I'm doing an rfid project that requires a read range of at least 5 inches or about 13cm. I'm using a nano with a usb port, a rfid reader from priority 1 design and a large antenna. The problem I'm running in to is that when I power the project from the usb port I get a pretty decent read range of 11-13cm. But when I power the project from a DC power supply the read range drops considerably.

After a lot of testing with various power supplies I believe that all my power adaptors are dirty enough to cause rf interference and reduce the range. Powering the project with a 9v battery and a 5v power regulator gives me really good range: 18cm with the largest antenna. But I need plug in power.

So how can I power this project in a way that gives me ultra clean power?

  1. I tried that same 5v regulator with a 5v adaptor and still had short read range.

  2. Someone suggested adding a smoothing capacitor between V+ and Gnd but I wouldn't know what size. I'm not an EE - I'm just a coder with a soldering iron.

  3. Is there a better power supply I can get?

[EDIT] Also just FYI, when I connect even only the ground wire the the computers USB port I get a much farther read range. So..

  1. Could it be a case of needing to ground the circut better as the switching power supplies only have 2 prongs and probably no real "ground". (Or am I making stuff up?)

[EDIT 2] What about a rechargable situation where my wall adaptor keeps a litium charged and the litium supplies clean power to the circuit?

Also I now know that I can charge my whole circuit with anything from 5 to 12 volts. In fact it would be better to have more in the 7 to 12 range than strictly 5v.

[UPDATE 3] Since this project is in my wall I was able to ground essentially to "earth" and that cleared up the noise. See this question as well: https://electronics.stackexchange.com/questions/230024/grounding-dc-project-to-clean-rf-noise

  • You can't use a 5V regulator with a 5V adaptor.. You can experiment with 1) a USB wall adapter 2) a >7V wall adapter and a linear 5V regulator (e.g. 7805, with the two 100nF capacitors) 3) a >6V wall adapter and a switching step-down regulator, set to 5V 4) any wall adapter (maybe >3V) and a switching buck-boost regulator. I think that with the USB wall adapter you will already solve your problems, then I'd test a switching regulator (step downs are better, I think) and the last choice is a linear regulator. you can find switching regulator modules for a few dollars on ebay, if you need them
    – frarugi87
    Commented Apr 21, 2016 at 14:00
  • I did experiment with the USB wall adaptor and got the same short read ranges. This is the regulator that I have today: robotshop.com/en/… I just tried that one with a 12v power supply and it won't read at all. And the only thing I'm changing in these experiments is the power source. I ordered this one which will be here tomorrow: amazon.com/SMAKN®-LM7805-3-Terminal-Voltage-Stabilizer/dp/…
    – badweasel
    Commented Apr 21, 2016 at 20:24
  • Strange.. No reads at all? Did you check the ouitput voltage with a multimeter? Anyway your "new" board is a linear regulator, so use a 7V+ voltage source (12V should be fine) and be aware that you won't be able to draw much current from it..
    – frarugi87
    Commented Apr 22, 2016 at 9:52
  • Yeah it essentially dropped the read range to nothing. Too much rf interference I suspect. Of course it's hard to tell what's happening when I can't see a console output. I do have it blinking out some return values as a type of "morse code". So I can tell that it's taking to the rfid reader and read that the measured antenna frequency is near 125khz.
    – badweasel
    Commented Apr 22, 2016 at 11:04
  • @frarugi87 The new stabilizer didn't help. I also realized that I can power the whole thing (arduino and rfid reader) from 9-12v. With a 9v battery my range is optimal. With a 12v switched power adaptor direct I get very short range. If I then connect the ground of my circuit to the ground wire at the wall my range goes up again to closer to the 9v battery range (about 1-2cm shorter). I want to try the 2 100nF caps but where do I put them? In series between vcc and gnd? Or in parallel? Can you write this up as an answer?
    – badweasel
    Commented Apr 22, 2016 at 22:58

2 Answers 2


This is what I use, although admittedly not in your application. I run it from a 12V wall adaptor and adjust the output voltage to suit my requirements. The input should ideally be at least a couple of volts above the output(depends on current and temperature at 25degC and 20mA it's > 1.6V and 1.5A at the same temp it's > 2.7V).

Circuit Diagram

Another alternative if you want to really go to town would be to repurpose an old computer ATX or similar PSU. There are plenty of guides out there and it's a straightforward job.

  • Please specify that the input voltage should be at least X volts, where X is 5V + the dropout voltage of the LM317 (I don't remember how much it was)
    – frarugi87
    Commented Apr 22, 2016 at 15:14
  • My issue appears to be RF noise coming from the 12v power supply is messing up the RFID application. Something in the power supply is noisy enough that it makes the RFID not read at all. Where as when I use a 9v power supply everything works golden. What I'm looking for is a solution that will give me ultra clean power in the 5.5v -12v range without having to swap out 9v batteries every few months. Even if it were a rechargable situation where the 12v wall supply charged a litium that then gave clean power to my circuit.
    – badweasel
    Commented Apr 22, 2016 at 22:39

I think the problem is not the voltage but the magnetic field, so maybe better to use a faraday cage (a metal box) around the power supply and connect it to output DC ground.

  • Grounding it to actual ground cleaned up the power.
    – badweasel
    Commented Feb 5, 2017 at 18:59

Your Answer

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

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