Powering Passive OLED Displays in Handhelds

Feb. 1, 2006
OLEDs in display applications are usually arranged in a passive matrix. Specialized ICs can simplify the power design and the implementation of power management features for these passive OLED displays.

For the PDF version of this article, click here.

Organic light-emitting diode (OLED) displays are an emerging technology that promise to revolutionize the display industry. Based on organic materials that emit light when a current passes through them, OLEDs offer several advantages compared to the incumbent liquid-crystal display (LCD) technology. Among these is ease of manufacture, which will eventually lead to lower cost displays. Performance advantages include faster response time, wider viewing angles, lower power consumption, and brighter or higher-contrast images. One final advantage is that OLEDs are an emissive technology and require no backlight. This not only saves power, but allows for displays to be as thin as 1 mm.

Similar to LCDs, OLED displays come in both passive matrix and active matrix configurations. In a passive matrix OLED (PMOLED) display, the diodes are connected in a grid, each diode comprising an individual OLED pixel. The rows of the grid are lit one at a time using external drive circuitry. In contrast, active matrix displays include transistors within the display enabling pixels to be continuously illuminated. OLEDs, however, unlike LCDs, are current driven. This adds to the complexity of active matrix design and has led to the majority of the volume today consisting of the passive matrix variety. PMOLEDs are used in an array of applications, including cell phones, car stereos, MP3 players and other consumer products.

Powering OLED Displays

Because many OLED displays are used in portable applications, power consumption is extremely important. Any power IC must be designed to operate with the highest efficiencies and conserve as much power as possible to maximize battery life, especially when the display is not operating.

The power requirement for OLED displays depends on several factors. Because the display is current driven, the peak current requirement is dependant on the total number of pixels that need to be illuminated at one time and the maximum current with which they can be driven. Additional current is also consumed by the display drive electronics. The voltage required depends on the forward drop of the diodes, the drop across interconnects within the display — which tend to be quite resistive — and any drop developed across the activated display drivers (Fig. 1).

In this application, the maximum voltage needed is given by:

VIN = VDIODE + IDIODE (RCOL + RROW) + VCD + VRD (Eq. 1)

where VDIODE is the forward drop of the diode; IDIODE is the current in the diode; RCOL is the resistance of the column interconnection; RROW is the resistance of the row interconnection metal; VCD is the voltage developed across the column driver; and VRD is the overhead needed in the row driver.

In a typical application, VIN will be around 20 V. The peak current is given by:

IDIODE × NPIXELS + ICD + IRD (Eq. 2)

where IDIODE is the current in a typical diode; NPIXELS is the number of pixels lit at one time; ICD is the supply current to the column driver; and IRD is the supply current to the row driver.

Power Savings in Portable Displays

In portable equipment with LCD displays, it is common practice to turn off the backlight during a period of inactivity. This is then followed a few seconds later by the display being powered off completely. In contrast, OLED displays have no backlights, so the display typically is dimmed after a period of inactivity and then turned off some time later. Eq.1 shows that if the current is reduced in the display, the maximum voltage needed also is reduced. In a typical application where the supply voltage is constant, this extra voltage will be dropped across the column driver, leading to extra power dissipation and wasted energy. By reducing the supply voltage, this energy is no longer dissipated in the column driver and system efficiency is improved.

OLED Power ICs

Several new kinds of devices are now coming to market specifically for powering PMOLED displays in portable applications. The ideal device for this application must feature a very efficient boost converter capable of operating from the battery voltages used in portable applications, or from preregulated supplies within the device. Features like output load disconnect and low standby current are also important to reduce the drain on the battery when the display is not illuminated. The ideal device also requires few external components and a small package size to minimize the solution size in today's compact handheld devices. The complex control schemes used in the ISL97702 from Intersil represent an example of a power IC suited for this application. A typical circuit for this device is shown in Fig. 2.

Boost Converter

The boost used should operate from 2.4-V to 5.5-V supplies. This covers the full Li-ion input range and also provides operation from preregulated 3-V or 5-V rails. Output voltages required in this type of application can range from 12 V to 25 V. Another desirable feature is the full integration of the boost FET and Schottky diode, reducing the need for external components. As an example, the ISL97702 incorporates a 1.2-A FET, supporting output voltages up to 28 V with efficiencies of up to 90%.

For optimum operation of the boost circuit, correct component selection is important. The main components to be considered are the inductor and output capacitor, which affect the stability of the boost control loop. Some boost converters use external compensation, which also requires correct selection of the compensation components. However, the ISL97702 features an internal compensation network. Such designs require the inductor and capacitor values to be within a certain range, and tables in the data sheet are usually provided to aid in component selection.

The inductor value also affects the inductor size. The ISL97702 is designed to work with inductors as low as 3.3 µH for small component size. However, low inductor values may cause the device to operate in discontinuous mode, which may lead to higher output ripple. Preferably, a value should be chosen to maintain continuous mode of operation. The inductor chosen must also be selected to handle the peak and average currents needed by the application. These values are given by the following equations:

where ΔIL is the peak-to-peak inductor current ripple in amps; L is the inductor value in henrys; and fOSC is the switching frequency in hertz.

The output capacitor should be chosen to maintain stable operation of the boost loop. Higher values of output capacitance provide lower output voltage ripple. Selection is made based on the tradeoff between ripple and component cost.

The input side capacitor is used to isolate the input supply current from the switching currents through the inductor. For this application, values in the 10-µF to 15-µF range are recommended.

Dual-Output Voltage Selection

As previously mentioned, significant power savings can be achieved by reducing the output voltage when the OLED is operated in dim mode. This is implemented using two separate feedback paths, which can be selected by the use of a simple logic input. This enables simple support of the bright-dim-off power-saving technique used with PMOLEDs.

The output voltage is set using a potential divider connected from the output pin to the feedback reference pin. This feedback voltage VFB is compared to an internally set reference to control the output voltage. The accuracy of the output voltage depends on the accuracy of the feedback reference and the resistor values used in the feedback network.

The ISL97702, for example, has a feedback voltage set to 1.15 V ±2%. When the select pin (SEL) is set low, the FB0 feedback pin is compared to the reference and the FB1 pin is grounded to provide a feedback ground reference. When SEL is high, FB1 is used as a reference and FB0 is grounded. The output voltages for the two paths are calculated as in Eqs. 6 and Eq. 7:

Fault Detection

It also is useful to integrate several protection circuits to ensure that both the IC and external components are protected. For example, undervoltage lockout ensures the device only operates when the input voltage is above the minimum required for correct operation. Overvoltage lockout stops the device from operating if the output voltage exceeds the maximum allowed for the device. For current, overcurrent protection monitors the switching currents and limits these to the maximum allowed with the device. Thermal protection is provided by overtemperature protection, which generates a shutdown when die temperature exceeds a preset maximum.

Clock Synchronization

In portable equipment especially, clock noise and cross talk can become a major concern. The ability to synchronize a switching device to an external clock enables the product designer to reduce these issues by locking all clocks to a single frequency. For applications where this is not a concern, the power circuit should also be capable of self-clocking. High-clocking frequencies in the 1-MHz range typically provide the best efficiencies while offering small component size. These devices should self-clock at 1 MHz, but are synchronizable to an external clock between 600 kHz and 1.4 MHz just by connecting that clock to a sync input pin.

Soft-Start Control

When a power IC first starts operating, the current needed to charge the capacitance in the system can produce significant input current requirements. If this current is too high, the battery voltage can drop, leading to devices in the system entering reset, or causing erratic operation.

To overcome this, a soft-start scheme is used to limit the current at startup. The current capability of the IC is slowly raised until full current capability is reached. Such schemes are typically used in many of today's boost converters.

Input Voltage Disconnect

To improve battery life, an integrated disconnect switch on the input side of the boost circuit offers a major advantage. When the device is disabled, this switch opens to disconnect the OLED display, the drivers and the feedback networks so that no leakage currents can flow. In this power-down mode, the internal IC power consumption should also be reduced to a minimum.

When the device is enabled and the load is connected to the output, a dc path is generated from input to output and a large current spike can flow as the output capacitance is charged. The ISL97702 disconnect switch also features a soft-start mode that limits current while the output capacitance is charged, enhancing the soft-start schemes found in other dc-dc converters.

Fig. 3 shows the soft-start operation of the ISL97702. During interval A, the current through the disconnect switch is limited to reduce in-rush current as the load capacitors are charged. During interval B, the boost converter starts with the current limit set to 25% of the maximum allowed current. During both intervals C and D, the current limit is increased by 25%, and reaches 100% in interval E.

Power Solutions

Power solutions for handheld devices are becoming more critical as suppliers continue to offer more features and consumers continue to demand longer battery life. OLED displays are just one of the new technologies generating requirements for specific power ICs with increased functionality. The use of these ICs can aid designers in meeting the physical and power management challenges of OLED displays.

Sponsored Recommendations

Comments

To join the conversation, and become an exclusive member of Electronic Design, create an account today!