We have mutiple same-sized, same-tech RFID Tags (Mifare Classic), in the Field we have discoverd that some tags are less readable than others. For example, the distance is smaller or it takes more read operations to sucessfully read the tag.

The current test station only checks if the tag is readable and then marks as "good". Since there are different readers in the field (tolerance) the test method is not ideal.

I would prefer a test method that gives a value, rather than a binary result (good/bad). For example, the tag should say "10%, or 45%" of a known good reference.

The test method should be as easy as possible to operate by the tester. For example the tester should place the tag in the reader and press start.

Any ideas how to achieve this?

  • 1
    \$\begingroup\$ Sure, there are options. The first thing you need to do is create a holder with multiple places, each place is marked, place 1 = 10cm, place 2 = 20cm and so on. The reader must be at a static location. An even easier solution is to have 1 place, a static reader and the machine to just move the holder closer to the reader. You can adjust the distance as you want. \$\endgroup\$
    – CFCBazar
    Oct 7, 2020 at 7:33
  • \$\begingroup\$ I would do what CFCBazar said, and set up the test jig to move the reader closer to the tag until it gets a read. The distance where it succeeds would be your output value. It's not fancy, but it's simple and it does a good job of reflecting real-world performance. \$\endgroup\$
    – Drew
    Oct 7, 2020 at 7:39
  • \$\begingroup\$ You need a test reader that can be coerced into using a range of test frequencies to mimic the variations in test frequencies used by the readers in the field. This is probably where the main problem lies. \$\endgroup\$
    – Andy aka
    Oct 7, 2020 at 8:04
  • \$\begingroup\$ @Andyaka there are many other variables besides f that determine BER \$\endgroup\$ Oct 7, 2020 at 17:39
  • \$\begingroup\$ I wasn't making a complete answer Tony. I was pointing out that distance isn't the only parameter to be tested. \$\endgroup\$
    – Andy aka
    Oct 7, 2020 at 17:40

1 Answer 1


I have never tested cards but I have designed BERT testers for 2 way traffic. In this case the message error rate or transaction error rate margin for any card and reader is what you want. But to define margin, you must know the specs. This includes the errors for Xtal carrier frequency , data clock f error, and bit jitter in each direction for biphase enc/decoder.

1st you must have good specs for an error margin and the range for f (ppm) distance in each direction.

My approach would be test and measure:

1). amplitude path loss margin with switchable losses in each direction 2) frequency error & margin 3) Bit shift margin by injecting random jitter in each direction using PRSG controlled delay taps for early, nominal and late by some tester to measure phase margin 4) go/no go test with injected jitter and attenuation final test

The PRSG pseudo random sequence generator uses a linear XOR sum feedback from a shift register to create a random pattern using the binary data converted to a random tertiary value -1,0,+1 times some delay time in us which is a percentage of the ideal 25% margin of a simple biphase decoder.


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.