Universally Compatible Wireless Power Using the Qi Protocol
Wireless charging of portable electronic devices is here now. It will become ubiquitous when all such devices adhere to the same standard.
The Wireless Power Consortium (WPC) has developed a standard for wireless power systems referred to as Qi (pronounced as “chee”). This Qi-compliant wireless power system allows compatibility between devices from multiple manufacturers. It also allows handheld equipment from original equipment manufacturers (OEMs) to focus exclusively on the design of their equipment without having to design a customized wireless power pad, since multiple Qi-compliant power sources are already available.
The key to interoperability within the Qi standard is the communication protocol. This article explains the fundamentals of how the receiver (RX) device (handheld equipment) communicates with the transmitter (TX) device (charging pad providing energy). Communication packets sent by the receiver (and the corresponding response from the transmitter) are illustrated. Communication from the receiver to the transmitter allows the closed-loop control and regulation of the receiver circuit’s output voltage.
Wireless power systems are emerging as a practical option for conveniently recharging mobile phones and other handheld devices. Implementing an industry standard interface allows a common charging pad (TX) to recharge multiple types of battery-operated devices (RX). The WPC developed the Qi standard for wireless power systems with up to 5W of output power. This allows complete interoperability between transmitters and receivers independent of device manufacturer.
This standard defines the means of implementing a number of functions that enhance the utility and efficiency of a wireless power system, for example:
- The charging pad does not consume significant amounts of standby power when idle (no device placed on the pad).
- The TX can detect the presence of an object placed on the pad, and further determine that it is a valid, Qi-compliant RX device.
- Once an object is placed on the charging pad, the transmitter can output a variable power level based on the TX’s requirements.
- The RX unit communicates its power needs back to the TX unit over the same magnetic coupling used for power transmission.
Figure 1. Qi-compliant wireless power system.
Figure 1 shows a block diagram of the overall Qi-compliant wireless power system. References 1 and 3 provide a more complete description of the wireless power system.
Basics of Communication
The resonant converter in the TX circuit generates a quasi-sinusoidal AC signal in the range of approximately between100-200 KHz across the primary TX coil. This signal is magnetically coupled into the secondary RX coil, where additional circuitry rectifies and regulates it to provide DC output to the handheld device.
Communication from the RX back to the TX uses the same magnetic coupling path as the forward power transfer. A simple load modulation method is used to communicate status and commands back to the TX controller. When the receiver circuit is powered-up, it can apply a controlled pulsed load across the secondary coil (Figure 2). This results in an amplitude modulation of the primary coil voltage which is detected and demodulated by the TX controller.
The load modulation can use either a resistive or capacitive load element. The modulated load is internal to the RX side circuitry and independent of the actual system load (battery or portable device).
Figure 2. Receiver-to-transmitter communication.
Figure 3. Communication pulses on RX side and corresponding modulation of TX coil voltage.
Figure 3 shows an example of the actual effect seen on the primary TX coil voltage as a result of the modulation pulses on the secondary RX side. These waveforms correspond to the ideal waveforms illustrated in Figure 2. The modulating signal on the RX side is measured at the point labeled “COMM DRV” in Figure 1.
The data in Figure 3 shows that load pulses on the RX side correspond to an amplitude modulation effect on the primary side (TX coil voltage and/or current). The communication from the RX side uses a “differential bi-phase” bit encoding scheme. Since there is no separate clock line or control signal path, a fixed clock frequency of approximately 2 Khz is used with a start bit before each 8-bit transmission, followed by parity and stop bits. Figure 4 illustrates the bit / byte encoding schemes as defined in the WPC Specification v1.0.2 (Reference 1).
Figure 4. WPC Qi specification bit encoding and format.
A communications packet consists of four specific sections:
- Preamble: a fixed sequence of several “1” bits, which allows the TX circuit to detect the start of communications, synchronize to the bit stream, and detect the start bit of the header byte to follow.
- Header byte: defines what type of information is to be transmitted (“packet type”) and the length of the message to follow.
- Message field: typically this is one byte, but could be larger depending on the type of information required; the size of the message field is determined by the packet type.
- Checksum byte: used at the end of each packet to allow the TX side to verify that no errors occurred after each packet transmission.
A number of specific functions are defined by the types of packets that can be sent. As of now, not all possible options are implemented, but room is still available for expansion of functionality with future versions of the Qi standard. Reference 1 provides a complete description of existing packet types. The most common packets and their functions are:
- Signal strength: used to help align the RX unit on the charging pad. This allows the charging pad to provide a visible or audible indication to the user when the signal strength sent back from the RX unit is good enough to allow power transfer (for example, the RX coil is properly positioned with respect to the TX coil).
- Control error packet: this returns a signed integer value (–128 to +127) that indicates the degree of error between the value of the input voltage seen by the RX and its desired input voltage. The TX circuit adjusts its output using a proportional-integral-differential (PID) algorithm in response to the signed value of the control error packet. When a large error exists between the actual and desired value of the RX coil voltage, the RX controller sends the error packets at a faster rate of approximately 32 ms intervals. As the coil voltage gets closer to the desired setpoint, the error packets are sent at a reduced rate of approximately 250 ms intervals (Figure 5).
- End power transfer packet: request by the RX unit instructing the TX to terminate power output. Typically this is due to a fault condition or if the receiver device no longer requires power, for example, when the RX device's battery has been fully charged.
- Rectified power packet: this is an unsigned integer value that communicates the amount of power the RX sees at the output of the rectifier circuit. The TX uses this information to determine the overall coupling efficiency as well as to determine when the RX is at its maximum power limit. The TX terminates power transfer when no packets are received for a fixed interval (350 ms minimum, 1800 ms maximum), indicating that the RX device has been removed from the pad.
Figure 5. Control error packet transmission and VRECT voltage response to a negative load transient (500 mA to 0 mA).
Optimizing load transient response using RX-to-TX communication
The ability of the RX side circuitry to communicate back to the TX side circuitry allows the overall wireless power system to act as a true closed-loop regulated power system, since the equivalent function of an analog error signal feedback is accomplished by the RX controller's control error packets being sent back to the TX controller. From an overall system point of view, the implementation shown for the wireless power TX/RX combination can be treated as a switch-mode converter with a low-dropout (LDO) post-regulator.
The raw input voltage to the RX coil can be highly variable in an "open loop" configuration as it will fluctuate significantly with variable load. To maintain good regulation of the final DC output (for example, +5V), the feedback provided from the RX side adjusts the input to the linear regulator (VRECT) up or down, based on the load current conditions.
Figure 6. Control error packet transmission and VRECT voltage response to positive load transient (0 mA to 500 mA).
When the output is lightly loaded, the RX circuit sends control error packets back to the TX controller to increase the VRECT input voltage applied to the LDO stage up to approximately +7V. The regulated output is set to +5V. Since the load current is light, the 2V input-output differential across the LDO does not represent significant power loss.
The reason for setting the VRECT level higher at light loads is to anticipate the effect of a low-to-high current transient. When this transient occurs, the VRECT voltage initially sags until the RX controller can respond. The error packets sent by the RX controller request the TX controller to raise the output voltage. Leaving the VRECT level high at light loads provides enough headroom to prevent the +5V regulated output from collapsing until the digital communication (feedback) can be sent.
At higher load currents, the VRECT voltage is kept as low as possible to minimize power loss across the LDO (and maximize total system efficiency). For example, at the maximum load current of 1.0A, the VRECT signal is set to approximately 5.20V. For example, the dropout performance of the LDO regulator stage within the bq51013 (wireless receiver IC) allows it to maintain a regulated 5.0V at 1.0A load current.
Figures 5 and 6 illustrate the adjustment in the VRECT setpoint based on load current described earlier. Note that the COMM signal bursts shown correspond to complete packets rather than individual bits due to the time scales of the plot. When the VREC deviates significantly from the desired setpoint by a large error, the COMM packets are sent at a faster interval. As the VRECT approaches the desired setpoint, the COMM packet transmission interval is decreased.
Figure 7. RX output voltage response to maximum load transient.
Figure 7 shows the system output voltage response to a large load transient (corresponding to the maximum load transient case of 0A à 1A). The maximum load transient results in less than 100 mV droop at the output, such as an approximate two percent deviation from the regulated output voltage.
Measurement of RX and TX signals
Qi-standard wireless power devices allow a system designer to implement a Qi-compliant power system using an integrated solution, such as the bqTESLA™, and requires no programming implement the communications protocol. Evaluation modules (EVMs) are available for both TX controller and RX controller sections.
Figures 8 and 9 are partial schematics from the bqTESLATM evaluation modules which highlight the measurement points used to collect the data in Figures 3, 5, 6, and 7. In the case of the discrete receiver circuit (>SLVU420), the COMM DRV signal can be directly measured since the load modulation FETs are external to the RX controller IC. When using the integrated RX circuit (SLVU477) the load modulation FETs are integrated within the RX controller and their gate drive signals cannot be accessed. However, the communication pulses from the RX controller still can be detected by measuring the differential voltage across the load capacitor C13 as shown. The complete schematics of the bqTESLATM EVM kits are provided in Reference 4 and 5.
Figure 8. Partial schematic of HPA687 evaluation module (discrete wireless power receiver).
Figure 9. Partial schematic of HPA725 evaluation module (integrated wireless power receiver).
When using a fully integrated Qi-standard chipset solution, all of the communication from receiver to transmitter is handled automatically with no user programming required. However, a basic understanding of the communication protocol can help the system designer know how to test and verify that the system is operating properly.
At a fundamental level, the communication can be thought of as amplitude modulation (AM) with a modulation frequency of 2 KHz and carrier frequency ranging from 100 to 200 KHz. This simple, robust protocol defined by the WPC Qi standard allows communication to occur along the same inductively coupled path as the forward power transfer, and does not require a separate set of contacts or magnetics.
The wireless power receiver’s ability to communicate its power needs back to the transmitter (based on load conditions) allows the system to maintain a stable output voltage under constant or transient load conditions. The closed-loop nature of the overall system is achieved by using a Qi communication protocol.
The authors would like to thank Steve Terry, Tony Antonacci, and Michael Day for their technical and editorial contributions to this article.
- “Qi low-power specification,” Wireless Power Consortium website.
- “An introduction to the Wireless Power Consortium standard and TI’s compliant solutions,” by Bill Johns, Texas Instruments, Analog Applications Journal, Texas Instruments, 1Q2011.
- “bq25046EVM-687 Evaluation Module User’s Guide,” SLVU420, Texas Instruments, June 2011.
- “bq51013EVM-725 Evaluation Module User’s Guide,” SLVU447, Texas Instruments, March 2011.
- “bqTESLA Wireless Power Transmitter Manager EVM User’s Guide,” SLVU467, Texas Instruments, June 2011.
- TI bqTESLA Home Page, Wireless Power Products, Links and FAQs.
About the Authors
Bill Johns is an Applications Engineer for bqTESA Wireless Power at Texas Instruments where he specializes in power conversion for portable battery powered applications. Bill has 20 years of design experience in the field of power conversion across a number of different markets. He received his BSEE from University of Texas in Dallas. Bill can be reached at email@example.com.
Upal Sengupta is a Portable Power Strategic Marketing Manager at Texas Instruments where he defines requirements for new analog / power integrated circuits, provides extensive technical support for customer designs related to portable power, battery monitoring, and charging applications, and develops reference designs and evaluation systems for power management IC devices. He received his BSEE from the University of Illinois, and a MSEE from Michigan State University. He can be reached at: firstname.lastname@example.org.