EL4093 ® Data Sheet January 1996, Rev B FN7159 300MHz DC-Restored Video Amplifier Features The EL4093 is a complete DCrestored video amplifier subsystem, featuring low power consumption and high slew rate. It contains a current feedback amplifier and a sample and hold amplifier designed to stabilize video performance. When the HOLD logic input is low, the sample and hold may be used as a general purpose op amp to null the DC offset of the video amplifier. When the HOLD input goes high the sample and hold stores the correction voltage on the hold capacitor to maintain DC correction during the subsequent video scan line. • High accuracy DC restoration for video The sample and hold amplifier contains a current output stage that greatly simplifies its connection to the video amplifier. Its high output impedance also helps to preserve video linearity at low supply voltages. For ease of interfacing, the HOLD input is TTL-compatible. This device has an operational temperature of -40°C to +85°C and is packaged in plastic 16-pin DIP and 16-pin SOIC. • TTL-compatible HOLD logic input • Low supply current of 9.5mA typ. • 300MHz bandwidth • 1500V/µs slew rate • 0.04% differential gain and 0.02° differential phase into 150Ω for NTSC • 1.5mV max. restored DC offset • Sample and hold amplifier with fast enable and low leakage Applications • Input amplifier in video equipment • Restoration amplifier in video mixers Ordering Information Pinout PART NUMBER EL4093 (16-PIN PDIP, SO) TOP VIEW TEMP. RANGE PACKAGE PKG. NO. EL4093CN -40°C to +85°C 16-Pin PDIP MDP0031 EL4093CS -40°C to +85°C 16-Pin SOIC MDP0027 Demo Board A demo PCB is available for this product. Request “EL4093 Demo Board.” 1 CAUTION: These devices are sensitive to electrostatic discharge; follow proper IC Handling Procedures. 1-888-INTERSIL or 321-724-7143 | Intersil (and design) is a registered trademark of Intersil Americas Inc. Copyright © Intersil Americas Inc. 2003. All Rights Reserved. Elantec is a registered trademark of Elantec Semiconductor, Inc. All other trademarks mentioned are the property of their respective owners. EL4093 Absolute Maximum Ratings (TA = 25°C) VS V+ to V- Supply Voltage . . . . . . . . . . . . . . . . . . . . . . . . 12.6V VHOLDVoltage at HOLD input (DGND-0.7) to (DGND+5.5V) VIN Voltage at any other input . . . . . . . . . . . . . . . . . . . . . V+ to V∆VIN Difference between Sample and Hold inputs . . . . . . . . . .±8V IOUT1 Video amplifier output current . . . . . . . . . . . . . . . . . . . ±30mA IOUT2 IIN PD TA TJ TST S/H amplifier output current . . . . . . . . . . . . . . . . . . . . ±10mA Maximum current into other pins. . . . . . . . . . . . . . . . . . ±6mA Maximum Power Dissipation . . . . . . . . . . . . . . . . See Curves Operating Ambient Temperature Range . . . . .-40°C to +85°C Operating Junction Temperature. . . . . . . . . . . . . . . . . . 150°C Storage Temperature Range. . . . . . . . . . . . .-65°C to +150°C CAUTION: Stresses above those listed in “Absolute Maximum Ratings” may cause permanent damage to the device. This is a stress only rating and operation of the device at these or any other conditions above those indicated in the operational sections of this specification is not implied. IMPORTANT NOTE: All parameters having Min/Max specifications are guaranteed. Typical values are for information purposes only. Unless otherwise noted, all tests are at the specified temperature and are pulsed tests, therefore: TJ = TC = TA Open-Loop DC Electrical Specifications Power supplies at ±5V, TA = 25°C Sample and Hold PARAMETER DESCRIPTION MIN TYP MAX UNITS IS,HOLD Total Supply current in HOLD mode 9.5 11.5 mA IS,SAMPLE Total Supply current in SAMPLE mode 8.5 10.5 mA TYP MAX UNITS Video Amplifier Section (Not Restored) PARAMETER DESCRIPTION MIN VOS Input Offset Voltage 10 110 mV IB+ Non-Inverting Input Bias Current 10 25 µA IB- Inverting Input Bias Current 15 50 µA ROL Transimpedance, VOUT = ±2.5V, RL = 150Ω 150 400 kΩ VO Output Voltage Swing, RL = 150Ω ±3 ±3.5 V ISC Output Short-Circuit Current 60 100 mA Open-Loop DC Electrical Specifications Power supplies at ±5V, TA = 25°C Sample and Hold Section PARAMETER DESCRIPTION MIN TYP MAX UNITS 0.5 1.5 mV VOS Input Offset Voltage TCVOS Average Offset Voltage Drift 6 IB Input Bias Current 1 2 µA IOS Input Offset Current 10 200 nA TCIOS Average Offset Current Drift 0.1 nA/°C VCM Common Mode Input Range ±2.5 ±2.8 V gM Transconductance (RL = 500Ω) 5 15 A/V CMRR Common Mode Rejection Ratio (VCM -2.5V to +2.5V) 70 90 dB VIL HOLD Logic Input Low (referenced to Digital GND) VIH HOLD Logic Input High (referenced to Digital GND) 2.0 VGND Digital GND Reference Voltage (V-) IDROOP Hold Mode Droop Current ICHARGE Charge Current Available to CHOLD ±5.5 ±8.5 mA VO Output Voltage Swing (RL = 10kΩ) ±3 ±3.5 V IO Output Current Swing (RL = 0Ω) ±4.5 ±5.5 mA 2 µV/°C 0.8 V V 10 (V+) - 4.0 V 70 nA EL4093 Closed-Loop AC Electrical Specifications Power supplies at ±5V, TA = 25°C, RF = RG = 750Ω, RL = 150Ω, CL = 5pF, CIN(parasitic) = 1.8pF Video Amplifier Section PARAMETER DESCRIPTION MIN TYP MAX UNITS BW, -3dB -3dB Small-Signal Bandwidth 300 MHz BW, ±0.1dB 0.1dB Flatness Bandwidth 50 MHz Peaking Frequency Response Peaking 0 dB SR Slew rate, VOUT between -2V and +2V 1500 V/µs dG Differential Gain Error, Voffset between -714mV and +714mV 0.04 % dθ Differential Phase Error, Voffset between -714mV and +714mV 0.02 ° Closed-Loop AC Electrical Specifications Power supplies at ±5V, TA = 25°C, RF = RG = 750Ω, RL = 150Ω, CL = 5pF, CHOLD = 2.2nF Sample and Hold Section PARAMETER DESCRIPTION MIN TYP MAX UNITS ∆ISTEP Change in Sample to Hold Output Current Due to Hold Step 0.1 µA ∆TSH Sample to Hold Delay Time 15 ns ∆THS Hold to Sample Delay Time 40 ns TAC Settling Time to 1% (DC Restored Amplifier Output) Video Amplifier Input from 0 to 1V 2.2 µs Typical Application 3 EL4093 Typical Performance Curves Non-inverting Frequency Response (Gain) Inverting Frequency Response (Gain) Frequency Response for Various RF and RG 4 Non-inverting Frequency Response (Phase) Inverting Frequency Response (Phase) Frequency Response for Various CIN Frequency Response for Various RL Frequency Response for Various CL 3dB Bandwidth vs Temperature (Video Amp) EL4093 Typical Performance Curves (Continued) Peaking vs Temperature (Video Amp) Output Voltage Swing vs Frequency Voltage and Current Noise vs Frequency Input Offset Voltage vs Die Temperature (Video Amp, 3 Sample) 5 2nd and 3rd Harmonic Distortion vs Frequency Supply Current vs Temperature Input Bias Current vs Temperature (Video Amp) Transimpedance vs Temperature (Video Amp) EL4093 Typical Performance Curves (Continued) Input Offset Voltage vs Die Temperature (Sample & Hold, 3 Samples) Input Bias Current vs Die Temperature (Sample & Hold) Transconductance vs Die Temperature (Sample & Hold) Droop Current vs Temperature (Sample & Hold) 6 Transconductance vs Temperature (Sample & Hold) Output Current Swing vs Temperature (Sample & Hold) Charge Current vs Temperature (Sample & Hold) Hold Step (∆IOUT) vs Temperature EL4093 Typical Performance Curves (Continued) Differential Gain and Phase vs DC Input Voltage at 3.58MHz Differential Gain and Phase vs DC Input Voltage at 3.58MHz Small-Signal Step Response Settling Time vs Settling Accuracy (Video Amp) Large-Signal Step Response Maximum Power Dissipation vs Ambient Temperature, 16-Pin PDIP Package Applications Information Product Description The EL4093 is a high speed DC-restore system containing a current feedback amplifier (CFA) and a sample & hold (S/H) amplifier. The CFA offers a wide 3dB bandwidth of 300MHz and a slew rate of 1500V/µs, making it ideal for high speed video applications such as SVGA. The CFA’s excellent differential gain and phase at 3.58MHz also makes it suitable for NTSC applications. Drawing only 9.5mA on ±5V supplies, the EL4093 serves as an excellent choice for those applications requiring both low power and high bandwidth. 7 Slew Rate vs Die Temperature (Video Amp) Maximum Power Dissipation vs Ambient Temperature, 16-Pin SO Package The connection between the CFA and sample & hold (the Autozero interface) has been greatly simplified. The output of the sample & hold is a high impedance current source, allowing direct connection to the CFA inverting input for autozero purposes. In addition, special circuitry within the sample & hold provides a charge current of 8.5mA in sample mode, resulting in a sample hold current ratio (ratio of charging current to droop current) of approx. 1,000,000. Theory of Operation In video applications, DC restoration moves the backporch or black level to a fixed DC reference. The EL4093 uses a CFA in feedback with a sample & hold to provide DC restoration. EL4093 Figure 1 shows how the two are connected to provide this function; the S/H compares the output of the CFA to a DC reference, and any difference between them causes an output current from the S/H. This “autozero” current is fed to the CFA inverting input, the effect of which is to move the CFA output towards the reference voltage. This autozero mechanism settles when the CFA output is one VOS away from the reference (the VOS here refers to the S/H offset voltage). bypassed to reduce the risk of oscillation. In the EL4093 there are two sets of supply pins: V+1/V-1 provide power for the CFA, and V+2/V-2 are for the S/H amplifier. Good performance can be achieved using only one set of bypass capacitors, although they must be close to the V+1/V-1 pins since that is where the high frequency currents flow. The combination of a 4.7µF tantalum capacitor in parallel with a 0.01µF capacitor has been shown to work well. Chip capacitors are recommended for the 0.01µF bypass to minimize lead inductance. For good AC performance, parasitic capacitance should be kept to a minimum, especially at the CFA inverting input. Ground plane construction should be used, but it should be removed from the area near the inverting input to minimize any stray capacitance at that node. Chip resistors are recommended for RF and RG, and use of sockets should be avoided if possible. Sockets add parasitic inductance and capacitance which will result in some additional peaking and overshoot. FIGURE 1. The autozero mechanism is typically active for only a short period of each video line. Figure 2 shows a NTSC video signal along with the EL4581 back porch output. The back porch signal is used to drive the HOLD input of the EL4093, and we see that the EL4093 is in sample mode for only 3.5µs of each line. It is during this time that the autozero mechanism attempts to drive the CFA output towards the reference voltage, at the same time putting a correction voltage onto the hold capacitor CHOLD. During the rest of the line (60µs) the EL4093 is in hold mode, but DC correction is maintained by the voltage on CHOLD. If the CFA is configured for non-inverting gain, then one should also pay attention to the trace leading to the +input. The inductance of a long trace (> 3’) can form a resonant network with the amplifier input, resulting in high frequency oscillations around 700MHz. In such cases a 50Ω–100Ω series resistor placed close to the +input would isolate this inductance and damp out the resonance. Capacitance at the Inverting Input Any manufacturer’s high-speed voltage or current feedback amplifier can be affected by stray capacitance at the inverting input. For inverting gains this parasitic capacitance has little effect because the inverting input is a virtual ground, but for non-inverting gains this capacitance (in conjunction with the feedback and gain resistors) creates a pole in the feedback path of the amplifier. This pole, if low enough in frequency, has the same destabilizing effect as a zero in the forward open-loop response. Hence it is important to minimize the stray capacitance at this node by removing the nearby ground plane. In addition, since the S/H output connects to this node, it is important to minimize the trace capacitance. Good practice here would be to connect the two pins with a short trace directly underneath the chip. Feedback Resistor Values FIGURE 2. Power Supply Bypassing and Printed Circuit Board Layout As with any high frequency device, good printed circuit board layout is necessary for optimum performance. Ground plane construction is highly recommended. Lead lengths should be as short as possible. The power supply pins must be well 8 The EL4093 has been optimized for a gain of +2 with RF = 750Ω. This value of feedback resistor gives a 3dB bandwidth of 300MHz at a gain of +2 driving a 150Ω load. Since the amplifier inside the EL4093 uses current mode feedback, it is possible to change the value of RF to adjust the bandwidth. Shown in the table below are optimum feedback resistor values for different closed loop gains. EL4093 come from the S/H output. Since the maximum that IAZ can be is 5.5mA, we can solve for VDC using the following: GAIN OPTIMUM RF BW (MHz) PEAKING (dB) +1 910 314 0.2 +2 750 300 0 +5 470 294 0.2 -1 680 300 0 Autozero Interface The autozero interface refers to the connection between the S/H output and the CFA inverting input. This interface has been greatly simplified compared to that of the EL2090, in that the S/H output is a high impedance current source. The S/H output can be connected directly to the inverting input, and its high impedance greatly reduces the interaction between the sample & hold and the gain setting resistors. Another virtue of this interface is better gain linearity as the autozero current changes. For example, at an autozero current of 0mA the output impedance is about 5MΩ, dropping to 1MΩ as the autozero current increases to 3mA. Using RF = RG = 750Ω, the closed loop gain changes only by 0.025% in this interval. V DC I AZ = ± 5.5mA = 2 --------------- 750Ω and see that VDC = ±2V. This range can easily accommodate most video signals. As another example, consider the case where we are restoring to a reference voltage of +0.75V. Using the same reasoning as above, a current IRF = (VDC - 0.75V)/RF must flow through RF, and a current IRG = VDC/RG must go into RG. Again, our boundary condition is that IRF + IRG ≤ ±5.5mA, and we can solve for the allowable VDC values using the following: V DC – 0.75V V DC ± 5.5mA = --------------------------------- + --------------750Ω 750Ω Hence VDC must be between +2.4V to -1.7V. This example illustrates that when the reference changes, the autozero range also changes. In general, the user should determine the autozero range for his/her application, and ensure that the input signal is within this range during the autozero period. Autozero Loop Bandwidth Autozero Range The autozero range is defined as the difference between the input DC level and the reference voltage to restore to. The size of this range is a function of the gain setting resistors used and the S/H output current swing. For a gain of +2 the optimum feedback resistor is 750Ω, and the available S/H output current is ±5.5mA minimum. To determine the autozero range for this case, we refer to Figure 3 below. The gain-bandwidth product (GBWP) of the autozero loop is determined by the size of the hold capacitor, the value of RF, and the transconductances (gm’s) of the S/H amplifier. To begin, the S/H amplifier is modeled as in Figure 4. First, the input stage transconductance is represented by gm1, with the compensation capacitor given by CHOLD. This stage’s GBWP is thus gm1/(2π • CHOLD) = 1/(2π • (350Ω)(2.2nF)) = 207kHz. Next, since the S/H has a current output, its output stage can be modeled as a transconductance gm2, in this case having a value of 1/(500Ω). The current from gm2 then flows through the I to V converter made up of the CFA and RF to produce a voltage gain. Thus the GBWP of the overall loop is given by: gm1 GBWP = --------------------------------- ( gm2 × R F ) 2π × C HOLD FIGURE 3. Suppose that the input DC level is +VDC, and that the reference voltage is 0V. We know that in feedback, the following two conditions will exist on the CFA: first, its output will be equal to 0V (due to autozero), and second, its VINvoltage is equal to the VIN+ voltage (i.e. VIN- = +VDC). So we have a potential difference of +VDC across both RF and RG, resulting in a current IRF = IRG = VDC/750Ω that must flow into each of them. This current IAZ = (IRF + IRG) must 9 EL4093 FIGURE 4. With RF = 750Ω, a GBWP of 310kHz is obtained. Note however that this is the small signal GBWP. As mentioned earlier, the sample and hold has special boost circuits built in which provides ±8.5mA of charge current during full slew. These boost circuits turn on when the S/H input differential voltage exceeds ±50mV. When the boosters are turned on, gm1 greatly increases and the circuit becomes nonlinear. Thus some stability issues are associated with the boosters, and they will be addressed in a later section. Charge Injection and Hold Step Charge injection refers to the charge transferred to the hold capacitor when switching to the HOLD mode. The charge should ideally be 0, but due to stray capacitive coupling and other effects, is typically 0.1pC in the EL4093. This charge changes the hold capacitor voltage by ∆V = ∆Q/CHOLD, and this ∆V is multiplied by the output stage transconductance (gm2) to produce a change in S/H output current. This last quantity is listed as the spec ∆ISTEP, and is calculated using the following: ∆Q ∆I SEP = -------------------- × gm2 C HOLD For CHOLD = 2.2nF and gm2 = 1/(500Ω), ∆ISTEP has a typical value of 100nA. This change in S/H output current flows through RF, shifting the CFA output voltage. However, as we shall soon see, this shift is negligible. Assuming RF = 750Ω, ∆ISTEP is impressed across RF to give (750Ω)(100nA) = 0.08mV of change at the CFA output. Droop Rate When the S/H amplifier is in HOLD mode, there is a small current that leaks from the switch into the hold capacitor. This quantity is termed the droop current, and is typically 10nA in the EL4093. This droop current produces a ramp in 10 the hold capacitor voltage, which in turn produces a similar effect at the CFA output. The Droop Rate at the CFA output can be found using the equation below: I DROOP Droop = ---------------------- ( gm2 × R F ) C HOLD Assuming RF = 750Ω and CHOLD = 2.2nF, the drift in the CFA output due to droop current is about 7µV/µs. Recall that in NTSC applications, there is about 60µs between autozero periods. Thus there is 7µV/µs(60µs) = 0.4mV, or less than 0.1 IRE, of drift over each NTSC scan line. This drift is negligible in most applications. Choice of Hold Capacitor The EL4093 has been designed to work with a hold capacitor of 2.2nF. With this value of CHOLD, the droop rate and hold step are negligibly small for most applications. In addition, with the special boost circuits inside the S/H, fast acquisition is possible even using a hold capacitor of this size. Figure 5 shows the input and output of the DC-restored amplifier while the S/H is in sample mode. Applying a +1V step to the non-inverting input of the CFA, the output of the CFA jumps to +2V. The S/H, however, then tries to autozero the system by driving the CFA output back to the reference voltage. Since the input differential across the S/H is initially +2V, the boost circuits turn on and supply 8.5mA of charge current to the hold capacitor. The boost circuit remains on until the CFA output has come to within 50mV of the reference. Note that this event took only 320ns; settling to within 1% of the final value takes another 2µs. Thus for a 1V input step, acquisition takes only one to two NTSC scan lines. EL4093 A remedy for this situation is to attenuate the colorburst before applying it to the S/H input. Figure 6 below shows a 3.58MHz chroma trap which would notch out the colorburst while preserving the video DC level. FIGURE 5. AUTOZERO MECHANISM RESTORES AMPLIFIER OUTPUT TO GROUND AFTER +1V STEP AT INPUT A natural question arises as to whether there are other CHOLD values that can be used. In one direction, increasing CHOLD will further reduce the droop and hold step, but lengthen the acquisition time. Since the droop and hold step are already small to begin with, there is no apparent advantage to increasing CHOLD. In the other direction, decreasing CHOLD would increase the droop and hold step but shorten the acquisition time. There is, however, a caveat to reducing CHOLD: too small a CHOLD would cause the autozero loop to oscillate. The reason is that when the S/H boost circuit turns on, the input stage gm increases drastically and the circuit becomes nonlinear. A sufficiently large CHOLD must be used to suppress the nonlinearity and force the loop to settle. For example, it has been found that a CHOLD of 470pF results in 1VP-P oscillation around 10MHz at the CFA output. The minimum recommended value for CHOLD is 2.2nF. With this value the loop remains stable over the entire operating temperature range (-40°C to +85°C). The greatest instability occurs at low temperatures, where we observe from the performance curves that the S/H gm’s, and hence the GBWP, are at their maximum. If the operating range is restricted to room temperature or above, then 1.5nF is sufficient to keep the loop stable. At this value of CHOLD the acquisition time reduces to about 1.5µs. Video Performance and Application Although the EL4093 is intended for high speed video applications such as SVGA, it also offers excellent performance for NTSC, with 0.04% dG and 0.02° dP at 3.58MHz. Some application considerations, however, are required for handling NTSC signals. Referring back to Figure 2, recall that typically, the autozero interval lies in the back porch portion of video containing the colorburst pulse. When the S/H compares the video to the reference voltage during this period, the colorburst (40 IREP-P) triggers the S/H boost circuit and prevents the autozero loop from settling. 11 FIGURE 6. COLORBURST TRAP FOR NTSC APPLICATIONS One may be tempted to use a RC lowpass filter to suppress the colorburst, as shown in Figure 7 below. This technique, however, poses several problems. First, to obtain enough attenuation, we need to set the pole frequency 10 to 20 times lower than 3.58MHz. This pole, being close to the auto zero loop pole, would destabilize the system and cause the loop to oscillate. FIGURE 7. CAUTION: LOWPASS FILTER DOES NOT WORK IN NTSC APPLICATIONS Although we can cancel this pole by introducing a zero, the RC network introduces a time delay between the CFA output and the S/H input. This has undesirable effects in some NTSC applications, as Figure 8 illustrates. There is only 0.6µs from the rising edge of sync to the colorburst. If we are autozeroing over the back porch, the autozero period would begin somewhere in this 0.6µs interval. Since the edge of sync is now delayed by the RC network, autozero begins before the video back porch reaches its final value. Consequently, the autozero loop performs a correction on every line and never settles. EL4093 FIGURE 8. LOWPASS FILTER DELAYS INPUT TO SAMPLE AND HOLD If the video does not contain any AC components during the autozero level (e.g. RGB video), then the above networks are not needed and the CFA output can be connected directly to the S/H input. where: VS = Supply Voltage ISMAX = Maximum Supply Current of Amplifier Power Dissipation VOUTMAX = Maximum Output Voltage of Application The EL4093 current feedback amplifier has an absolute maximum of ±30mA output current drive. This is slightly more than the current required to drive ±2V into 75Ω. To see how much the junction temperature is raised in this worst case, we refer to the equations below: RL = Load Resistance TJMAX = TMAX + (θJA • PDMAX) where: TMAX = Maximum Ambient Temperature θJA = Thermal Resistance of the Package PDMAX = Maximum Power Dissipation of the CFA and S/H amplifier in the Package For the EL4093, the maximum supply current is 11.5mA on VS = ±5V. Assume that in the worst case, the CFA output swings ±2V into 75Ω. Since the S/H has a current output, we assume that it is at maximum current swing (±5.5mA) but at a mid-rail output voltage (0V). With the above assumptions, PDMAX for the EL4093 is 223mW, and using the thermal resistance of a narrow SO package (120°C/W), this yields a temperature increase of 27°C. Since the maximum ambient temperature is 85°C, the resulting junction temperature of 112°C is still below the maximum. Please note that this in addition to metal migration problems. PDMAX for either the CFA or the S/H amplifier can be calculated as follows: PDMAX = (2•VS•ISMAX) + (VS - VOUTMAX) • (VOUTMAX/RL) All Intersil U.S. products are manufactured, assembled and tested utilizing ISO9000 quality systems. Intersil Corporation’s quality certifications can be viewed at www.intersil.com/design/quality Intersil products are sold by description only. Intersil Corporation reserves the right to make changes in circuit design, software and/or specifications at any time without notice. Accordingly, the reader is cautioned to verify that data sheets are current before placing orders. Information furnished by Intersil is believed to be accurate and reliable. However, no responsibility is assumed by Intersil or its subsidiaries for its use; nor for any infringements of patents or other rights of third parties which may result from its use. No license is granted by implication or otherwise under any patent or patent rights of Intersil or its subsidiaries. For information regarding Intersil Corporation and its products, see www.intersil.com 12