Generate PN5190 DPC Lookup Table with NFC Cockpit

[deniscollis]

I'm attempting to create a DPC Lookup Table with NFC Cockpit.  My board comprises a K8x, SPI connected PN5190, and my own tuned antenna.  I have ported the board firmware (VCOM/UcBalFW v03.08.00) to my board, and can successfully control the PN5190 with the NXP NFC Cockpit V6.7, V6.8 or V7.1 Windows application. In addition I have a Dobot Magician Robot, Oscilloscope, and EMV approved PICCs. 

I am using Renke Bienert's PN5190 Dynamic Power Control Quick Calibration and TxShaping Demo to perform the DPC calibration.

With the PICC positioned at 4.0.0 (4cm, center), and VDDPA at 5.7V maximum I get a transfer power measurement of around 3.4V with a TxLDO current (ITVDD) of  251mA, which is the Target Current that I saved to the EEPROM.

Going through the procedure of 1) reducing the VDDPA in steps, 2) lowering the PICC closer to the antenna until the target power transfer of 3.4V is measured, and 3) entering the ITVDD for that VDDPA; I end up with the following:

Now, when I click [Compute and Move to LUT], only the first 2 entries are populated:

The demo video instructs me to interpolate the 'in-between' values in the current reduction table, but the data is too sparse to do so.  Not even the corresponding VDDPA entries (5.5, 5.1, 4.1, 3.1, 2.1, 1.7 and 1.5) are populated.

Any idea what's going on here?

Thanks,

Denis 

[deniscollis]

Hi Kelly,

I agree with your annotation "...does not match the actual situation at all." That's the whole point. I'm trying to figure out why NFC Cockpit failed to generate the current reduction entries correctly.  I would expect at least the 5.7V entry to show a current reduction of 0.

You may have noticed that I have a 3.4V power transfer target, instead of the much higher 6V+ that Renke recommends. I'm deliberately throttling back on the target current from a possible max of about 300mA down to 251mA to avoid a PN5190 over-temperature condition. Could this be causing the the calculation issue?

Best,
Denis

[KellyLi]

Hello @deniscollis 

Yes, after discussing with the internal expert, suggestion to you like this: you can do that if you want, but still that does not sound like a good design. The Cockpit logic for DPC calibration cannot be changed unless special firmware is updated. Based on this situation, it's still recommended that DPC calibration under the full range of voltages. If you are concerned about overheating you can limit the target current and the DPC only controls the VDDPA based on the measured current.

Hope the above can be helpful to you.

BR

kelly

[deniscollis]

Hi @KellyLi 

NFC Cockpit. Target Current set to 250mA  &  DPC Status [Enabled]

If I move the PICC closer to the PCD the current exceeds 250mA.  I thought that the Target Current was the maximum limit. I even tried setting the target to 100mA and lowered the hysteresis values to see how the DPC would behave. The DPC does not appear to be working at all! 

This thing has defeated me. I think I'll go fishing instead

Best,
Denis

[KellyLi]

Hello @deniscollis 

Therefore, it is highly recommended that you calibrate over the full range of voltage to make the DPC conform to a normal logic. The calibration process is very brief, and that little time of high current will not cause overheating (The PN5190 can drive up to 2W at 5.7V@350mA). After the calibration is complete, if you are worried about overheating when your device is running normally, you can take some measures to limit the maximum power output. You can refer to Section 2 of AN13669 for details.

From theoretically and practically, after correct calibration and enable the DPC function, the current will not exceed the target current what you set. So, another suggestion to you, if possible, please running the complete logic of DPC using NXP's development board first.

Please feel free to contact us if any questions.

BR

kelly

[deniscollis]

Hello @KellyLi Kelly,

I'm going through the calibration exercise on the PNEV5190B dev board.  At first I used the NXP-provided EEPROM settings that are included with the secure firmware update V2.05.

The first thing I notice is that the TxLDO Current(mA) is set to 65514 after I [Start Calibration].  Clearly this indicates that the value has gone negative. (This does not happen on my board, but I have silicon version B1, and the dev board has version B0.)

Next thing I notice is that clicking [Load Protocol] does not activate the RF field.  (This happens with my board too.) However, toggling DPC Status to [Disable] and back to [Enable] fixes this, and also the negative TxLDO current, above.

Lastly, I notice is that the target current is set to 306mA in the default EEPROM settings.  But with the PICC at 4cm and RF active, with DPC enabled or disabled, the TxLDO current is 278mA.  Lowering the PICC to about 3cm I see the current hits 326mA just before the VDDPA switches from 5.7V all the way down to 4.2V, skipping the intermediate transitions. This is not the way it is described in the training video.

It appears that the PN5190 Dynamic Power Control Quick Calibration training video is inaccurate in its content, and produces inaccurate results.  Is there a step-by-step procedure document that I can use instead?

Thanks,
Denis

[KellyLi]

Hello @deniscollis 

Please update the below firmware and try again according to Renke's video.

BR

kelly

[deniscollis]

I Found the explanation for the higher than expected current in AN12549 - PN5190 Antenna Design Guide:

"The VDDPA is automatically reduced, as soon as the current exceeds the DPC_TARGET_CURRENT + DPC_HYSTERESIS_LOADING, and the VDDPA is automatically increased again, as soon as the current is below DPC_TARGET_CURRENT – DPC_HYSTERESIS_UNLOADING." 

Clearly, I failed to compensate for the hysteresis.  The document goes on to state (I'm not sure what 'dec' is, below, but assume decimal):

"The hysteresis shall not be less than 20 dec (loading) and 10 dec (unloading)."

This explains why I'm getting a higher than expected power transfer voltage.  It's simply that, when lowering the PICC, a current increase of 20mA is not achieved, so the transfer power just keeps increasing.

[KellyLi]

Hello @deniscollis 

Yes, "dec" means decimal. DPC_HYSTERESIS_LOADING and DPC_HYSTERESIS_UNLOADING can be set by register, which is described in detail in the datasheet. Therefore, it is recommended that you dump Eeprom and check if these parameters meet calibration. 

BR

Kelly

[KellyLi]

Hello @deniscollis 

Thanks for contacting us and choosing the NXP products.

Based on your present information, there are some comments as follows:

1. Did you tune your antenna? Before DPC calibration, antenna need to be tuned.

2. Are the calibration methods and steps correct? Judging from the point-by-point calibration value, the calibration result is inaccurate and too far from the target value, especially the last few points. The value of each calibration point should be very close to the target value.

3. Please check the below fig:

So, based on the above comments, it's strongly recommended that that you re-calibrate your device strictly according to Ranke's video.

Please don't hesitate to contact us if still have any questions.

BR

kelly