This post entry aims at explaining the debugging process oriented to EMVCo Contactless certification of a device integrating NXP's PN5180. The structure is the following:
PN5180 Antenna design considerations
Before going into the debugging process for the EMVCo Contactless Analog tests we will see some important considerations for an antenna design and impedance tuning oriented for an EMVCo compliant device.
Antenna tuning recommendations
The first recommendation is that with the Dynamic Power Control feature the PN5180 allows us to perform symmetrical antenna tuning instead of the typical asymmetrical tuning. This symmetrical tuning provides us with a better transfer function, being able to drive more power to the antenna. The following figure shows the Smith Chart with the S11 parameter plot of a device using a symmetrical antenna tuning:
The only disadvantage of the symmetrical tuning is that we need a current limiter to avoid destroying the chip because of exceeding the chip’s limits. In the case we are documenting today, the PN5180 DPC feature is used to limit the supply voltage and therefore the transmitter current depending on the load detected by the chip.
Regarding the EMC filter, the inductor should fit with the following condition to guarantee a good relation between the AGC and the ITVDD:
Another consideration is about the resistor used in the reception branch. This resistor controls the receiver sensibility and as a starting point is recommended to use a value to obtain an AGC in free air of:
- Reader Mode only design: AGC value in free air around 600dec
- Full NFC design: AGC value in free air around 300dec
Finally, EMV contactless transactions are performed at 106kbps which would allow us to work with a high Q factor of the overall system. This means that the power gain can be higher, but at the same time it might also lead to some issues because of the lower bandwidth. In light of this, we have to bear in mind, that if the Q factor is too high it may lead to problems in the waveform tests.
PN5180 DPC calibration
The Dynamic Power Control is a feature that uses the AGC value to establish different power configurations depending on the load applied to the antenna.
As I mentioned before, the main goal is to protect the chip from a transmitter current level that might destroy it. The first step before calibrating the DPC is to check the correlation between the AGC value and the transmitter current or ITVDD when different loads are applied to the antenna. Basically, we will play with the distance between the load and the device to get several points with different AGC values. Based on those measurements, we can plot a graph like the following:
Normally we would use a reference PICC and a metal plane or phone to check that the behavior is linear and with no big difference between those loads.
Once we have checked the correlation we can proceed with the calibration process, which can be done very easily with the NFC Cockpit software.
Here the important thing is to control the ITVDD and keep it always below the chip’s limit. As you can see in the figure below, without the DPC,
this symmetrical tuning would lead to a voltage above the limit for positions close to the reader antenna. However, with DPC we can control that voltage at any moment.
Another consideration is that we have to make sure that the DPC is calibrated to have maximum power when the reference PICC is far from the reader to avoid a lack of power in the tests at those positions.
EMV L1 Analog Tests
We are going to divide this debugging process into 3 main phases which are the power tests in the first instance, followed by the waveform tests and the reception tests.
The reason why we set this order is to first debug the tests that may require HW modifications which have a strong impact on the other tests. This way, for example, if you have passed all power and waveform tests, debugging the reception tests may not have an impact on the results obtained previously.
In order to debug the power tests, we will need just an oscilloscope and an EMVCo reference PICC. We will need to connect the outputs J9 and J1 of the EMVCo reference PICC to the oscilloscope and set the jumper J8 of the reference PICC in non-linear load mode. The J9 of the EMVCo reference PICC is the DC_OUT output that we will use to measure the power received by the antenna. The J1 is the LETI_COIL_OUT output and we will use it to capture the command in the oscilloscope. The overall setup is depicted in the figure below.
We have to use the trigger to capture the REQA command sent from the DTE when the reference PICC is in the position we want to test. This capture can be seen in the two figures below.
The yellow channel is the LETI_COIL_OUT of the EMVCo reference PICC and the blue channel represents the DC_OUT obtained from the J1 connector. As said previously, we will use the DC_OUT to measure the voltage in the period of the signal where there is no modulation, like this part highlighted with the red squared.
We will use this same procedure to evaluate the power tests in all positions. Depending on the position tested, the specifications define and certain range where the voltage measured should be fitted. In this sense, the maximum voltage level is common for all planes, but the minimum voltage allowed will decrease for positions further from the terminal.
In order to identify the critical positions for the power tests, we have to identify two different scenarios, the first one with the positions that might not reach the minimum voltage established, and the positions that might exceed the maximum value. For the first scenario the critical positions are the outer positions of the plane z = 4cm and the plane z=3cm as the external positions for plane z= 3cm have a bigger radius.
The other scenario is that where you can be exceeding the maximum level. This situation can happen in the central positions of the lower planes, like plane z=1 or z=0.
In order to overcome possible issues, we will give some tips that can be used for your design. Regarding a case of lack of power, first, we have to make sure that the DPC is correctly calibrated, meaning that you are operating in gear 0 for the external positions of planes 3 and 4 and that gear 0 is operating with full power.
If we have verified those two things and we still have issues, we would need to change the tuning of the antenna and reduce the target impedance. This is graphically represented in the following Smith Chart:
By reducing the impedance we increase the current that the PN5180 is driving to the antenna so the voltage would increase. Is important to always verify that we are working within the recommended operating range of the chip and that we are not exceeding the transmitter current limit. In a worst-case scenario, if we cannot achieve the voltage with these HW changes we would need to evaluate changes in the hardware design, like adding a ferrite sheet or changing the antenna dimensions or position.
On the other hand, if the problem comes because we are exceeding the maximum voltage allowed by the specifications we can easily solve it by reducing the power configuration of the gear used in that specific position.
For the waveform group of tests, we will use a setup consisting of the EMVCo reference PICC along with an oscilloscope and a PC software to evaluate the signal obtained from the oscilloscope. In our case, we will use the Wave Checker software from CETECOM. We need to connect the output J9 of the EMVCo reference PICC to the oscilloscope and set the jumper J8 of the EMVCo reference PICC in the fixed load position. The oscilloscope needs to be connected to the PC or laptop, so the software is able to get the waveform and analyze the parameters needed.
Type A tests
The waveform group of tests for Type A consists of the following test cases:
- TA121: t1
- TA122: Monotonic Decrease
- TA123: Ringing
- TA124: t2
- TA125: t3 and t4
- TA127: Monotonic Increase
- TA128: Overshoot
Some of these test cases are directly related to the parameters defined for the specific modulation phase for Type A at 106 kbps. This modulation phase along with the respective parameters is depicted in the figure below.
When the Wave Checker gets the oscilloscope capture, it automatically analyzes the signal, performing all the measurements and comparing them with the specifications limits.
Debugging hints for Type A
The PN5180 has a few registers and parameters to control the wave shape generated by the NFC chip and transmitted by the antenna. These are the most relevant ones:
- TX_CLK_MODE_RM (RF_CONTROL_TX_CLK register)
- Rise and Fall times (RF_CONTROL_TX register)
- TX_OVERSHOOT_CONFIG register
From all the different test cases we will show how to debug the t3 and t4 test case as it is usually the most problematic. For this purpose, we will start from a certain configuration where the waveform tests show the following results, with a fail in the t3 and t4 test case.
In order to tackle this problem, we will rely on the TAU_MOD_RISING parameter from the RF_CONTROL_TX register of the PN5180. In this case, as the timings are slightly above the maximum allowed in the specifications we will decrease the TAU_MOD_RISING 3 points and execute again the tests. The results after the modification show that all test are passing with a certain margin:
Another parameter that the PN5180 has and can be used for the waveform tests is the TX_CLK_MODE_RM parameter from the RF_CONTROL_TX_CLK register. Below you can see two graphs that clearly illustrate the effect of this parameter over the waveform.
As you can see from the two figures, by changing the default high impedance configuration of 001, to a low side pull configuration the waveform results in a smoother decay of the envelope.
Type B tests
For Type B waveform, the specifications define the following test cases:
- TB121: Modulation Index
- TB122: Fall time
- TB123: Rise time
- TB124: Monotonic Increase
- TB125: Monotonic Decrease
- TB126: Overshoots
- TB127: Undershoots
Again, these tests are based on the different parameters that can be identified for the modulation phase of the Type B commands:
Debugging hints for Type B
The register and parameters that the PN5180 includes to control the waveform for type B are:
- TX_RESIDUAL_CARRIER (RF_CONTROL_TX register)
- TX_CLK_MODE_RM (RF_CONTROL_TX_CLK register)
- TX_UNDERSHOOT_CONFIG register
- TX_OVERSHOOT_CONFIG register
For Type B, we will study the modulation index test case, as it is the one that needs to be adjusted more often. In this case, we start from a situation where the device presents problems in the modulation index at 1 cm, with a value below the limit.
In order to make corrections of the modulation index we will use the TX_RESIDUAL_CARRIER parameter from the RF_CONTROL_TX register. This parameter controls the amplitude of the residual carrier during the modulated phase. For the present problem, we will increase it by 4 points and rerun the test. As you can see in the picture below, the modulation index is within the specifications limits with margin.
Adaptative Waveform Control
The PN5180 has another interesting feature called Adaptative Waveform Control that is used to set a different transmitter configuration depending on the gear and protocol used at any moment. This way we can easily debug by positions and use specific configurations for a certain group of positions without the need of rerunning all the tests for the rest of the positions.
With the AWC feature we can control the:
- TX_RESIDUAL CARRIER
We can see in the table an example of an AWC configuration for Type B. Where we have changed the Residual Carrier from gear 2 onwards. As you can see, It is also configured with a change in the falling and rising times from Gear 1.
As you can see this Adaptative Waveform Control feature along with the DPC represent a powerful tool to easily debug waveform tests without a change in the HW.
The reception tests purpose is to evaluate the ability of the device to identify and correctly demodulate the responses from the PICC when this response comes in the limits of the specifications for amplitude and polarity of the modulation.
The tools and setup needed to debug the reception tests for EMVCo are depicted in the following figure:
- Oscilloscope to capture the signal received by the reference PICC.
- Arbitrary Waveform Generator to generate the response of the PICC.
- PC Software to control the AWG and load the EMVCo responses to the EMVCo reference PICC. For our case, we will use the Wave Player software from CETECOM.
- EMVCo reference PICC.
This time, we will use the output J9 of the reference PICC to the oscilloscope to capture the command from the reader and trigger the injection of the response from the waveform generator to reference PICC, connected to J2. We should connect the waveform generator to the computer that has the Wave Player software installed to load the EMVCo responses.
As said previously, the reception tests aim at testing the ability of the device to correctly interpret the response when it is generated at the limit of the amplitude and polarity of the modulation. Considering the positive and negative polarity and the maximum and minimum amplitude of the modulation we have the following four test cases that are performed both for Type A and Type B:
- Tx131: Minimum positive modulation
- Tx133 - Maximum positive modulation
- Tx135 - Minimum negative modulation
- Tx137 - Maximum negative modulation
To debug these tests with the PN5180 we will use:
- RX_GAIN (RF_CONTROL_RX register)
- RX_HPCF (RF_CONTROL_RX register)
- MIN_LEVEL (SIGPRO_RM_CONFIG register)
- MIN_LEVELP (SIGPRO_RM_CONFIG register)
The procedure is basically to use the Waveplayer to set the amplitude and polarity of the response and check in the device is the response was correctly received and demodulated.
To debug the reception we will test different configuration for the RX_GAIN and RX_HPCF parameters that control the reception filters, amplifier and ADC blocks from the receiver branch. These receiver blocks are pictured in the diagram below.
Depending on the values used for the RX_GAIN and RX_HPCF parameters, the filter will be defined accordingly. The following table shows the filter characteristics in relation to those values:
If we don’t find a correct value to pass the test at a certain position, we should modify the Rx resistor in order to increase or decrease the receiver sensibility.
Adaptative Receiver Control
In the same line as the Adaptative Waveform Control, the PN5180 includes the Adaptative Receiver Control that can be used to define different reception configurations depending on the gear and protocol used. With the ARC we can control all the registers involved in the reception and apply a correction to the preconfigured value depending on the gear used.
We can see an example of the Adaptative Receiver Control configuration in the following table, where we have defined a correction of -1 to the MIN_LEVEL and the HPCF parameters from gear 1. We can also see that the RX_GAIN parameter has a correction of +2 from gear 0.
The ARC is very useful when we can't find a proper configuration for all positions and we need a different set of values depending on the positions tested.
Rx Matrix tool
Another interesting tool for debugging the reception tests is the Rx Matrix tool. This tool is used to launch and tests different receiver configuration in an automated way. The Rx Matrix tool is integrated into NXP's NFC Cockpit and you can control the Arbitrary Waveform Generator to set the amplitude of the modulation used for the tests. We can select which parameters we want to change and in which range we want them to be tested and the Rx Matrix will automatically run all the possible combinations in a sweep.
With the Rx Matrix tool, we can select the expected response and the number of iterations we want to try for every possible configuration. That way we can obtain a success ratio for the communication and easily identify the best configuration for the position tested.
An example of the Rx Matrix is given in the figure below. We have fixed the RX_GAIN and RX_HPCF parameters and performed a sweep for the MinLevel, testing it from a value of 0 to 8. We have set the Rx Matrix to execute 50 iterations for every configuration, obtaining the success ratio results plotted below.
As you can see the Rx Matrix along with a Waveform Generator is a powerful tool to find the optimum receiver configuration in a short time and in an effortless way.
The PN5180 comes with a complete and useful product support package including:
- The demokit, that can be used to get introduced to the product and check its features.
- The NFC Cockpit, that we have talked about during this article, and that represents a powerful tool to control the PN5180 with a very intuitive and useful interface. We srongly recommend that you integrate this tool in your final device as it may save you a lot of time during the debugging phase.
- A complete documentation including the updated product datasheet, or a set of application notes to guide you through all the designing process, from the antenna design guide to the DPC configuration or use of the Rx Matrix tool.
- Last but not least, the NFC Reader library which is the recommended software stack for NXP's NFC frontends and NFC controllers with customizable firmware.
NFC Reader Library
The NFC Reader Library comes with built-in MCU support, but it can also run on different MCU platforms, as well as non-NXP. The library has been built in such a way that you can adapt it and implement the required driver for your host platform. Other characteristics are:
- It is free of charge and you can download the latest release from NXP’s website.
- It is a complete API for developing NFC and MIFARE-based applications.
- Includes an HTML-based API documentation for all the components, which is generated from source-code annotations.
- Finally, the release includes several examples and applications.
Among the examples and applications included in the NFC Reader Library we can highlight two applications that are very useful for the preparation of the Device Test Environment required for the EMVCo certification:
- The SimplifiedAPI_EMVCo for the digital testing
- The SimplifiedAPI_EMVCo_Analog for the Analog testing.
You can control all the parameters involved in both applications using the phNxpNfcRdLib_Config.h configuration file. The identification and modification of these parameters should be very easy as the code is well documented, like you can see in the code chunk in the image: