is there a difference between ST25R3911 and ST25R3911B?

SD.5 · ‎2021-04-19

We are moving from ST25R3911 to the latest version ST25R3911B with the firmware and the other circuitry remains the same. But we are seeing the reduction in operating distance for the cards.

Ulysses HERNIOSUS · ‎2021-05-12

Hi Xavier,

this procedure sounds right, I assume you delay sufficiently/wait for the termination interrupt in between?!

Any chance that you can do smith charts as in AN4914 for some trim values?

The AN4914 presents also that the similar algorithms can easily be implemented in SW (measuring 16 points and picking the best - I recommend the one with highest amplitude). Could this be an acceptable change which solves the issue for you?

I am suspecting a minor change of behavior of the internal algorithm (between ST25R3911 and ST25R3911B) when the target phase is not reachable, but would need to do more investigation here.

Regards, Ulysses

View solution in original post

Ulysses HERNIOSUS · ‎2021-04-19

Hi SD.5,

ST25R3911B compared to ST25R3911 brought some bug fixes and minor changes in the register layout. My expectation would be that a firmware would still work (only few people have codes in a way to be hit by the small differences). I don't recall changed analog performance.

Can you share the marking of the ST25R3911B and the content of IC Identity register (3Fh)?
How do you observe changed operating distance: Wake-up or card reading?
Can you check the resulting regulator value of 3911 vs 3911B?

Best Regards, Ulysses

SD.5 · ‎2021-04-26

Hi Ulyses,

1.Can you share the marking of the ST25R3911B and the content of IC Identity register (3Fh)?

The marking on the ST25R3911B is 25R3911BFN04D S9THA 046 and the content of 3Fh is 0x0D

2.How do you observe changed operating distance: Wake-up or card reading?

we are seeing a drastic change in the card reading distance.

3.Can you check the resulting regulator value of 3911 vs 3911B?

We have measured the voltage on VSP_D, VSP_A, VSP_RF. All of them are same between 3911 and 3911B.

Can you share me the errata sheet for the bug fixes and minor changes in the register layout?

Thanks

Sarav

Ulysses HERNIOSUS · ‎2021-04-27

Hi Sarav,

ad 1) You have latest silicon - confirmed.

ad 2) Can you share some more details here: Which technology? What is the difference in reading distance? Does the ST25R3911B signal errors in this case? Can you provide a logic analyzer trace?

ad 3) I meant reading out registers 2Ah and 2Bh. Are you running on 5V - I assume bit sup3V=0?

I am trying to collect more information on the fixes and changes - ongoing.

Regards, Ulysses

SD.5 · ‎2021-05-02

Hi Ulysses,

1) Thanks for the confirmation.

2) We are seeing error on the Antenna calibration display register(23H) and the value is 0xF8. With the same firmware on the old chip, the value is 0x00.

It happens during wakeup on card detection(Phase wakeup).

And also we measured the RF voltage using calibration coil, it is less compared to the old chip.

We probed the trim pins as well. We are seeing the RF signal on the old Chip but not on the new chip.

On reading distance, More than 2cm in the old chip, on new chip it is detecting only at on top of the antenna

3) Register Values

2Ah is 0xC0

2Bh is 0x80

Thanks

Sarav

Ulysses HERNIOSUS · ‎2021-05-03

Hi Sarav,

still not sure about the reasons. 2Ah: C0 means you are using manual configuration of regulator. So I can not imagine a difference here.

Antenna calibration values (23h: F8 vs 00) would means completely different trim values. I can imagine here a major difference in performance.

Can you monitor registers 21h, 22h and 23h?

Are you issuing any SetDefault commands in your software?

In order to rule out differences in matching and trim pin configuration: Please set trim_s=1 sweep tre bits from 0 to 15, measure and compare amplitude and phase by means of internal measurements....

Regards, Ulysses

SD.5 · ‎2021-05-03

Hi Ulysses,

On the registers

21h is set to 0x00

22h - We are not setting it explicitly and hence default value is being set and it is 0x80

23h is the display register and we are seeing difference in this register between old chip(0x00) and new chip(0xF8).

If direct command (Calibrate Antenna -D8) is set then 23h is 0xF8, and we are seeing the problem. Now we commented out D8 in the code and then 23h is 0x00 and power is boosted and reading distance is good. We want to understand why this is happening in the new chips?

We are using the SetDefault command only during the Power up.

We will set the calibration registers for different values and test.

Thanks

Sarav

Ulysses HERNIOSUS · ‎2021-05-03

Hi Sarav,

I have re-visited the changes and datasheet of the chips in question. I don't see a reason for such a change in behavior.

Please compare the sequences of Calibrate Antenna - D8. It should achieve on both chips the same result ( given that target register 22h is the same).

Please also provide tables of amplitude and phase results over all 16 values for both your 3911 and 3911b board.

Regards, Ulysses

SD.5 · ‎2021-05-10

Hi Ulysses,

On the reference code, There is a comment says that the Antenna calibrate command needs to be run twice( Errata #1.5). (Screenshot attached)

Can you share the errata 1.5 with us?

Please find the Amplitude and Phase values for both the chips.

Thanks

Sarav

Ulysses HERNIOSUS · ‎2021-05-10

Hi Sarav,

can you double check your phase measurements it is always 0x00 (>=150deg)? Which function did you use? From RFAL?

With such a phase for all trim values the Calibrate Antenna should always have produced an error (also in the old ST25R3911) as the target phase could not be achieved.

I will need to discuss internally, it would be helpful if you could use VNA to produce smith charts of at least trim values 0, 7 and 15 and share them here or with me.

Regards, Ulysses