EEVblog Electronics Community Forum
Electronics => RF, Microwave, Ham Radio => Topic started by: madhu.wesly01 on June 27, 2018, 04:31:38 am
-
Hello,
I am using TI's Switching power supply for my RF transceiver, but when I observed the packets at Base station there is a lot of packet duplication(Duplicate packet with different strengths(RSSI and SNR) but at the same time and at different frequencies). Please let me know what might be the reason?? Is power supply the issue since since power supplies are not recommended for RF??
Thanks in advance.
-
With switching supplies a LOT of care must be taken around RF.
With a badly designed PSU you can not only interfere with your own radio but even interfere with other radios half way across the room.
Its best to stick with linear regulators for RF, but when a switching regulator is needed it needs to be well designed and filtering placed on its outputs and inputs to keep the noise from getting out, sometimes it might be nececery to put it into a shielded can. Best tool for tracking down such interference is a spectrum analyzer.
-
I suspect what you might be seeing is the supply switching noise acting as a second LO at some point. You can do two things:
If they are (far enough) outside your band of interest you can just filter them out (sometimes the antenna itself can be selective enough to do this, even).
If not, you need to better plan your frequency stuff.
It all comes down to frequency planning. Know what frequencies you are expecting switching noise at. Are the frequencies (and their harmonics) in your RF/IF band? Then you have to be really, really careful. If not, you can relax the filtering constraints.
Keep in mind also that if you have a power supply though the form of a bias-T, it will work both ways - keep your RF from leaking onto your DC line, but also stop your DC noise getting into the RF path easily.
What I believe is also common is a switching regulator with a linear post regulator. What you do there is use the switching regulator to give you the majority of your DC/DC conversion, and then use a LDO regulator with just a few hundred millivolt drop to smooth everything out and improve the high-frequency loop response.
-
Repeat the same measurements with a linear supply instead of the switching supply you have. If the problems are gone, you got the culprit.
I had the similar RFI problem, and I killed a lot of "wall warts" and replaced them by retro-style linear power supplies, like here:
https://electronicprojectsforfun.wordpress.com/2018/01/31/kill-your-wall-wart/
Now the fuzz is gone.
-
Could be your SMPS, but not necessarily, e.g. bad PCB layout and/or bad partitioning. Since you're asking for help, maybe you could provide more information: what's the SMPS you are using? can you show us your PCB layout? etc.