Thesis Organization

In the following chapters, an integrated LED driver with fast reference tracking techniques including the CR and BSBR control mechanism are presented in this thesis. In

addition, the current regulator and current balance circuit are also discussed in this thesis. The Chapter 2 would describe the architecture of constant current regulator for LED Driver with the SAR-controlled adaptive off-time technique. The characteristics of LED backlight module and LCD panel are introduced in this chapter. In addition, the advantages and the disadvantages of these prior arts would be discussed in Chapter 2. In the Chapter 3, the analysis of the reference tracking procedure with FRT and CR technique is presented. The FRT technique is utilized for rapidly switching between two different output voltages and the CR technique is proposed for saving much power dissipation during the transition between two different output voltages. Furthermore, the stability and transient response of the LED driver with FRT technique is also discussed in this chapter. The circuit implementation composed of the voltage control current source (VCCS) compensator, the PWM generator, and the one-shot generator would be detailed illustrated in Chapter 3. In Chapter 4, a new charge recycling buck-store and boost-restore (BSBR) technique is proposed to reduce the power dissipation on the LED driver circuit and is applied on the boost converter to control the voltage, 9.3V for 4 series R-LED and 12.4V for G- or B- LED in the LED backlight module. The architecture of the proposed BSBR control, the BSBR tracking algorithm and the energy transforming efficiency are also discussed in detail. Experimental results shown in Chapter 5. Finally, conclusions are made in Chapter 6.

.

Chapter 2

The Architecture of Constant

Current Regulator for LED Driver

2.1 The Characteristic of LED

The LED I-V curve is shown in Figure 14. Because LED can be manufactured with smaller mismatch, the forward voltage variation of LED is expected. In addition, the forward voltage also varies with temperature and time. In order to get high quality image for LCD TV, it is impossible to regulate the forward voltage of LED to dimming the LED for changing the backlight brightness. In other words, the luminance is proportional to the level of driving current. The higher driving current will cause the higher brightness. As a result, by using the current to dimming the LED can prevent the variation of forward voltage and increase the brightness uniformity of LED backlight to get the high quality image of LCD TV.

Forward Current (mA)

Figure 14 I-V curve characteristic of LED

2.2 The Basic LED Current Regulators

The most common method to drive LED current is shown in Figure 15. A simple current regulator is implemented for LED strings. This circuit includes operational amplifier, a reference voltage, V_REF, and the value of the external resistor, R_EXT, to determine the LED current. It is uses the constant-current source to regulate LED strings [19-20]. The constant-current source eliminates LED current changes due to variations in forward voltage.

By using the constant-current source produces the constant LED brightness and strings uniformly. In Figure 15, LED can connect in a series and parallel to keep an identical current flowing in each LED, because the LED current ILED1~ ILEDn are produced by the value as VREF

/R_EXT. Therefore, if the external resistances are matched, this circuit can increase the current matching ability between channels.

Figure 15. A simplified current regulator for LED driver.

Another important issue is LED dimming control. LED dimming control is needed in many applications. In applications as LCD backlighting, dimming provides brightness and contrast adjustment. In general, two types of dimming methods can be achieved, analog and pulse width modulation (PWM) [21]. In analog dimming, the changing of LED’s forward current can change the brightness. For example, if an LED is at full brightness with 20 mA of

forward current, then 50% of the brightness is achieved by applying 50% of the maximum current to the LED. However, the drawback with analog dimming is that changes in forward current cause LED’s color shift. This color shift may become unacceptable in displays requiring a true color representation. On the other hands, PWM dimming is achieved by applying full current to the LED at a modulated duty cycle. The LED brightness is controlled by adjusting the relative duty cycle. For example, 50% brightness level is achieved by turning the LED on time at full current for 50% of each period. The advantage of PWM dimming is that the forward current is always constant, so we just have to decide the maximum current for all the LED strings. Instead of analog dimming, by using this method, LED color does not vary with brightness. In order to keep the human eyes from seeing the LED turn on and off, the switching speed must be above 100 Hz. Therefore, the proposed method also includes the PWM dimming control circuit to maintain the benefits of PWM dimming. In order to eliminate the inrush current occurred at the instance of string turn on, we also proposed a delay method to reduce it. By using the delay method, the on time of all the strings will be split into several parts. In other words, turning on the strings gradually can reduce the charge current at the moment. The proposed circuit not only balances the current for LED strings but also is suitable for PWM dimming control.

2.3 The Structure of LED Lighting System with HCC and PCC Technique

There is another structure of LED driver for lighting system. The design of the LED lighting system needs the regulated driving current technique to flow through the LED for uniform brightness. The prior arts of the LED driver are the hysteretic current control (HCC) and peak current control (PCC) techniques. The PCC technique uses a constant off-time to reduce the need for the connection of the sensing resistor in series with the LEDs, sacrificing

the accuracy. The PCC technique connects the sensing resistor at the source node of the N-type power MOSFET. As a result, the PCC technique has the advantage of high efficiency but low accuracy. Therefore, how to get high efficiency and accuracy at the same time becomes an important design issue in an LED lighting system. The implementation of the conventional PCC technique is shown in Figure 16(a). It includes an oscillator to periodically turn on the N-type power MOSFET M_N. When the inductor current is increased to the predefined peak current level, the N-type power MOSFET MN will be turned off. As a result, the inductor current is discharged by the freewheel-diode. The inductor current also flows through the LEDs; thus, the average inductor current will determine the brightness of the LEDs. Considering the inductor current waveform in the steady state as shown in Figure 16(b), the inductor current ripple ∆IL can be expressed as (5).

IN o o D

V_F and V_D are the forward voltages of the LED and the freewheel-diode, respectively. The V_o is equal to the summation of the total forward voltage of LED in series. The ton and toff are the on-time and off-time of the N-type power MOSFET M_N, respectively. Owing to the constant switching frequency, ton and toff can be approximately described as (6) and (7), respectively.

The forward voltage of freewheel-diode is ignored in (6).

o

respectively [22]. Since the bottom current level is determined by the switching frequency, the average inductor current I_L(avg.) can be calculated as (8).

( ) 1 sensing resistor R_s and the diode during phases I and II, respectively. In phase I, the inductor current passes through the external sensing resistor Rs and the equivalent resistance of the MN

(R_on) results in the energy consumption calculated as (9).

( )

2

( ) ( )

phI PCC L avg on s on

E = I × R + R × t

(9)

Moreover, the inductor current flows through the freewheel-diode to decrease the inductor current during phase II. Therefore, the freewheel-diode also brings the energy dissipation described as (10).

( ) ( )

phII PCC L avg D off

E = I × V × t

(10)

(a)

≈

≈ ≈

(b)

Figure 16. The prior art for LED lighting system. (a) The implementation of the LED driver with the PCC technique. (b) The inductor current waveform of the PCC technique.

The HCC technique utilizes two threshold current levels to accurately control the average inductor current I_L(avg), thereby defining the switching frequency [22]. The

implementation of the fundamental HCC technique is shown in Figure 17(a). It includes the simplest implementation of the current sensing circuit composed of an external resistor, R_S. The inductor current waveform in the steady-state is shown in Figure 17(b). The two threshold current levels are defined as I_Hth and I_Lth indicating the high and low threshold current levels, respectively. When the switch MS and the N-type power MOSFET MN are turned on, the inductor current increases at a rate determined by (V_IN – V_o)/L. Thus, the current sensing circuit generates the sensing current via R1 and the switch MS to produce the ramp voltage V_s. When the voltage V_s is higher than the V_REF, the switch M_S and the N-type power MOSFET MN are turned off by the output of the comparator. Therefore, Vs has a step variation since the sensing current passes through both the resistors R₁ and R₂, not only through R1. In addition, the inductor current is discharged via the freewheel-diode back to VIN, as such, V_s decays at a rate decided by the inductor current. The current ripple can be defined as (11):

The HCC technique can accurately design the low threshold current I_Lth and the inductor current ripple by defining the high threshold current IHth and utilize the suitable resistors R1

and R₂. Hence, the average inductor current can be accurately described as (12).

( )

As a result, the HCC technique can achieve a more accurate average current than that of the PCC technique. Hence, the brightness of the LED can be effectively controlled by the HCC technique. However, the large inductor current flows through the sensing resistor Rs in the full period since the whole switching period needs the information of the inductor current

to compare it with the two threshold current levels. Thus, the energy dissipation by the conduction loss in phase I and phase II can be expressed as (13) and (14), respectively

( )

2

( ) ( )

phI HCC L avg on s on

E = I × R + R × t

(13)

(

²

)

( ) ( ) ( )

phII HCC L avg D L avg s off

E = I × V + I × R × t

(14)

According to (9) and (10), the power consumption depends on the sensing resistor Rs and the forward voltage of freewheel-diode V_D because the R_s is much larger than R_on. However, comparing (10) and (14), the power efficiency of the PCC technique is generally higher than the HCC technique because the inductor current passes through the sensing resistor during phase I instead of the full period. Thus, the conduction loss of the HCC technique is larger than that of the PCC technique. There is a trade-off between accuracy and efficiency in the design of the LED driver. The HCC technique can have more accurate average current but larger conduction loss. On the other hand, the PCC technique can achieve higher efficiency at the expense of accuracy. To achieve higher accuracy and efficiency at the same time, the current regulator with the SAR-controlled adaptive off-time technique is proposed.

(a)

≈ ≈

≈

(b)

Figure 17. The prior art for LED lighting system. (a) The implementation of the LED driver with the HCC technique. (b) The inductor current waveform of the HCC technique.

2.4 The Successive Approximation Register (SAR) Design Methodology

The conduction loss of phase I due to the sensing resistor [7], [21] can be reduced by means of the on-chip low-side current sensing circuit that can accurately define the peak current level [23]. Additionally, the removal of the external sensing resistor can also save the cost and footprint area, but this causes an inability to sense inductor current during phase II [16]. Thereby, the average inductor current is difficult to accurately control because the bottom level of inductor current depend on the constant off-time or fixed frequency.

Therefore, the accuracy of the LED lighting system is deteriorated. The value of the off-time needs to be adaptively adjusted to ensure the accurate inductor current that can be controlled between the peak and bottom current levels. As a result, the average inductor current can be independent of the variation of the input voltage and the numbers of LED in series. To adaptively adjust the value of the off-time [24], it is proposed that the SAR-controlled adaptive off-time calibrate the off-time value. That is, the duration of the off-time can be adjusted to regulate the bottom current level.

The flow chart of the SAR-controlled adaptive off-time technique is shown in Figure 18.

The 8-bit SAR code A[7:0] is used to decide the duration of the off-time. At the beginning, the SAR code A[7:0] has an initial value of “1000,0000” and the gain code G[7:0] is set to

“0100,0000”. Adding or subtracting the gain code G[7:0] leads to the accurate calibration values of the SAR code A[7:0] in the following eight switching cycles. When the duration of the off-time is too short, the current sensing signal Vsense at the beginning of the next switching cycle is larger than the expected value of VREF2 as shown in Figure 19. Thus, the average inductor current is larger than expected value to have an influence on the brightness of the LED. At this time, the gain codes G[7:0] would be added to the SAR code A[7:0] to

prolong the off-time. On the other hand, if the off-time is too long, the current sensing signal, Vsense, at the beginning of the next switching cycle will be smaller than the V_REF2. That is, the lumen of LED is relatively small due to the smaller average inductor current, and therefore, the SAR code A[7:0] will subtract the gain code G[7:0].

START

Figure 18. The inductor current waveform at different the off-time values.

Figure 19. The inductor current waveform at different the off-time values.

After the eight switching cycles, the SAR code A[7:0] can be dynamically adjusted by a minimum value of the gain code G[7:0], which is “0000,0001” according to the value of the comparison result of the voltage Vsense and the reference voltage VREF2. Consequently, after the calibration duration, the adaptive off-time can ensure that the bottom level of the inductor

current will be close to the expected value. That is, the average inductor current can be independent of the variation of the input voltage.

The external sensing resistor is not needed any more since the on-chip low-side current sensing circuit can detect the peak current level and the SAR code A[7:0] adaptively adjusts the off-time. Therefore, the efficiency can be improved owing to the removal of the conduction loss of the external sensing resistor. Simultaneously, the accuracy is also guaranteed because the adaptive off-time can ensure the average inductor is independent of the variation of the input voltage. In other words, the SAR-controlled adaptive off-time has the advantages of high efficiency like the PCC technique and high accuracy like the HCC technique.

2.5 The Successive Approximation Register (SAR) Circuit Implementations

The implementation of the SAR-controlled adaptive off-time technique is illustrated in Figure 20. The external sensing resistor is replaced by the on-chip low-side current sensing circuit. As a result, the power dissipation of the on-chip low-side current sensing circuit is much smaller than that of the external sensing resistor. Furthermore, the freewheel-diode [25]

is substituted by the active diode, which is the P-type power MOSFET M_Diode, in order to reduce the power dissipation during phase II. The digital signals EN and CLR are used to enable the LED driver and reset the initial digital code. The digital signal, Digital PWM, is the dimming signal for adjusting the brightness of the LED.

Digital PWM sensing circuit with the blanking time circuit. During phase I, the N-type power MOSFET is turned on to increase the inductor current. The current sensing circuit thus increases the sensing voltage Vsense. Once Vsense is larger than VREF1, the output of the comparator ‘CMP2’

triggers the signal Reset to turn off the N-type power MOSFET M_N. The on-time duration t_on is decided. The peak level of inductor current can be expressed as:

×

1

K refers to the sensing ratio. At this particular time, the active diode M_Diode will be turned on to discharge the inductor current. The forward voltage of the active diode is the source-drain voltage V_SD of P-type power MOSFET. The off-time duration t_off is controlled by the 8-bit

SAR code A[7:0] from the SAR-controlled modulator. After the off-time duration, the adaptive off-time module sends a signal Set to turn on the N-type power MOSFET to charge the inductor current again.

At the beginning, the SAR-controlled adaptive off-time technique calibrates the off-time duration depending on the output ‘Count’ of the comparator ‘CMP1’. The signal ‘Count’

equaled to ‘1’ or ‘0’ means that the sensing current signal Vsense is higher or lower than the reference voltage V_ref2. After the calibration process, the bottom level of inductor current can be determined by the voltage VREF2. Thus, the bottom level of the inductor current can be

Therefore, the current ripple divided by the average current can be calculated as:

1 2

As a result, the current ripple ratio can be maintained by the reference voltages V_ref1 and Vref2 and would not be influenced by the input voltage or the numbers of LED in series.

Smaller current ripple has a more stable LED lumen. However, the ratio of the current ripple also defines the switching frequency as:

According to (18), a small inductor current ripple results in a faster switching frequency and increases power consumption. On the other hand, a high inductor current ripple results in

a slower switching frequency Therefore, power efficiency is increased because the switching loss is further reduced. However, the conduction power is increased due to large current ripple.

Therefore, the inductor current ripple, which is about ±15% of the average inductor current has a better power conversion efficiency as shown in Figure 21. The power efficiency would be reduced while the current ripple is smaller than ±15%. Moreover, the current ripple can be increased to reduce the size of the inductor at the cost of reduced efficiency and, possibly, LED lifetime [26]. As a result, the suitable ratio of the inductor current ripple is ±15%. This is popularly used in today’s LED lighting products.

Figure 21. The relation of the ratio of inductor current ripple (Iripple / Iavg), the power efficiency (η), and the switching frequency (fs).

The LED driver needs to provide an accurate driving current for LED lighting systems. It should be noted that the SAR-controlled modulator can still dynamically adjust the off-time to rapidly react to the variations of the input voltage and the numbers of LED in series. In addition, the power dissipation during phases I and II can be reduced and expressed as (19) and (20), respectively.

2

( ) ( )

phI SAR L avg on on

E = I × R × t

⁽¹⁹⁾

( ) ( )

phII SAR L avg SD off

E = I × V × t

⁽²⁰⁾

According to (15), the power dissipation in phase I is much less than those in (9) and (13). During phase II, the power dissipation can also be reduced through the use of small V_DS because of the active diode. Hence, the technique can minimize the conduction power loss and improve the accuracy of the driving current. The following subsections will describe the implementations of the sub-modules.

2.5.1 The Implementation of the SAR-Controlled Modulator

The implementation of the SAR-controlled modulator is composed of three sub-modules as shown in Figure 22. The three sub-modules are the up-down 8-bit counter, the 8-bit SAR gain code generator, and the over-control logic circuit. The up-down 8-bit counter is used to

The implementation of the SAR-controlled modulator is composed of three sub-modules as shown in Figure 22. The three sub-modules are the up-down 8-bit counter, the 8-bit SAR gain code generator, and the over-control logic circuit. The up-down 8-bit counter is used to

在文檔中 具有電荷儲存器和定電流機制的升壓電路用來改善RGB發光二極體背光模組效率 (頁 36-0)