AN146 H I G H - S PEED L I T H I U M I O N B ATTERY C HARGER Introduction • On-chip comparator and PWM provides a means to implement a high speed buck converter with a small external inductor. On-chip temperature sensor provides an accurate and stable drive voltage for determining battery temperature. An external RTD (resistive temperature device) can also be accommodated. A single C8051F300 provides full product range for multi-chemistry chargers, expediting time to market and reducing inventory. Driven by the need for untethered mobility and ease of use, many systems rely on rechargeable bat- • teries as their primary power source. The battery charger is typically implemented using a fixedfunction IC to control the charging current/voltage profile. • The C8051F300 family provides a flexible alternative to fixed-function linear battery chargers. This note discusses how to use the C8051F300 device in Li-Ion battery charger applications. The Li-Ion charging algorithms can be easily adapted to other Charging Basics battery chemistries. Batteries are exhaustively characterized to determine safe yet time-efficient charging profiles. The The code accompanying this application note was optimum charging method for a battery is depenoriginally written for C8051F30x devices. The dent on the battery’s chemistry (Li-Ion, NiMH, code can also be ported to other devices in the Sili- NiCd, SLA, etc.). However, most charging stratecon Labs microcontroller range. gies implement a 3-phase scheme: Key Points 1. Low-current conditioning phase • 2. Constant-current phase On-chip high-speed ADC provides superior accuracy in monitoring charge voltage (critical to prevent overcharging in Li-Ion applications), maximizing charge effectiveness and battery life. 3. Constant-voltage phase/charge termination Figure 1. Lithium Ion Battery Charger Block Diagram. V Pos (+) LDO 8k FLASH, PWM, Temp Sensor, Precision Time Base 8051F300 LED Cygnal Integrated Products Buck Converter PWM Out Rev. 1.3 7/13 Li-Ion Cells AIN1 - Voltage AIN2 - Current V Neg (-) Resistor Divider Amplifier Sense Resistor Copyright © 2013 by Silicon Laboratories Silicon Laboratories Confidential. Information contained herein is covered under non-disclosure agreement (NDA). AN146 AN146 All batteries are charged by transferring electrical energy into them. The maximum charge current for a battery is dependent on the battery’s rated capacity (C). For example, a battery with a cell capacity of 1000mAh is referred to as being charged at 1C (1 times the battery capacity) if the charge current is 1000mA. A battery can be charged at 1/50C (20 mA) or lower if desired. However, this is a common trickle-charge rate and is not practical in fast charge schemes where short charge-time is desired. Most modern chargers utilize both trickle-charge and rated charge (also referred to as bulk charge) while charging a battery. The trickle-charge current is usually used in the initial phases of charging to minimize early self heating which can lead to premature charge termination. The bulk charge is usually used in the middle phase where the most of the battery’s energy is restored. During the final phase of battery charge, which generally takes the majority of the charge time, either the current or voltage or a combination of both are monitored to determine when charging is complete. Again, the termination scheme depends on the battery’s chemistry. For instance, most Lithium Ion battery chargers hold the battery voltage constant, and monitor for minimum current. NiCd batteries use a rate of change in voltage or temperature to determine when to terminate. Hardware Description Li-Ion batteries are currently the battery chemistry of choice for most applications due to their high energy/space and energy/weight characteristics when compared to other chemistries. Most modern linear Li-Ion chargers use the tapered charge termination, minimum current (see Figure 2) method to ensure the battery is fully charged, as does the example code provided at the end of this application note. Buck Converter The most economical way to create a tapered termination linear charger is to use a buck converter. A buck converter is a switching regulator that uses an inductor and/or a transformer (if isolation is desired), as an energy storage element to transfer energy from the input to the output in discrete packets. Feedback circuitry regulates the energy transfer via the transistor, also referred to as the pass switch, to maintain a constant voltage or constant current within the load limits of the circuit. While charging, some of the electrical energy is converted to thermal energy, until the battery reaches full charge, at which time all the electrical energy is converted to thermal energy. If charging isn’t terminated, the battery can be damaged or destroyed. Fast chargers (chargers that charge batteries fully in less than two hours) compound this issue, as these chargers use a high charge current to minimize charge time. Therefore, monitoring a battery’s temperature is critical especially for Li-Ion batteries which may explode if overcharged. Temperature is monitored during all phases and charge is terminated immediately if the temperature exceeds a preset maximum limit. 2 Rev. 1.3 AN146 Figure 2. Lithium Ion Charge Profile. Charge Voltage Charge Current Conditioning Current regulation Phase Voltage regulation Time Figure 3. Buck Converter. Switch + VIN Supply Current L Switch Path of current flow from VIN - - VIN D C + Battery D L Inductor Current VBatt + Path of current flow from L and C C Battery - VBatt + VREF VREF Comparator Comparator Switch OFF when VBatt > VREF Switch ON when VBatt < VREF a) Switch ON b) Switch OFF Rev. 1.3 3 AN146 Buck Regulator Operation The buck regulator operates by controlling the duty cycle of a transistor switch. The duty cycle is automatically increased to dispense more current into the battery. A comparator closes the switch when VBATT < VREF. As shown in Figure 3a, current flows into the battery and capacitor C. This current is also stored in inductor L. VBATT rises until it exceeds VREF at which time the comparator turns the switch off (Figure 3b). The current stored in the inductor rapidly decreases until diode D is forward biased, causing inductor current to flow into the battery at a decreasing rate. Capacitor C begins discharging after the inductor current has decayed and eventually VBATT begins to fall. When VBATT falls below VREF, the comparator again turns the switch on and another cycle begins. On a larger scale, if the duty cycle is decreased (shorter “on” time), the average voltage decreases and vice versa. Therefore, controlling the duty cycle allows one to regulate the voltage or the current to within desired limits. Selecting the Buck Converter Inductor To size the inductor in the buck converter, one first assumes a 50 percent duty cycle, as this is where the converter operates most efficiently. Duty cycle is given by Equation 1, where T is the period of the PWM (in our example T = 10.5S). ton DutyCycle = --------T Equation 1. Duty Cycle. With this established, select a PWM switching frequency. As Equation 2 shows, the larger the PWM switching frequency, the smaller (and more cost effective) the inductor. Our example code config- 4 ures the ‘F300’s hardware to generate a 510kHz switch rate. Vi – Vsat – Vo ton L = ---------------------------------------------------2Iomax Equation 2. Inductor Size. Now we can calculate the inductor’s size. Assuming Vi, the charging voltage, is 15V, Vsat, the saturation voltage, is 0.5V, the desired output voltage, Vo, is 4.2V, and I0MAX, the maximum output current, is 1500 mA, the inductor should be at least 4H. Note that the capacitor in this circuit is simply a ripple reducer. The larger it is the better as ripple is inversely proportional to the size of the capacitor. High Speed Charger As AN037, Lithium Ion Battery Charger Using C8051F300, illustrates, the F300’s 8-bit PWM can be configured to generate a 96kHz PWM with no external components. This PWM output can be used to drive the pass switch in a buck converter and charge a battery. However, a 96kHz frequency requires a buck converter to utilize a relatively large 18H inductor. For some applications, this is too large and costs too much. To reduce the size and cost of this inductor requires that the switch rate of the buck converter increase. The beauty of the F300 lies in its flexible feature set. As mentioned, included in the device is a PCA (Programmable Counter Array) that has three 16-bit capture/ compare modules with corresponding output drives that can be configured to provide numerous functions. We can use two of the PCA’s modules, along with two external single-pole low-pass filters, and the on-chip comparator to generate an 8-bit, 510kHz PWM (refer to Figure 4 for details). By setting the switch rate to 510kHz, the inductor required to satisfy the buck converter equations is reduced by a factor of five to approximately 4H. Rev. 1.3 AN146 To create a 510kHz PWM with the F300, Module 0 of the PCA is configured to provide a 510kHz square wave via the Frequency Output Mode. This square wave is then filtered through a low-pass filter with a 500kHz corner frequency to provide approximately a 2 Volt peak-to-peak pseudo triangle wave to the positive input of the on-chip comparator. For the minus input of the comparator, Module 1 is configured as an 8-bit PWM at 96kHz switch rate. This PWM output is then low pass filtered with a corner frequency of approximately 15Hz to create a simple DC digital-to-analog converter. By comparing the pseudo triangle wave to a DC input, the output of the comparator’s output becomes a 510kHz PWM. The DC control path, Module1’s output in this example, controls the duty cycle of the 510kHz PWM output from the comparator. By varying the duty cycle of the 8-bit 96kHz PWM, the minus input to the comparator can be varied from 0 volts to the supply, typically 3.3V. The accuracy of the DC control path is limited by the settling time of the external RC filter. For this example, components were selected to minimize errors contributed from this path. For more details on component selection for the DC path, refer to AN010, 16-Bit PWM Using an On-Chip Timer. The overall accuracy of the 510kHz PWM output from the comparator is mostly limited by the pseudo triangle path to the comparator. Assuming the DC path is error free, to create a true 8-bit 510kHz PWM output from the comparator requires that a perfectly linear full-scale triangle wave be input to the positive input of the comparator. A true full scale triangle wave refers to a triangle wave that linearly ramps from 0 volts to the positive supply and then returns back the negative supply in a similar fashion. However, the charge and discharge profile of a capacitor in the low pass configuration is not linear past a time constant. Moreover, it is desirable not to allow this capacitor to fully charge as the pseudo triangle wave becomes more nonlinear towards its peaks. Unfortunately, limiting the overall charge time/voltage limits the overall accuracy of the 510kHz PWM. For example, if we compare a triangle wave with a peak-to-peak voltage of 1/4 the supply to the DC control path’s voltage, we could generate a 2-bit 510kHz PWM output from the comparator. In practical applications, a 2-bit 510kHz has very limited use. To improve the accuracy requires either one of two changes: 1) increase the voltage of the pseudo triangle wave or 2) increase the resolution of the DC path’s PWM control. As mentioned earlier, increasing the peak-topeak voltage of the pseudo triangle wave can be easily accomplished by adjusting its low pass filter components accordingly. Our example is designed Figure 4. High Speed Charger. C8051F300 VBAT Module1 IBAT TBAT Module0 Comparator CMP0 CEX1 DC Control CEX0 CMP+ CMPHigh Speed PWM Rev. 1.3 Pseudo Triangle Wave Buck Converter 5 AN146 to provide approximately a 2 volt peak-to-peak pseudo triangle wave. When compared to the 8-bit PWM dc control path, we achieve approximately 7.5-bits of performance at 510kHz switch rate. The overall resolution of the high speed PWM output can be increased very easily by configuring Module 1 to a 16-bit PWM. On an aside, if a faster PWM output, greater than 510kHz, is desired to reduce the inductor size further, the user can reconfigure Module 0 to provide a faster square wave up to ½ the internal oscillator frequency or approximately 12MHz. If this is desired, the external low pass filter for the pseudo triangle wave path will have to be modified to accommodate the faster square wave. Other limitations, like comparator speed and voltage induced across the inductor due to the higher current transients will also have to be considered. Software Description takes these two points, calculates a slope and an offset for both the current and voltage channels, and stores the results in FLASH. All future conversions are scaled relative to these slope and offset calculations. Temperature To monitor the temperature, the algorithms use the on-chip temperature sensor. The sensor is left uncalibrated, but still provides a sufficiently accurate temperature measurement. For more accurate temperature measurement, one or two-point temperature calibration is required. An external temperature sensor can be used if desired. The AMUX can be reconfigured to accommodate this additional input voltage. Current The current delivered to the battery cells is monitored by measuring the voltage across a sense resistor (typically tens of mohms; our example uses a 0.24 ohm resistor). To maximize current measurement accuracy, this reference design uses an external amplifier with a gain of 10. This provides about 11-bits of current measurement accuracy (8-bits from the ADC and 3-bits from the external gain amplifier). To further maximize current measurement accuracy, the raw current measurements are scaled via the slope and offset calibration coefficients every time a measurement conversion is taken. The software example that follows demonstrates a Li-Ion battery charger using the C8051F300. The algorithms discussed are written entirely in “C” making them easily portable. Refer to the F300’s datasheet for a full description of the device. Note that the software architecture for the low speed (96kHz) charger discussed in AN037 and the high speed charger (510kHz) are essentially the same (i.e. the flow charts that follow can be used for either hardware configuration). The main difference are the control mechanisms. For the slow speed charger in AN037, Module 0 (CEX0) is used to control the duty cycle. For the high speed char- To determine the minimum current resolution, ger Module 1 (CEX1) is used to control the duty recall that the output code of a ADC is given by cycle. Equation 3 n Ain 2 Dout = -------------------Vref Calibration To ensure accurate voltage and current measurements, the algorithms use a two-point system calibration scheme. In this scheme, the user is expected to apply two known voltages and two known currents, preferable, one point near ground and the other point near full-scale. The algorithm then 6 Rev. 1.3 Equation 3. Digital Output Code. AN146 Accounting for the external amplifier, Equation 4 Ain = Iin Rs 10 The battery’s voltages are divided down and monitored via external resistors. Note that this example uses the supply voltage as the ADC voltage reference. Any monitored voltage above the reference voltage must be divided down for accurate monitoring. If a more accurate reference is required, an external voltage reference can be used. Adjustment to the divide resistors must be made accordingly. Equation 4. Input Current with 10x Gain. states that Ain is Iin is then given by Equation 5 Dout Vref Iin = -----------------------------------n 2 10 Rs Equation 5. Input Current. Charging - Phase1 In Phase 1, (for description purposes, we assume the battery is initially discharged), the ‘F30x regulates the battery’s current to ILOWCURRENT (typically 1/50 C) until the battery’s voltage reaches VMINVOLTBULK. Note that the battery’s charge current is current limited to ILOWCURRENT to ensure safe initial charge and to minimize battery selfheating. If at any time the temperature increases out of limit, charging is halted. Assuming • • • • • Voltage Rs = 0.24 VREF = 3.3 V 2N = 256, an 8-bit converter External Gain = 10 No External Gain = 1 When Dout = 1, IMIN is given by Equation 6. Charging - Phase 2 mA Imin = 5.37 --------------Code Once the battery reaches VMINVOLTBULK the charger enters Phase 2, where the battery’s algorithm controls the PWM pass switch to ensure the output voltage provides a constant charge-current IBULK to the battery (rate or bulk current is usually 1C and is definable in the header file as is ILOWCURRENT and VMINVOLTBULK). Equation 6. IMIN. When Dout=256, Imax is given by Equation 7. Imax = 1.37 Amps Equation 7. Imax. Charging - Phase 3 It is important to note that if one chooses to modify the algorithm, the order of mathematical operations is important. To minimize truncation error effects, be sure to perform multiply operations first making the numerator as large as possible, before performing divide operations. Further recall that a long type variable, which is the limit for the F300’s compiler, is limited to 232-1 or approximately 4 billion. After the battery reaches VTop (typically 4.2 V in single cell charger), the charger algorithm enters Phase 3, where the PWM feeds back and regulates the battery’s voltage. In Phase 3, the battery continues to charge until the battery’s charge current reaches IMINIBULK, after which, the battery is charged for an additional 30 minutes and then charge terminates. Phase 3 typically takes the majority of the charging time. Rev. 1.3 7 AN146 Note that in most practical applications, such as a portable PC, the batteries may be in any of the three phases when charging is activated. This doesn’t really affect the charger as it simply monitors the battery’s current condition and starts charging from that point. To charge more than 1 battery cell, both the hardware and software will need to be modified. For example, to charge two 4.2 V batteries simultaneously, resistors R11 and R12 will need to be switched. Then, the header file will need these modifications: Getting Started CELLs = 2 The reference design that accompanies this application note is designed to charge a single cell 4.2 V lithium ion battery. To accommodate numerous power supplies and batteries, it charges at 250mA bulk current. To charge a battery, first connect power to the board by applying an 8V to 15V supply to JP1. The power supply should be able to supply a minimum of 500mA. Once the appropriate supply is connected, connect the battery to JP3. Connect the positive lead of the battery to pin 1 and connect the negative lead to pin 3. Terminal 2 of JP3 can be left unconnected as no code has been developed to monitor temperature via an external temperature sensor at this time. Finally, press the reset switch and the battery will begin charging. The PWM charge control signal can be monitored by probing pin 5 on the C8051F300. RESB = 10 Recommended Operating Conditions • • Supply: 8V to 15V Battery: One cell, 4.2V, with <1000mAh rating Default Charging Parameters Once the header file is modified, recompile the software and down load the new source code to the charger board. A similar scheme can be used to modify the board for any number of cells. Conclusion The C8051F300’s high level of analog integration, small form-factor, integrated FLASH memory, and low power consumption make it ideal for flexible next generation battery charging applications. This note discussed how to use the C8051F300 in Lithium Ion battery charger applications at 510kHz switch rate. Example code is provided as well. Reference Applications of Linear Integrated Circuits. Eugene Hnatek, John Wiley and Sons, 1975. • Trickle Current = 135mA • Bulk Current = 250mA • Regulation Voltage = 4.2V • Termination Current = 125mA Efficiency of Charger • • • Switching Efficiency > 80% Voltage Accuracy > 1% Current Accuracy > 2% Charging 2 Cells or More 8 Rev. 1.3 AN146 Appendix A - Schematic Figure 5. High Speed Charger Schematic. Rev. 1.3 9 Value Package Notes C8051F300 MLP-11 Cygnal Integrated Products C8051F300 LPV321M5 SOT-23-5 National (LPV321M5 or equivalent) N-Channel SOT-23 Zetex, N-Channel 30-V (D-S) MOSFET, (2N7002CT or equivalent) P-Channel SOT-23 Zetex, P-Channel 30-V (D-S) MOSFET, (ZXMP3A13FCT-ND or equivalent) 22uH SMD Coil Craft Inductor, 22uH, 1.5 A, (DO3316P-223 or equivalent) Schottky SMC 3A 40V Power Rectifier Diode (MBRS340CT or equivalent) 0.1uF 0805 Cap X7R 50V 5% (Kemet C0805C104J5RACTU or equivalent) 33pF 0805 Cap X7R, 50V 10% (Kemet C0805330J5GACTU or equivalent) 1 uF 0805 Cap X7R, 10V 10% (Kemet C0805105K8RACTU or equivalent) 22uF EIA6032-28 Cap Tantalum, 16V, 10% (Kemet T491C226K016AS or equivalent) 100pF 0805 Cap X7R 50V 5% (Kemet C0805C101K5GACTU or equivalent) 100k 0805 Resistor 1/10W, 5% (Panasonic P100KCCT-ND or equivalent) 10k 0805 Resistor 1/10W, 5% (Panasonic P10.0KCCT-ND or equivalent) 1k 0805 Resistor 1/10W, 5% (Panasonic P1.0KCCT-ND or equivalent) 200 0805 Resistor 1/10W, 5% (Panasonic P200CCT-ND or equivalent) 20k 0805 Resistor 1/10W, 5% (Panasonic P20.0KCCT-ND or equivalent) Switch 6MM, SQ Momentary switch (Panasonic P8007S-ND or equivalent) 0.24 ohm 0805 Resistor 1/4W, 2% (Panasonic RL12T0.24GCT or equivalent) BOM TOTAL Included on Demo Board, but NOT Part of Battery Charger BOM Item QTY Part Value Package Notes 1 1 U2 MIC5235 SOT-23-5 Micrel Semiconductor (MIC5235-3.3M5) 2 1 C1 1 uF 0805 Cap X7R, 10V 10% (Kemet C0805105K8RACTU or equivalent) 3 1 C2 2.2 uF EIA3216-18 Cap Tantalum, 16V, 10% (Kemet T491A225K016AS or equivalent) 4 2 R1,R2 475 ohm 0805 Resistor 1/10W, 5% (Panasonic P475CCT-ND or equivalent) 5 2 R5,7 1k 0805 Resistor 1/10W, 5% (Panasonic P1.0KCCT-ND or equivalent) 6 1 R6 10k 0805 Resistor 1/10W, 5% (Panasonic P10.0KCCT-ND or equivalent) 7 1 D1 LED, Red 0.1" thru hole T-1 3/4 (Panasonic LN21RPHL or equivalent) 8 1 D2 LED, Green 0.1" thru hole T-1 3/4 (Panasonic LN31GPHL or equivalent) 9 1 Shunt Shunt 0.1" Shunt (929957-08 or equivalent) 10 2 JP1,JP2 1x2 Header 0.1" thru hole Sullins (S2105-02 or equivalent) 11 1 JP3 1x3 Header 0.1" thru hole Sullins (S2105-03 or equivalent) 12 1 JP4 2x5 Header 0.1" thru hole Protected with central polarizing key slot (3M 2510-6002UB or equiv.) 13 1 P1 RAPC722 2x5.5mm Jack Switchcraft (SC1153-ND or equivalent) 14 1 Board 2-Layer 2"x1.75" PCBEXPRESS (board manufacturing services) Item QTY Part 1 1 U1 2 1 U3 3 1 Q1 4 1 Q2 5 1 L1 6 2 D3,4 7 6 C3,5,7,9,10,12 8 1 C4 9 1 C6 10 1 C8 11 1 C11 12 2 R3,15 13 5 R4,10,11,13,14 14 1 R9 15 2 R8,16 16 1 R12 17 1 S1 18 1 RSENSE AN146 Appendix B - Bill Of Materials Figure 6. High Speed Charger Bill of Materials. 10 Rev. 1.3 AN146 Appendix C - PCB Layout Figure 7. High Speed Charger Layout (Silk Screen) JP3: Battery Input Terminal 3.3V LDO & Power LED JP1: 8V-15V Input Supply Terminal (+)(-) Battery(+) Temp Battery(-) C2 Interface Buck Regulator Sub-circuit C8051F300 Reset Switch Current and Voltage Feedback Monitoring Sub-circuits Rev. 1.3 11 AN146 Figure 8. High Speed Charger Layout (Top Layer) 12 Rev. 1.3 AN146 Figure 9. High Speed Charger Layout (Bottom Layer) Rev. 1.3 13 14 Value C8051F300 LPV321M5 N-Channel P-Channel 22uH Schottky 0.1uF 33pF 1 uF 22uF 100pF 100k 10k 1k 200 20k Switch 0.24 ohm Package MLP-11 SOT-23-5 SOT-23 SOT-23 SMD SMC 0805 0805 0805 EIA6032-28 0805 0805 0805 0805 0805 0805 6MM, SQ 0805 p ( Notes ) Cygnal Integrated Products C8051F300 National (LPV321M5 or equivalent) Zetex, N-Channel 30-V (D-S) MOSFET, (2N7002CT or equivalent) Zetex, P-Channel 30-V (D-S) MOSFET, (ZXMP3A13FCT-ND or equivalent) Coil Craft Inductor, 22uH, 1.5 A, (DO3316P-223 or equivalent) 3A 40V Power Rectifier Diode (MBRS340CT or equivalent) Cap X7R 50V 5% (Kemet C0805C104J5RACTU or equivalent) Cap X7R, 50V 10% (Kemet C0805330J5GACTU or equivalent) Cap X7R, 10V 10% (Kemet C0805105K8RACTU or equivalent) Cap Tantalum, 16V, 10% (Kemet T491C226K016AS or equivalent) Cap X7R 50V 5% (Kemet C0805C101K5GACTU or equivalent) Resistor 1/10W, 5% (Panasonic P100KCCT-ND or equivalent) Resistor 1/10W, 5% (Panasonic P10.0KCCT-ND or equivalent) Resistor 1/10W, 5% (Panasonic P1.0KCCT-ND or equivalent) Resistor 1/10W, 5% (Panasonic P200CCT-ND or equivalent) Resistor 1/10W, 5% (Panasonic P20.0KCCT-ND or equivalent) Momentary switch (Panasonic P8007S-ND or equivalent) Resistor 1/4W, 2% (Panasonic RL12T0.24GCT or equivalent) Included on Demo Board, but NOT Part of Battery Charger BOM Item QTY Part Value Package p ( Notes ) 1 1 U2 MIC5235 SOT-23-5 Micrel Semiconductor (MIC5235-3.3M5) 2 1 C1 1 uF 0805 Cap X7R, 10V 10% (Kemet C0805105K8RACTU or equivalent) 3 1 C2 2.2 uF EIA3216-18 Cap Tantalum, 16V, 10% (Kemet T491A225K016AS or equivalent) 4 2 R1,R2 475 ohm 0805 Resistor 1/10W, 5% (Panasonic P475CCT-ND or equivalent) 5 2 R5,7 1k 0805 Resistor 1/10W, 5% (Panasonic P1.0KCCT-ND or equivalent) 6 1 R6 10k 0805 Resistor 1/10W, 5% (Panasonic P10.0KCCT-ND or equivalent) 7 1 D1 LED, Red 0.1" thru hole T-1 3/4 (Panasonic LN21RPHL or equivalent) 8 1 D2 LED, Green 0.1" thru hole T-1 3/4 (Panasonic LN31GPHL or equivalent) 9 1 Shunt Shunt 0.1" Shunt (929957-08 or equivalent) 10 2 JP1,JP2 1x2 Header 0.1" thru hole Sullins (S2105-02 or equivalent) 11 1 JP3 1x3 Header 0.1" thru hole Sullins (S2105-03 or equivalent) 12 1 JP4 2x5 Header 0.1" thru hole Protected with central polarizing key slot (3M 2510-6002UB or equiv.) 13 1 P1 RAPC722 2x5.5mm Jack Switchcraft (SC1153-ND or equivalent) 14 1 Board 2-Layer 2"x1.75" PCBEXPRESS (board manufacturing services) Item QTY Part 1 1 U1 2 1 U3 3 1 Q1 4 1 Q2 5 1 L1 6 2 D3,4 7 6 C3,5,7,9,10,12 8 1 C4 9 1 C6 10 1 C8 11 1 C11 12 2 R3,15 13 5 R4,10,11,13,14 14 1 R9 15 2 R8,16 16 1 R12 17 1 S1 18 1 RSENSE AN146 Figure 10. High Speed Charger Bill of Materials Rev. 1.3 AN146 Figure 11. main() Flow Chart main() Config_F300() CalibrateADCfor Measurement() Enable Interrupts Infinite Loop Yes/No Clear Termination Flags Clear Charge Status Flags Yes Error Detected ? No Error Detected ? No Status = BULK ? Yes Turn on LED0 Turn off LED0, Error BULK_charge() Infinite Loop No Yes/No Status = LOWCURRENT ? Yes LOWCURRENT_charge() Rev. 1.3 15 AN146 Figure 12. CalibrateADCforMeasurement() Flow Chart CalibrateADCforMearurement() Setup ADC0's AMUX, Throughput, Gain, for near zero-scale voltage cal point Acquire 16-bit Measurement Setup ADC0's AMUX, Throughput, Gain, for near zero-scale Current cal point Setup ADC0's AMUX, Throughput, Gain, for near full-scale voltage cal point Acquire 16-bit Measurement Acquire16-bit Measurement Setup ADC0's AMUX, Throughput, Gain, for near full-scale Current cal point Calculate Voltage Slope Coefficient Acquire16-bit Measurement Calculate Voltage Offset Coefficient Calculate Current Slope Coefficient Erase Memory Page 0x1A00 Calculate Current Offset Coefficient Store Voltage Offset and Slope Coefficients in FLASH Memory Store Current Offset and Slope Coefficients in FLASH Memory END 16 Rev. 1.3 AN146 Appendix D - Software Flow Charts Figure 13. Monitor_Battery() Flow Chart Monitor_Battery() Measurement Type ? Current AMUX = Current Charge Voltage Temperature Battery Voltage AMUX = Volt Stop PWM Stop PWM AMUX = Temperature AMUX = Volt AV = 0 I=0 No I? Voltage w/ or w/out PWM Calculate Voltage w/ Calibration Coefficients Yes Current Calculate Current w/ Calibration Coefficients Temperature Calculate Temperature w/ Calibration Coefficients Start ADC0 No ADC0 Done? Yes AV = AV + ADC0 Turn PWM on Return Desired Parameter AV = AV/10 END Rev. 1.3 17 AN146 Figure 14. Bulk_Charge() Flow Chart (Part 1) Bulk_Charge() Start PWM w/ Zero Output Status = const_C T Within Limits ? No Yes V <max_V & > min_Bulk ? No Yes Set Appropriate Flags Calculate bulk_finish_time Green LED On Status = BULK & No Error? No Yes Status = const_c ? No Yes Regulate Battery Current Read Charge Voltage Yes Change Status from const_C to const_V B 18 C Charge Voltage Within Limits ? No A Rev. 1.3 D AN146 Figure 15. BULKCurrent() Flow Chart (Part 2) B C A No D Status = const_V & Current within Limits ? Reset Flags for Low Current Constant Current Mode Yes Regulate Voltage() Yes Time Overflow ? Stop PWM & Flag Error No Yes Stop PWM & Flag Error Temp. Overflow ? No No 60 Sec. Over ? Yes No const_V, NOT Delay & Current Below Threshold ? Yes Calculate bulk_finish_time Status = Delay Delay Time Over ? Yes No Stop PWM Status = const_C Status = LOWCURRENT Green LED Off END Rev. 1.3 19 AN146 Figure 16. LowCurrent_Charge() Flow Chart LOWCURRENT_charge() No ResetTimeBase() V <BulkThreshold ? Yes Prepare Flags to enter Bulk Mode Calculate Finish_time No No ERROR & LOWCURRENT =1 ? Yes Change Status from const_C to const_V Yes No Temp within Limits ? Charge Voltage Within Limits ? No Regulate Current Yes No No V <max_V ? Yes Yes Status = Delay Green LED Blinking Regulate Voltage Yes Green LED Off Stop PWM and flag error END 20 Status = const_V Rev. 1.3 Lowcurrent Finish Time reached? No AN146 Figure 17. Turn_PWM_Off() Flow Chart Turn_PWM_Off() No CEX1 Counter <0xF0? Yes Increment CEX1 Counter CEX1 Counter <0xF0? Yes No Disable PWM Mode END Rev. 1.3 21 AN146 Figure 18. Measure() Flow Chart measure() Set accumulator and counter i variables to zero Clear End of Conversion Flag Start New Conversion Conversion Complete ? No Yes accumulator = accumulator + ADC0 Increment i No i=0 ? Yes Return 16-bit Measurement END 22 Rev. 1.3 AN146 Figure 19. Regulate_Voltage() Flow Chart Regulate_Voltage() Measure Battery's voltage Voltage < VOLT_BULK & PCA not max ? No Yes No Make Duty Cycle Larger Voltage > VOLT_BULK & PCA not 0 Yes Make Duty Cycle Smaller Voltage < VOLT_BULK + Tolerence & > VOLT_BULK ? No Yes END Rev. 1.3 23 AN146 Figure 20. Regulate_Current() Flow Chart Regulate_Current() Measure Current Current < passed current & PCA not max ? No Yes No Make Duty Cycle Larger Current > passed current & PCA not 0 Yes Make Duty Cycle Smaller Current = passed value ? No Yes Monitor Voltage w/ PWM off Voltage < VOLT_LOWCURRENT ± Tolerence ? No Yes CHARGE_STATUS = const_V END 24 Rev. 1.3 AN146 Figure 21. PCA_OVERFLOW_ISR() Flow Chart PCA_OVERFLOW_ISR() Reset PCA Counter and PCA Interrupts Decrement time.count 0 = time.count No Yes Reset time.count to overflow value Increment time.sec No 60 = time.sec ? Yes Reset time.sec Increment time.min No LOW CURRENT charge & no errors ? No 60 = time.min ? Yes Increment time.hour Yes No odd second ? Reset time.min Yes Turn on LED No 24 = time.hour Yes Turn Off LED Reset time.hour END Rev. 1.3 25 AN146 Appendix E - Firmware (Header File) //----------------------------------------------------------------------------// // Copyright 2003 Cygnal Integrated Products, Inc. // // Filename: F300_HighSpeed_BC.h // Target Device: 8051F300 // Created: 1 MAR 2003 // Created By: DKC // Tool chain: KEIL Eval C51 // // This header file is used to define all preprocessor directives, prototypes, // and global variable for F300_HighSpeed_BC.c. // // The user should modify this header file before proceeding as key // battery parameter limits are set here. // //----------------------------------------------------------------------------// Function Prototypes //----------------------------------------------------------------------------void Config_F300(void); void Reset_Time_Base(void); void CalibrateADCforMeasurement(void); void Regulate_Current(int); void Regulate_Voltage(void); void Turn_PWM_Off(void); int Monitor_Battery(unsigned char); void Bulk_Charge(void); void Lowcurrent_Charge(void); unsigned int Measure(void); //----------------------------------------------------------------------------// UNIONs, STRUCTUREs, and ENUMs //----------------------------------------------------------------------------typedef union LONG { // byte-addressable LONG long l; unsigned char b[4]; } LONG; typedef union INT { int i; unsigned char b[2]; } INT; typedef struct { unsigned long int t_count; int sec; int min; int hour; }time_struct; // byte-addressable INT // global seconds // global minutes // global hour //----------------------------------------------------------------------------// Global Variable Definitions //----------------------------------------------------------------------------time_struct TIME; // Global Struct to Track Time 26 Rev. 1.3 AN146 char bdata TERMINATION; char bdata CHARGE_STATUS; INT code CHECK_BYTE _at_ LONG code VOLT_SLOPE _at_ LONG code VOLT_OFFSET _at_ LONG code I_NOAMP_SLOPE _at_ LONG code I_NOAMP_OFFSET _at_ LONG temp_LONG_1,temp_LONG_2; INT temp_INT_1,temp_INT_2; int Current = 0; int Voltage = 0; 0x1A00; 0x1A60; 0x1A64; 0x1A70; 0x1A74; // // // // // // // // // // // // Global Variable to Track Termination Global Variable to Track Charging 0x0A0A Default value, for later use Volt Slope Register Volt Offset Register Current Slope Register,ext. amp off Current Offset Register,ext. amp.off Temporary Storage Variables Temporary Storage Variables Most recent Current Measurement used to account for voltage drop across sense resistor //----------------------------------------------------------------------------// Bit maskable CHARGE STATUS Register Definition //----------------------------------------------------------------------------sbit BULK = CHARGE_STATUS^0; // bit 0 : BULK charge status bit sbit LOWCURRENT = CHARGE_STATUS^1; // bit 1 : LOWCURRENT charge status bit sbit ERROR = CHARGE_STATUS^2; // bit 2 : ERROR before/during charging sbit CONST_V = CHARGE_STATUS^3; // bit 3 : charged w/ constant VOLTAGE sbit CONST_C = CHARGE_STATUS^4; // bit 4 : charged w/ constant CURRENT sbit DELAY = CHARGE_STATUS^5; // bit 5 : BULK charge DELAY for LiIon // after CURRENT threshold detection sbit READY = CHARGE_STATUS^6; // bit 6 : Lowcurrent charge is // terminated; battery is charged sbit FREE1 = CHARGE_STATUS^7; // bit 7 : Not Currently used //----------------------------------------------------------------------------// Bit Maskable TERMINATION Register Definition //----------------------------------------------------------------------------sbit TEMP_MIN = TERMINATION^0; // bit 0 : minimum TEMPERATURE overflow sbit TEMP_MAX = TERMINATION^1; // bit 1 : maximum TEMPERATURE overflow sbit I_MIN = TERMINATION^2; // bit 2 : minimum CURRENT overflow sbit I_MAX = TERMINATION^3; // bit 3 : maximum CURRENT overflow sbit TIME_MAX = TERMINATION^4; // bit 4 : maximum time overflow sbit VOLT_MAX = TERMINATION^5; // bit 5 : maximum VOLTAGE overflow sbit VOLT_MIN = TERMINATION^6; // bit 6 : minimum VOLTAGE overflow sbit FREE2 = TERMINATION^7; // bit 7 : Not Currently used //----------------------------------------------------------------------------// Bit maskable PORT Definitions //----------------------------------------------------------------------------sbit LED0 = P0 ^ 2; // bit 2 : LED0, Pin P0.2 sbit CMPOUT = P0 ^ 3; // bit 3 : Comparator Output sbit CMPIN1 = P0 ^ 4; // bit 4 : Comparator + Input sbit CMPIN2 = P0 ^ 5; // bit 5 : Comparator - Input sbit CEX0 = P0 ^ 6; // bit 6 : Frequency Output Mode. sbit CEX1 = P0 ^ 7; // bit 7 : 8-bit PWM // AMUX Selections; Analog Inputs #define VBAT 0xF0; // bit 0 : Voltage Ch.; Analog In #define IBAT 0xF1; // bit 1 : Current Ch.; Analog In #define TBAT 0xF8; // bit 2 : Temp. Ch.; Analog In //----------------------------------------------------------------------------// 8051F300 PARAMETERS //----------------------------------------------------------------------------#define SYSCLK 24500000 // System clock frequency #define TEMP_SENSOR_GAIN 3300 // Temp Sensor Gain in (uV / degC) #define TEMP_GAIN 2 // PGA gain setting Rev. 1.3 27 AN146 #define #define #define #define #define INT_CURRENT_GAIN EXT_CURRENT_GAIN VREF SCRATCH_PAGE PWM_CLOCK 1 10 3300 0x1C00 SYSCLK/255 // // // // // PGA gain setting External gain setting ADC Voltage Reference (mV) FLASH page used for temp storage PWM frequency is 96 kHz //----------------------------------------------------------------------------// Calibration/Calculation PARAMETERS //----------------------------------------------------------------------------#define V1_CAL 67 // 1st cal point for 2 point cal. #define V2_CAL 2800 // 2nd cal point for 2 point cal. #define I1_CAL 33 // 1st cal point for 2 point cal. #define I2_CAL 2800 // 2nd cal point for 2 point cal. #define RSENSE 24 // RSENSE is default to 240mohm #define RESB 20 // 20k Ohms,Voltage Divide Resistor #define RESAB 30 // 30k Ohms,Sum of Divide Resistor #define TEMP_SLOPE ((long) TEMP_GAIN * TEMP_SENSOR_GAIN * 65536 / 100 / VREF) // An estimate of the Temperature<SLOPE> // in [tenth codes / K] // The temperature measurement is // within 3 degrees of accuracy. //----------------------------------------------------------------------------// Monitor_Battery Switch PARAMETERS //----------------------------------------------------------------------------#define TEMPERATURE 7 // Value for Switch Statement #define VOLTAGE 5 // Value for Switch Statement #define VOLTAGE_PWM_OFF 3 // Value for Switch Statement #define CURRENT 1 // Value for Switch Statement //----------------------------------------------------------------------------// Battery/Pack Parameters //----------------------------------------------------------------------------#define CELLS 1 // Number of cells in the battery pack #define CAPACITY 250 // mAh, Battery Capacity (LiIon) #define LiIon_CELL_VOLT 4200 // mV, Nominal Charge Voltage #define I_BULK (unsigned int)(CAPACITY) #define I_LOWCURRENT (unsigned int)(135) #define VOLT_BULK (unsigned int)(CELLS*LiIon_CELL_VOLT) #define VOLT_LOWCURRENT (unsigned int)(CELLS*LiIon_CELL_VOLT) #define #define #define #define (unsigned int)(CELLS*LiIon_CELL_VOLT/100)// 1 Percent Acc (unsigned int)(CAPACITY/10) // 10 Percent Acc 100 // Minium Battery Charging is 100 mA 1350 // Maximum Allowed Current to Protect Hardware VOLT_TOLERANCE CURRENT_TOLERENCE IMIN IMAX //----------------------------------------------------------------------------// Battery Characteristics: Charge TERMINATION Limits //----------------------------------------------------------------------------#define MIN_TEMP_ABS 26300 // Abs. min. TEMPERATURE = -10 C, 263K #define MAX_TEMP_ABS 35300 // Abs. max. TEMPERATURE = 70C, 323K: #define MIN_VOLT_BULK (unsigned int)(CELLS*LiIon_CELL_VOLT*2/3) // Minimum BULK Voltage #define MAX_VOLT_ABS (unsigned int)(CELLS * LiIon_CELL_VOLT) #define MIN_I_BULK (unsigned int)(125) #define MAX_TIME_LOWCURRENT 30 // Max Lowcurrent Charge Time = 90min #define MAX_TIME_BULK 90 // Maximum BULK Charge Time = 90 min 28 Rev. 1.3 AN146 #define BULK_TIME_DELAY END OF FILE 30 // at 1C CURRENT // DELAY = 30min after “MIN_I_BULK” Rev. 1.3 29 AN146 Appendix F - Firmware (Source File) //----------------------------------------------------------------------------// // Copyright 2003 Cygnal Integrated Products, Inc. // // Filename: F300_HighSpeed_BC.c // Target Device: 8051F300 // Created: 1 March 2003 // Created By: DKC // Tool chain: KEIL Eval C51 // // This is a stand alone battery charger for a Lithium ION battery. // It utilizes a buck converter, controlled by the on-chip 8-bit PWM, // to provide constant current followed by constant voltage battery charge. // The High Frequency Output Mode is used to generate the switch rate. // The default rate is 510 kHz. // //----------------------------------------------------------------------------// Includes //----------------------------------------------------------------------------#include <c8051f300.h> #include “F300_HighSpeed_BC.h” // Battery Hearder File //----------------------------------------------------------------------------// Functions //----------------------------------------------------------------------------void Config_F300(void) { RSTSRC = 0x02; XBR0 = 0x37; XBR1 = 0x90; XBR2 = 0x40; // // // // // // Enable VDD Monitor Skip P0.0,1,2,4,5; they’re analog In Enable P0.6, P0.7, as CEX0 and CEX1 Make CEX0 an 8-Bit PWM and CEX1 Frequency Output Mode Also, Enable crossbar and weak pull-ups CMPIN2 CMPIN1 = 1; = 0; // Make Comparator Output Initially low // to minimize current spikes on start-up P0MDOUT P0MDIN = 0xC8; = 0xC8; // Set P0.3,6,7 output to push-pull // Configure P0.0,1,2,4,5 as Analog Inputs OSCICN = 0x07; // Set SYSCLK to 24.5MHz, internal osc. ADC0CN = 0xC0; // Turn on the ADC Module; // enable low power mode for settling REF0CN = 0x0C; // Configure ADC’s to use VDD for // Voltage Reference, // Enable On-chip Temperature Sensor //---------------------------------------------------------------// Comparator Register Configuration //---------------------------------------------------------------CPT0MX = 0x22; 30 // Comparator 0 MUX Selection Register // P0.4,5 Input to Comparator Rev. 1.3 AN146 CPT0MD = 0x00; CPT0CN = 0x80; // P0.3 Output of Comparator // Comparator 0 Mode Selection Register // Comparator 0 Control Register, Turn on //----------------------------------------------------------------------------// PCA Configuration //----------------------------------------------------------------------------PCA0MD = 0x00; // Disable WDT PCA0MD = 0x08; // Set PWM Time base = SYSCLK PCA0L PCA0H = 0x00; = 0x00; // Initialize PCA Counter to Zero PCA0CN = 0x40; //Module PCA0CPM0 PCA0CPL0 PCA0CPH0 0 = 0x00; = 0x28; = 0x28; // Enable PCA Counter // Clear PCA Counter Overflow flag //Module PCA0CPM1 PCA0CPL1 PCA0CPH1 1 = 0x42; = 0xE0; = 0xE0; //Module 2 PCA0CPM2 = 0x49; PCA0CPL2 = 0xFF; PCA0CPH2 = 0x00; } EIE1 = 0x08; // // // // Configure CCM0 to Frequency Initialize PCA PWM to small 0x18 makes output frequency 0x28 makes output frequency Output Mode duty cycle ~510kHz ~306kHz // Configure CCM0 to 8-bit PWM mode // Initialize PCA PWM to small duty cycle // 0xB9 Ensures a Soft Initial Charge // // // // // // Configure Module 1 as software timer Initialize to 255 so that Interrupt is generated when PCA ends 8-bit PWM Cycle PCA0CPH is the high byte of the Output Compare Module // Enable PCA Overflow Interrupt //----------------------------------------------------------------------------// Reset_Time_Base - Resets all Time Counting Values //----------------------------------------------------------------------------void Reset_Time_Base() { TIME.sec = 0x00; TIME.min = 0x00; TIME.hour = 0x00; TIME.t_count = PWM_CLOCK; } //----------------------------------------------------------------------------// Initialize CalibrateADCforVoltageMeasurement //----------------------------------------------------------------------------// This function calibrates the voltage channel and stores the calibration // coefficients in the parameters volt_slope and volt_offset. // void CalibrateADCforMeasurement() // This calibration routine uses a 2 point cal. { unsigned char xdata *pwrite; // FLASH write pointer long i=0; Rev. 1.3 31 AN146 EA = 0; // Disable All Interrupts // Wait until 1st calibration voltage is ready for cal //while (SW0 == 1); // Wait until SW0 pushed for (i=0;i<100000;i++); // Wait for Switch Bounce // Once ready, Get the first calibration voltage AMX0SL = VBAT; // Select appropriate input for AMUX ADC0CF = (SYSCLK/5000000) << 3; // ADC conversion clock = 5.0MHz ADC0CF &=0xF8; // Clear any Previous Gain Settings ADC0CF |= 0x01; // PGA gain = 1 temp_INT_1.i = Measure(); // Wait until 2nd calibration voltage is ready for cal //while (SW0 == 1); // Wait until SW0 pushed //for (i=0;i<100000;i++); // Wait for Switch Bounce // Once ready, Get the 2nd calibration voltage AMX0SL = VBAT; //Change Mux for second point temp_INT_2.i = Measure(); // Calculate the SLOPE // V1 and V2 are in tenth of a degree temp_LONG_1.l = (unsigned)(temp_INT_2.i-temp_INT_1.i); temp_LONG_1.l *= (unsigned)100; // Account for Math Truncation Error temp_LONG_1.l /= (unsigned)(V2_CAL - V1_CAL); // Calculate the OFFSET temp_LONG_2.l = (unsigned)temp_INT_1.i; temp_LONG_2.l -= (signed)(temp_LONG_1.l * V1_CAL/100); temp_LONG_1.l = 2050; temp_LONG_2.l = 0; // If no cal. use these // as default values // Erased memory at page 0x1A00 pwrite = (char xdata *)&(CHECK_BYTE.b[0]); PSCTL = 0x03; // MOVX writes target FLASH memory; // FLASH erase operations enabled FLKEY = 0xA5; FLKEY = 0xF1; *pwrite = 0x00; // FLASH key sequence #1 // FLASH key sequence #2 // initiate PAGE erase // Write the Volt SLOPE and OFFSET to Flash PSCTL = 1; // MOVX writes to Flash pwrite = (char xdata *)&(VOLT_SLOPE.b[0]); FLKEY = 0xA5; FLKEY = 0xF1; // enable flash write *pwrite = temp_LONG_1.b[0]; pwrite = (char xdata *)&(VOLT_SLOPE.b[1]); FLKEY = 0xA5; FLKEY = 0xF1; // enable flash write *pwrite = temp_LONG_1.b[1]; pwrite = (char xdata *)&(VOLT_SLOPE.b[2]); FLKEY = 0xA5; FLKEY = 0xF1; // enable flash write *pwrite = temp_LONG_1.b[2]; 32 Rev. 1.3 AN146 pwrite = (char xdata *)&(VOLT_SLOPE.b[3]); FLKEY = 0xA5; FLKEY = 0xF1; // enable flash write *pwrite = temp_LONG_1.b[3]; pwrite = (char xdata *)&(VOLT_OFFSET.b[0]); FLKEY = 0xA5; FLKEY = 0xF1; // enable *pwrite = temp_LONG_2.b[0]; pwrite = (char xdata *)&(VOLT_OFFSET.b[1]); FLKEY = 0xA5; FLKEY = 0xF1; // enable *pwrite = temp_LONG_2.b[1]; pwrite = (char xdata *)&(VOLT_OFFSET.b[2]); FLKEY = 0xA5; FLKEY = 0xF1; // enable *pwrite = temp_LONG_2.b[2]; pwrite = (char xdata *)&(VOLT_OFFSET.b[3]); FLKEY = 0xA5; FLKEY = 0xF1; // enable *pwrite = temp_LONG_2.b[3]; PSCTL = 0; flash write flash write flash write flash write // MOVX writes target XRAM //----------------------------------------------------------------------------// Initialize CalibrateADCforCurrentMeasurement_NOAMP //----------------------------------------------------------------------------// This function calibrates the current channel with no external amp // and stores the calibration coefficients in the // parameters i_noamp_slope and i_noamp__offset. // // This calibration routine uses a 2 point cal. // Wait until calibration voltage is ready for cal //while (SW0 == 1); // Wait until SW0 pushed //for (i=0;i<100000;i++); // Wait for Switch Bounce // Once ready, Get the first calibration voltage AMX0SL = IBAT; // Select appropriate input for AMUX ADC0CF = (SYSCLK/5000000) << 3; // ADC conversion clock = 5.0MHz ADC0CF &=0xF8; // Clear any Previous Gain Settings ADC0CF |= 0x03; // Set PGA gain = 4 temp_INT_1.i = Measure(); // Acquire 16-bit Conversion temp_INT_1.i *= 2; // Account for Differential Mode // Wait until 2nd calibration voltage is ready for cal //while (SW0 == 1); // Wait until SW0 pushed //for (i=0;i<100000;i++); // Wait for Switch Bounce // Once ready, Get the 2nd calibration voltage temp_INT_2.i = Measure(); // Acquire 16-bit Conversion temp_INT_2.i *=2; // Account for Differential Mode // Calculate the temp_LONG_1.l = temp_LONG_1.l *= temp_LONG_1.l /= temp_LONG_1.l /= SLOPE (unsigned)(temp_INT_2.i - temp_INT_1.i); (unsigned)100; // Account for Math Truncation Error (unsigned)(I2_CAL - I1_CAL); (unsigned)INT_CURRENT_GAIN;// Account for Gain // Calculate the OFFSET Rev. 1.3 33 AN146 temp_LONG_2.l = (signed)(temp_INT_1.i/INT_CURRENT_GAIN); temp_LONG_2.l -= (signed)(temp_LONG_1.l * V1_CAL/100); temp_LONG_1.l = 2050; temp_LONG_2.l = 0; // If no cal. use these // as default values // Memory at 0x1A00 is already erased // Write the Volt SLOPE and OFFSET to Flash PSCTL = 1; // MOVX writes to Flash pwrite = (char xdata *)&(I_NOAMP_SLOPE.b[0]); FLKEY = 0xA5; FLKEY = 0xF1; // enable *pwrite = temp_LONG_1.b[0]; pwrite = (char xdata *)&(I_NOAMP_SLOPE.b[1]); FLKEY = 0xA5; FLKEY = 0xF1; // enable *pwrite = temp_LONG_1.b[1]; pwrite = (char xdata *)&(I_NOAMP_SLOPE.b[2]); FLKEY = 0xA5; FLKEY = 0xF1; // enable *pwrite = temp_LONG_1.b[2]; pwrite = (char xdata *)&(I_NOAMP_SLOPE.b[3]); FLKEY = 0xA5; FLKEY = 0xF1; // enable *pwrite = temp_LONG_1.b[3]; pwrite = (char xdata *)&(I_NOAMP_OFFSET.b[0]); FLKEY = 0xA5; FLKEY = 0xF1; // enable *pwrite = temp_LONG_2.b[0]; pwrite = (char xdata *)&(I_NOAMP_OFFSET.b[1]); FLKEY = 0xA5; FLKEY = 0xF1; // enable *pwrite = temp_LONG_2.b[1]; pwrite = (char xdata *)&(I_NOAMP_OFFSET.b[2]); FLKEY = 0xA5; FLKEY = 0xF1; // enable *pwrite = temp_LONG_2.b[2]; pwrite = (char xdata *)&(I_NOAMP_OFFSET.b[3]); FLKEY = 0xA5; FLKEY = 0xF1; // enable *pwrite = temp_LONG_2.b[3]; } PSCTL = 0; flash write flash write flash write flash write flash write flash write flash write flash write // MOVX writes target XRAM //----------------------------------------------------------------------------// Measure //----------------------------------------------------------------------------// // This routine averages 65536 ADC samples and returns a 16-bit unsigned // result. // unsigned int Measure (void) { unsigned i; // sample counter unsigned long accumulator=0L; // here’s where we integrate the // ADC samples 34 Rev. 1.3 AN146 // read the ADC value and add to running total i = 0; do { AD0INT = 0; // clear end-of-conversion indicator AD0BUSY = 1; // initiate conversion while(!AD0INT); // wait for conversion to complete accumulator += ADC0; // read adc value and accumulate i++; // update counter } while (i != 0x0000); } // the accumulator now contains 16 added bits of which 8 are usable return (unsigned int) (accumulator >> 8); //----------------------------------------------------------------------------// Regulate_Current //----------------------------------------------------------------------------// This routine monitors the battery’s current and adjusts // the PWM (i.e. duty cycle) to keep the current at a known value // void Regulate_Current(int passed_current) { unsigned int temp = 0,delay_count = 0; do{ temp = Monitor_Battery(CURRENT); // Measure Current if (temp < passed_current) { PCA0CPH1--; for(delay_count = 0;delay_count<2500;delay_count++); } if (temp > passed_current) { PCA0CPH1++; for(delay_count = 0;delay_count<2500;delay_count++); } }while ((temp < (passed_current - CURRENT_TOLERENCE)) || (temp > (passed_current + CURRENT_TOLERENCE))); // I_BULK or I_LOWCURRENT is set now temp = Monitor_Battery(VOLTAGE_PWM_OFF); // If VOLTAGE within range, // change from constant CURRENT charge // mode to constant VOLTAGE charge mode if ((temp >= (VOLT_LOWCURRENT - VOLT_TOLERANCE*2)) && (temp <= (VOLT_LOWCURRENT + VOLT_TOLERANCE*2))) { CONST_C = 0; CONST_V = 1; } } //----------------------------------------------------------------------------// Regulate_Voltage //----------------------------------------------------------------------------// This routine monitors the battery’s voltage and adjusts // the PWM (i.e. duty cycle) to keep the voltage at a known value // void Regulate_Voltage(void) { unsigned int temp = 0,delay_count = 0; Rev. 1.3 35 AN146 do{ temp = Monitor_Battery(VOLTAGE); // set VOLT_BULK (with “soft start”) if (temp < VOLT_BULK) { PCA0CPH1--; for(delay_count = 0;delay_count<2500;delay_count++); } if (temp > VOLT_BULK) { PCA0CPH1++; for(delay_count = 0;delay_count<2500;delay_count++); } } }while ((temp < (VOLT_BULK - VOLT_TOLERANCE)) || (temp > (VOLT_BULK + VOLT_TOLERANCE))); // VOLTAGE is set now //----------------------------------------------------------------------------// Turn_PWM_Off //----------------------------------------------------------------------------// This routine peforms a soft charge turn off by taking the PWM’s // duty cycle slowly to zero. // void Turn_PWM_Off(void) { do{ if (PCA0CPH1 < 0xF0) PCA0CPH1++; }while (PCA0CPH1 < 0xF0); // Duty Cycle is now small and safe to turn off. } PCA0CPM0 = 0x00; // Disable PWM //----------------------------------------------------------------------------// Monitor_Battery //----------------------------------------------------------------------------// This routine acts as a switch when gathering different conversion types. // It adjusts the throughput, adjust the AMUX and returns the current in mA, // voltage in mV, and temperature in C, 2% accurate. // int Monitor_Battery(unsigned char value) { char i; unsigned long av =0,delay_count=0; long signed result; ADC0CF = (SYSCLK/5000000) << 3; ADC0CF &= 0xF8; switch (value) { case TEMPERATURE: //Turn_PWM_Off(); AMX0SL = TBAT; ADC0CF |= 0x02; break; 36 // ADC conversion clock = 5.0MHz // Clear any Previous Gain Settings // Turn PWM Off // Select appropriate input for AMUX // Set PGA gain = 2 Rev. 1.3 AN146 case VOLTAGE: AMX0SL = VBAT; ADC0CF |= 0x01; break; // Select appropriate input for AMUX // Set PGA gain = 1 case VOLTAGE_PWM_OFF: //Turn_PWM_Off(); AMX0SL = VBAT; ADC0CF |= 0x01; break; // Turn PWM Off // Select appropriate input for AMUX // Set PGA gain = 1 case CURRENT: AMX0SL = IBAT; ADC0CF |= 0x01; break; // Select appropriate input for AMUX // Set PGA gain = 1 } //Compute average of next 10 A/D conversions for(delay_count = 0;delay_count<2500;delay_count++);// Allow Settling Time for(av=0,i=10;i;--i){ AD0INT = 0; // clear end-of-conversion indicator AD0BUSY = 1; // initiate conversion while(!AD0INT); // wait for conversion to complete av = av+ADC0; } av = av/10; av = av<<8; // Compute the average // Convert to 16-bit conversion // ...to account for 16-bit cal. // coefficients PCA0CPM0 = 0x46; // Turn on PWM switch (value) { case TEMPERATURE: result = (long) av * 1000/TEMP_SLOPE; break; } case VOLTAGE: case VOLTAGE_PWM_OFF: result = (av - VOLT_OFFSET.l); // Account for System Errors result *= 100; // Account for Math Truncation Error result *= RESAB; // Account for Divide Resistors result /= VOLT_SLOPE.l; // Convert to Voltage in Millivolts result /= RESB; result -= ((RSENSE*Current)/100); // Account for Sense Resistor Voltage Drop break; case CURRENT: result = (av - I_NOAMP_OFFSET.l); // Account for System Errors result *= 100; // Account for Math Truncation Error result *= 100; // Account for Sense Resistor result /= I_NOAMP_SLOPE.l; // Convert to Milliamps result /= RSENSE; // Account for Sense Resistor result /= EXT_CURRENT_GAIN; // Account for external Amplifier Current = (int) result; break; Rev. 1.3 37 AN146 } return (int) result; //----------------------------------------------------------------------------// Bulk_Charge Function //----------------------------------------------------------------------------void Bulk_Charge(void) { unsigned int temp = 0; unsigned int bulk_finish_hour = 0; unsigned int bulk_finish_min = 0; unsigned int delay_hour = 0; unsigned int delay_min = 0; unsigned int last_min = 0; Reset_Time_Base(); // Reset Time Base to zero // Calculate BULK charge finish time bulk_finish_min = (TIME.min + MAX_TIME_BULK); bulk_finish_hour = TIME.hour; while (bulk_finish_min > 60) { bulk_finish_min = bulk_finish_min - 60; bulk_finish_hour++; } CONST_C = 1; DELAY = 0; // Start in constant current charge mode // Reset timer DELAY temp = Monitor_Battery(TEMPERATURE); // Monitor Temperature // Is temperature within range? if ((temp > MIN_TEMP_ABS) && (temp < MAX_TEMP_ABS)) { temp = Monitor_Battery(VOLTAGE); // Monitor Voltage // Is Voltage within range? Voltage = temp; // for Debug if ((temp <= (MAX_VOLT_ABS + VOLT_TOLERANCE)) && (temp > MIN_VOLT_BULK)) { PCA0CPM0 = 0x46; // Turn on PWM // Enter main loop in Bulk_Charge() while ((BULK == 1) && (ERROR == 0)) { if (CONST_C == 1) Regulate_Current(I_BULK); else if (CONST_V == 1) { Current = Monitor_Battery(CURRENT); // Measure Current if((Current < IMIN)||(Current > IMAX)) { CONST_V = 0; // Exit CONST_V CONST_C = 1; // Prepare to enter CONST_C BULK = 0; // Prepare to exit BULK mode LOWCURRENT = 1; // Prepare to enter LOWCURRENT Mode if (Current < IMIN) I_MIN = 1; // Indicate Specific Error for Debug 38 Rev. 1.3 AN146 else I_MAX = 1; // Indicate Specific Error for Debug } else if ((Current < IMAX) && (Current > IMIN)) { I_MAX = 0; // Reset Error Flag I_MIN = 0; // Reset Error Flag Regulate_Voltage(); // Charge with Constant Voltage } } // Now, Check for error and charge termination conditions // If above max charge time, flag error // Test for BULK Charge Time Out // Monitor Time if ((TIME.hour == bulk_finish_hour) && (TIME.min == bulk_finish_min) && (DELAY == 0)) { Turn_PWM_Off(); // Turn Off PWM TIME_MAX = 1; // Set Time max error flag ERROR = 1; // Set general error flag } { // Monitor Temperature temp = Monitor_Battery(TEMPERATURE); if ((temp < MIN_TEMP_ABS) && (temp > MAX_TEMP_ABS)) Turn_PWM_Off(); if (temp < MIN_TEMP_ABS) TEMP_MIN = 1; else TEMP_MAX = 1; } ERROR = 1; // Turn Off PWM // Set Temperature below minimum flag // Set Temperature exceeds maximum flag // Set general error flag // Minute elapsed? // Check for minimum current // if reached, enter last DELAY charge if (TIME.min != last_min) { last_min = TIME.min; if ((CONST_V == 1) && (DELAY == 0) && (Monitor_Battery(CURRENT) <= MIN_I_BULK)) { // Calculate TOP OFF Battery Time finish time delay_min = (TIME.min + BULK_TIME_DELAY); delay_hour = TIME.hour; while (delay_min > 60) { delay_min = delay_min - 60; delay_hour++; } DELAY = 1; // Set Delay Flag } // Monitor Delay time, time up? Rev. 1.3 39 AN146 } } } } if ((TIME.hour == delay_hour)&&(TIME.min == delay_min) && (DELAY == 1)) { Turn_PWM_Off(); // Turn Off PWM CONST_V = 0; // Exit CONST_V CONST_C = 1; // Prepare to enter CONST_C BULK = 0; // Prepare to exit BULK mode LOWCURRENT = 1; // Prepare to enter LOWCURRENT Mode } else if(ERROR == 0) { if (temp > (MAX_VOLT_ABS + VOLT_TOLERANCE)) { VOLT_MAX = 1; // Set Max Voltage error flag ERROR = 1; // Set general error flag } else if(temp < MIN_VOLT_BULK) { VOLT_MIN = 1; // Set Minimum bulk voltage error flag LOWCURRENT = 1; // Switch to LOWCURRENT mode BULK = 0; // Exit Bulk Charge mode } // battery’s voltage very low } else if(ERROR == 0) { if (temp < MIN_TEMP_ABS) TEMP_MIN = 1; else TEMP_MAX = 1; } } // End Main While loop ERROR = 1; // Absolute temperature out of range? // Set Temperature below minimum flag // Set Temperature exceeds maximum flag // Set general error flag //----------------------------------------------------------------------------// Lowcurrent_Charge //----------------------------------------------------------------------------void Lowcurrent_Charge(void) { unsigned int temp = 0; unsigned int lowcurrent_finish_min = 0; unsigned int lowcurrent_finish_hour = 0; Reset_Time_Base(); // Reset Time base to zero // Calculate LOWCURRENT finish time lowcurrent_finish_min = (TIME.min + MAX_TIME_LOWCURRENT); lowcurrent_finish_hour = TIME.hour; while (lowcurrent_finish_min > 60) { lowcurrent_finish_min = lowcurrent_finish_min - 60; lowcurrent_finish_hour++; } 40 Rev. 1.3 AN146 // Enter Main Lowcurrent Loop. // Only exits are upon error and full charge while ((LOWCURRENT == 1) && (ERROR == 0)) { temp = Monitor_Battery(TEMPERATURE);// Get Temperature Reading // Is TEMPERATURE within limits if ((temp > MIN_TEMP_ABS) && (temp < MAX_TEMP_ABS)) { // Is Battery’s Charge Voltage below max charge voltage temp = Monitor_Battery(VOLTAGE); // Get Voltage Reading if (temp <= (VOLT_LOWCURRENT + VOLT_TOLERANCE)) { if (CONST_C == 1) // CONST_C ?, charge w/ constant current Regulate_Current(I_LOWCURRENT); if (CONST_V == 1) Regulate_Voltage(); // CONST_V?, charge w/ constant voltage if ((temp >= MIN_VOLT_BULK) && (DELAY == 0))// Bulk Threshold voltage met? { LOWCURRENT = 0; // Exit LOWCURRENT mode BULK = 1; // Switch to Bulk Charge mode } // Check elapsed time if ((TIME.hour == lowcurrent_finish_hour) && ( TIME.min == lowcurrent_finish_min)) { TIME_MAX = 1; // Set Time MAX error flag ERROR = 1; // Set general error flag } } else if(ERROR == 0) { VOLT_MAX = 1; ERROR = 1; } } else if(ERROR == 0) { if (temp < MIN_TEMP_ABS) TEMP_MIN = 1; else TEMP_MAX = 1; } } } ERROR = 1; // Voltage to high? // Set Max voltage error flag // Set general error flag // Absolute temperature out of range? // Set Temperature below minimum flag // Set Temperature exceeds maximum flag // Set general error flag //----------------------------------------------------------------------------// Main Function //----------------------------------------------------------------------------void main(void) { EA = 0; // Disable All Interrupts Reset_Time_Base(); Config_F300(); // Config F300 //Turn_PWM_Off(); // Turn PWM off before Calibration //CalibrateADCforMeasurement(); // Calibrate F300 Rev. 1.3 41 AN146 EA = 1; // Enable All Active Interrupts while(1) { LED0 = 0; // Turn LED0 off TERMINATION = 0x00; CHARGE_STATUS = 0x00; LOWCURRENT = 0; BULK = 1; CONST_C = 1; while (ERROR == 0) { if (BULK == 1) { Bulk_Charge(); } if (LOWCURRENT == 1) Lowcurrent_Charge(); } if (ERROR == 1) { Turn_PWM_Off();; EA = 0; while (1); } } } // Reset Termination Flags // Reset Charge Status Flags // Start in LOWCURRENT Charge Mode // Enter Bulk Charge Mode // Enter Lowcurrent_Charge function // Toggle LED0 at 1 Hz rate via ISR // // // // Turn PWM Off Disable All Interrupts Enter a eternal loop No recovery except “reset-button” //----------------------------------------------------------------------------// Interrupt Service Routines //----------------------------------------------------------------------------//----------------------------------------------------------------------------// PCA_ISR //----------------------------------------------------------------------------// This routine counts the elapsed time in seconds, minutes, hours. // It also toggles LED0 every second when in Lowcurrent Charge Mode. // This routine interrupts every time the PCA counter overflows, every 256 // SYSCLK cycles. After SYSCLK/256 interrupts, one second has elapsed. // void PCA_OVERFLOW_ISR (void) interrupt 9 { int time_sec; PCA0CN = 0x40; // Reset all PCA Interrupt Flags PCA0H = 0x00; if (0x0000 == --TIME.t_count) { TIME.t_count = PWM_CLOCK; if ( 60 == ++TIME.sec ) { TIME.sec = 0x00; if ( 60 == ++TIME.min ) 42 // Reset High Byte of PCA Counter // of 8-bit PWM we are using Module1 // Reset 1 Second Clock // Account for elapsed seconds // Reset second counter every minute // Account for elapsed minutes Rev. 1.3 AN146 { TIME.min = 0x00; if ( 24 == ++TIME.hour ) TIME.hour = 0x00; } } time_sec = TIME.sec; } } // Reset minute counter every hour // Account for elapsed hours // Reset hour counter every day if ((LOWCURRENT == 1) && (ERROR == 0)) { // Blink LED0 at 1 Hz if in Lowcurrent //if (TIME.sec % 2) // LED0 = 0; // Turn on LED every odd second //else // LED0 = 1; // Turn on LED every even second } END OF FILE Rev. 1.3 43 AN146 CONTACT INFORMATION Silicon Laboratories Inc. 400 West Cesar Chavez Austin, TX 78701 Tel: 1+(512) 416-8500 Fax: 1+(512) 416-9669 Toll Free: 1+(877) 444-3032 Please visit the Silicon Labs Technical Support web page: https://www.silabs.com/support/pages/contacttechnicalsupport.aspx and register to submit a technical support request. Patent Notice Silicon Labs invests in research and development to help our customers differentiate in the market with innovative low-power, small size, analogintensive mixed-signal solutions. Silicon Labs' extensive patent portfolio is a testament to our unique approach and world-class engineering team. The information in this document is believed to be accurate in all respects at the time of publication but is subject to change without notice. Silicon Laboratories assumes no responsibility for errors and omissions, and disclaims responsibility for any consequences resulting from the use of information included herein. Additionally, Silicon Laboratories assumes no responsibility for the functioning of undescribed features or parameters. Silicon Laboratories reserves the right to make changes without further notice. Silicon Laboratories makes no warranty, representation or guarantee regarding the suitability of its products for any particular purpose, nor does Silicon Laboratories assume any liability arising out of the application or use of any product or circuit, and specifically disclaims any and all liability, including without limitation consequential or incidental damages. Silicon Laboratories products are not designed, intended, or authorized for use in applications intended to support or sustain life, or for any other application in which the failure of the Silicon Laboratories product could create a situation where personal injury or death may occur. Should Buyer purchase or use Silicon Laboratories products for any such unintended or unauthorized application, Buyer shall indemnify and hold Silicon Laboratories harmless against all claims and damages. Silicon Laboratories and Silicon Labs are trademarks of Silicon Laboratories Inc. Other products or brandnames mentioned herein are trademarks or registered trademarks of their respective holders. 44 Rev. 1.3