# Power supply filtering RCL-filter vs alternatives (for LF-RFID driver)

simulate this circuit – Schematic created using CircuitLab

This schematic is the manufacturers dev-board design for the RFID-IC power Supply, using a large power inductor at 4A I_RMS and 22uH.

Calculating this design as an RLC filter, yields fc~2kHz (low-pass). Since for this design the power supply noise is superimposed on the antenna signal, any noise in the range of up to 20kHz (per datasheet) is critical for high read sensitivity. The RFID-tag modulates the Antenna field at ~4kHz, which to me seems the most critical frequency range.

1. The current rating of the inductor seems excessive, the antenna driver has a maximum at ~250mA which is well below the 4A rating of the inductor. Can a smaller inductor be chosen? I think the I_RMS rating is only relevant for maximum power handling and doesn't take into account effects of saturation. As far as I understand, the filtering properties will be lost when the inductor reaches saturation (however not very clear on this, if someone can recommend some literature I'd be grateful). The ripple on the input is in the range of +-150mV.

2. Is there a better/smaller/more elegant way to achieve comparable or better filtering between, let's say 200-20kHz?

• Don't forget in-rush currents.
– jonk
Feb 12, 2020 at 17:44
• Good point, this will be added on the next iteration Feb 21, 2020 at 10:35

the antenna driver has a maximum at ~250mA which is well below the 4A rating of the inductor. Can a smaller inductor be chosen?

A smaller inductor will almost certainly have a higher ESR. This will cause a higher DC voltage drop through the filter.

If you can accept that limitation, you can probably use a smaller inductor.

You should also consider whether that Schottky diode is there for a reason or not. If there is some expected reverse polarity event for it to protect the load against, you will need to be sure the inductor won't be blown out when that event occurs.

I think the I_RMS rating is only relevant for maximum power handling and doesn't take into account effects of saturation.

Depending on the inductor design, the current limitation might be due to saturation, or it might be due to resistive loss. A good datasheet will tell you which. On the other hand, not all magnetics vendors provide highly detailed datasheets.

Is there a better/smaller/more elegant way to achieve comparable or better filtering between, let's say 200-20kHz?

You might consider using two (possibly lower-valued) inductors with capacitors in between to make a two-section filter instead of this one-section filter. That will get you 40 dB/decade roll-off rather than 20-dB per decade. Whether the overall performance is better for your application (and whether it's worth the increased cost and/or board area) is up to you to evaluate.

• Thanks for the feedback, I made a mistake on the diode, I meant Zener not Schottky, thus the 5.6V rating. I imagine this will act as an over-voltage protection on the output. I don't have any reason to believe the driver will feed back the higher antenna voltage on the inputs. Feb 13, 2020 at 10:31