good enough to hide problems I didn't know I had.

 

I poked around on what's available on digikey... this one seems to be the highest inductance I can practicaly work with https://www.digikey.com/product-detail/en/kemet/SCF-03-650/399-10777-ND/4290843 if I understand everything so far correctly.

That should work.