MF6 6th Order Switched Capacitor Butterworth Lowpass Filter General Description The MF6 is a versatile easy to use, precision 6th order Butterworth lowpass active filter. Switched capacitor techniques eliminate external component requirements and allow a clock tunable cutoff frequency. The ratio of the clock frequency to the lowpass cutoff frequency is internally set to 50 to 1 (MF6-50) or 100 to 1 (MF6-100). A Schmitt trigger clock input stage allows two clocking options, either self-clocking (via an external resistor and capacitor) for stand-alone applications, or an external TTL or CMOS logic compatible clock can be used for tighter cutoff frequency control. The maximally flat passband frequency response together with a DC gain of 1 V/V allows cascading MF6 sections for higher order filtering. In addition to the filter, two independent CMOS op amps are included on the die and are useful for any general signal conditioning applications. Features n n n n n n No external components Cutoff frequency accuracy of 0.3% typical Cutoff frequency range of 0.1 Hz to 20 kHz Two uncommitted op amps available 5V to 14V total supply voltage Cutoff frequency set by external or internal clock Block and Connection Diagrams All Packages DS005065-2 Top View Order Number MF6CWM-50 or MF6CWM-100 See NS Package Number M14B DS005065-1 TRI-STATE (R) is a registered trademark of National Semiconductor Corporation. (c) 1999 National Semiconductor Corporation DS005065 www.national.com MF6 6th Order Switched Capacitor Butterworth Lowpass Filter June 1999 Absolute Maximum Ratings (Note 11) Soldering Information Vapor Phase (60 sec.) 215C Infrared (15 sec.) 220C See AN-450 "Surface Mounting Methods and Their Effect on Product Reliability" (Appendix D) for other methods of soldering surface mount devices. If Military/Aerospace specified devices are required, please contact the National Semiconductor Sales Office/ Distributors for availability and specifications. Supply Voltage Voltage at Any Pin Input Current at Any Pin (Note 13) Package Input Current (Note 13) Power Dissipation (Note 14) Storage Temperature ESD Susceptibility (Note 12) 14V V- - 0.2V, V+ + 0.2V 5 mA 20 mA 500 mW -65C to +150C 800V Operating Ratings (Note 11) TMIN TA TMAX 0C TA +70C 5V to 14V Temperature Range MF6CWM-50, MF6CWM-100 Supply Voltage (VS = V+-V-) Filter Electrical Characteristics The following specifications apply for fCLK 250 kHz (Note 3) unless otherwise specified. Boldface limits apply for TMIN to TMAX; all other limits TA = TJ = 25C. Parameter Conditions Typical Tested Design (Note 8) Limit Limit (Note 9) (Note 10) Units V+ = +5V, V- = -5V fc, Cutoff MF6-50 Frequency Range MF6-100 (Note 1) Min 0.1 Max 20k Min 0.1 Max 10k fCLK =250 kHz Total Supply Current Maximum Clock 4.0 Filter Output Feedthrough 8.5 mA mV Op Amp 1 Out 25 (peak-to- Op Amp 2 Out 20 peak) 0.0 0.30 0.30 MF6-50 49.27 0.3% 49.27 1% 49.27 1% MF6-100 98.97 0.3% 98.97 1% 98.97 1% Rsource DC Gain 2 k Clock to Cutoff 6.0 30 Ho, fCLK/fc Hz dB Frequency Ratio DC Offset Voltage MF6-50 -200 MF6-100 -400 RL =10 k Minimum Output Voltage Swing Maximum Output mV +4.0 +3.5 +3.5 -4.1 -3.8 -3.5 V Source 50 Sink 1.5 mA MF6-50 83 dB MF6-100 81 Short Circuit Current (Note 6) Dynamic Range (Note 2) Additional MF6-50 fCLK =250 kHz Magnitude f=6000 Hz -9.47 -9.47 0.6 -9.47 0.75 Response Test f=4500 Hz -0.92 -0.92 0.6 -0.92 0.4 Points (Note 4) MF6-100 Attenuation Rate MF6-50 dB fCLK =250 kHz f=3000 Hz -9.48 -9.48 0.3 -9.48 0.75 f=2250 Hz -0.97 -0.97 0.3 -0.97 0.4 -36 -36 octave -36 -36 octave fCLK =250 kHz dB dB/ f1 =6000 Hz f2 =8000 Hz MF6-100 fCLK =250 kHz dB/ f1 =3000 Hz f2 =4000 Hz V+ = +2.5V, V- = -2.5V fc, Cutoff MF6-50 Frequency Range (Note 1) www.national.com MF6-100 Min 0.1 Max 10k Min 0.1 Max 5k 2 Hz Filter Electrical Characteristics (Continued) The following specifications apply for fCLK 250 kHz (Note 3) unless otherwise specified. Boldface limits apply for TMIN to TMAX; all other limits TA = TJ = 25C. Parameter Conditions Typical Tested Design (Note 8) Limit Limit (Note 9) (Note 10) 4.0 4.0 Units V+ = +2.5V, V- = -2.5V fCLK =250 kHz Total Supply Current Maximum Clock Feedthrough 2.5 mA Filter Output 20 mV Op Amp 1 Out 15 (peak-to- Op Amp 2 Out 10 Rsource2 k peak) 0.0 0.30 0.30 MF6-50 49.10 0.3% 49.10 2% 49.10 3% MF6-100 98.65 0.3% 98.65 2% 98.65 2.25% MF6-50 -200 MF6-100 -400 Ho, DC Gain dB fCLK/fc, Clock to Cutoff Frequency Ratio DC Offset Voltage RL =10 k Minimum Output Voltage Swing Maximum Output Source Short Circuit mV +1.5 +1.0 +1.0 -2.2 -1.7 -1.5 V 28 Sink 0.5 mA 77 dB Current (Note 6) Dynamic Range (Note 2) Additional MF6-50 fCLK =250 kHz Magnitude f=6000 Hz -9.54 -9.54 0.6 -9.54 0.75 Response Test f=4500 Hz -0.96 -0.96 0.3 -0.96 0.4 Points (Note 4) MF6-100 Attenuation MF6-50 fCLK =250 kHz f=3000 Hz -9.67 -9.67 0.6 -9.67 0.75 f=2250 Hz -1.01 -1.01 0.3 -1.01 0.4 -36 -36 octave -36 -36 octave fCLK =250 kHz dB dB/ f1 =6000 Hz Rate dB f2 =8000 Hz MF6-100 fCLK =250 kHz dB/ f1 =3000 Hz f2 =4000 Hz Op Amp Electrical Characteristics Boldface limits apply for TMIN to TMAX; all other limits TA = TJ = 25C. Parameter Conditions Typical Tested Design (Note 8) Limit Limit (Note 9) (Note 10) 20 20 Units V+ = +5V, V- = -5V 8.0 Input Offset Voltage Input Bias Current 10 VCM1 = 1.8V, CMRR (Op Amp #2 Only) mV pA 60 55 dB +4.0 +3.8 +3.6 -4.5 -4.0 -4.0 VCM2 = -2.2V RL = 10 k Output Voltage Swing Maximum Output Short Source 54 65 80 Circuit Current (Note 6) Sink 2.0 4.0 6.0 Slew Rate 7.0 V mA V/s DC Open Loop Gain 72 dB Gain Bandwidth Product 1.2 MHz V+ = +2.5V, V- = -2.5V 8.0 Input Offset Voltage 3 20 20 mV www.national.com Op Amp Electrical Characteristics (Continued) Boldface limits apply for TMIN to TMAX; all other limits TA = TJ = 25C. Parameter Conditions Typical Tested Design (Note 8) Limit Limit (Note 9) (Note 10) Units V+ = +2.5V, V- = -2.5V Input Bias Current 10 VCM1 = +0.5V, CMRR (Op-Amp #2 Only) pA 60 55 dB +1.5 +1.3 +1.1 -2.2 -1.7 -1.7 VCM2 = -0.9V RL = 10 k Output Voltage Swing Maximum Output Short Source 24 Circuit Current (Note 6) Sink 1.0 V mA Slew Rate 6.0 DC Open Loop Gain 67 V/s dB Gain Bandwidth Product 1.2 MHz Logic Input-Output Electrical Characteristics (Note 5) The following specifications apply for V- = 0V unless otherwise specified. Boldface limits apply for TMIN to TMAX; all other limits TA = TJ = 25C. Parameter Conditions Typical Tested Design (Note 8) Limit Limit Units (Note 9) (Note 10) 0.8 0.8 V 2.0 2.0 V 2.0 2.0 A V TTL CLOCK INPUT, CLK R PIN (Note 7) Maximum VIL, Logical "0" Input Voltage Minimum VIH, Logical "1" Input Voltage Maximum Leakage Current L Sh Pin at at CLK R Pin Mid- Supply SCHMITT TRIGGER VT+, Positive Going Min Threshold Voltage Max Min V+ = 10V 7.0 V+ = 5V 3.5 Max VT-, Negative Going Min Threshold Voltage Max Min V+ = 10V 3.0 V+ = 5V 1.5 Max Hysteresis (VT+ - VT-) Min V+ = 10V 4.0 Max Min V+ = 5V 2.0 Max Minimum Logical "1" Output Voltage (Pin 11) Maximum Logical "0" Output Voltage (Pin 11) Io = -10A Io = 10A 6.1 6.1 8.9 8.9 3.1 3.1 4.4 4.4 1.3 1.3 3.8 3.8 0.6 0.6 1.9 1.9 2.3 2.3 7.6 7.6 1.2 1.2 3.8 3.8 V+ = 10V 9.0 9.0 V+ = 5V 4.5 4.5 V+ = 10V 1.0 1.0 V+ = 5V 0.5 0.5 Minimum Output Source CLK R Tied V+ = 10V 6.0 3.0 3.0 Current (Pin 11) to Ground V+ = 5V 1.5 0.75 0.75 Maximum Output Sink CLK R Tied V+ = 10V 5.0 2.5 2.5 Current (Pin 11) to V+ V+ = 5V 1.3 0.65 0.65 V V V V V V V mA mA Note 1: The cutoff frequency of the filter is defined as the frequency where the magnitude response is 3.01 dB less than the DC gain of the filter. Note 2: For 5V supplies the dynamic range is referenced to 2.82 Vrms (4V peak) where the wideband noise over a 20 kHz bandwidth is typically 200 Vrms for the MF6-50 and 250 Vrms for the MF6-100. For 2.5V supplies the dynamic range is referenced to 1.06 Vrms (1.5V peak) where the wideband noise over a 20 kHz bandwidth is typically 140 Vrms for both the MF6-50 and the MF6-100. Note 3: The specifications for the MF6 have been given for a clock frequency (fCLK) of 250 kHz and less. Above this clock frequency the cutoff frequency begins to deviate from the specified error band of 1.0% but the filter still maintains its magnitude characteristics. See Application Hints, Section 1.5. www.national.com 4 Logic Input-Output Electrical Characteristics (Continued) Note 4: Besides checking the cutoff frequency (fc) and the stopband attenuation at 2 fc, two additional frequencies are used to check the magnitude response of the filter. The magnitudes are referenced to a DC gain of 0.0 dB. Note 5: For simplicity all the logic levels have been referenced to V- = 0V and will scale accordingly for 5V and 2.5V supplies (except for the TTL input logic levels). Note 6: The short circuit source current is measured by forcing the output that is being tested to its maximum positive voltage swing and then shorting that output to the negative supply. The short circuit sink current is measured by forcing the output that is being tested to its maximum negative voltage swing and then shorting that output to the positive supply. These are the worst-case conditions. Note 7: The MF6 is operating with symmetrical split supplies and L.Sh is tied to ground. Note 8: Typicals are at 25C and represent most likely parametric norm. Note 9: Tested limits are guaranteed to National's AOQL (Average Outgoing Quality Level). Note 10: Design limits are guaranteed, but not 100% tested. These limits are not used to calculate outgoing quality levels. Note 11: Absolute Maximum Ratings indicate limits beyond which damage to the device may occur. DC and AC electrical specifications do not apply when operating the device beyond its specified conditions. Note 12: Human body model, 100 pF discharged through a 1.5k resistor. Note 13: When the input voltage (VIN) at any pin exceeds the power supply rails (VIN < V- or VIN > V+) the absolute value of current at that pin should be limited to 5 mA or less. The 20 mA package input current limits the number of pins that can exceed the power supply boundaries with a 5 mA current limit to four. Note 14: The maximum power dissipation must be derated at elevated temperatures and is dictated by TJMAX, JA, and the ambient temperature, TA. The maximum allowable power dissipation at any temperature is PD = (TJMAX - TA)/JA or the number given in the Absolute Maximum Ratings, whichever is lower. For this device, TJMAX = 125C, and the typical junction-to-ambient thermal resistance is 78C/W. For the MF6CJ this number decreases to 62C/W. For MF6CWM, JA = 78C/W. Typical Performance Characteristics Schmitt Trigger Threshold Voltage vs Power Supply Voltage Crosstalk from Filter to Op-Amps (MF6-100) DS005065-41 DS005065-40 Crosstalk from Filter to Op-Amps (MF6-50) Crosstalk from Either Op-Amp to Filter Output (MF6-50) Crosstalk from Either Op-Amp to Filter Output (MF6-100) DS005065-43 DS005065-42 Equivalent Input Noise Voltage of Op-Amps DS005065-44 DS005065-45 5 www.national.com Typical Performance Characteristics Positive Voltage Swing vs Power Supply Voltage (Op Amp Output) (Continued) Positive Voltage Swing vs Power Supply Voltage (Filter Output) Positive Voltage Swing vs Temperature (Filter and Op Amp Outputs) DS005065-46 Negative Voltage Swing vs Power Supply Voltage (Filter and Op Amp Outputs) DS005065-47 Negative Voltage Swing vs Temperature (Filter and Op Amp Outputs) DS005065-48 Power Supply Current vs Clock Frequency DS005065-51 DS005065-49 Power Supply Current vs Temperature DS005065-50 Power Supply Current vs Power Supply Voltage DS005065-53 DS005065-52 www.national.com fCLK/fc Deviation vs Clock Frequency 6 DS005065-54 Typical Performance Characteristics fCLK/fc Deviation vs Temperature (Continued) fCLK/fc Deviation vs Power Supply Voltage DS005065-55 fCLK/fc Deviation vs Temperature fCLK/fc Deviation vs Clock Frequency DS005065-56 fCLK/fc Deviation vs Power Supply Voltage DS005065-58 DC Gain Deviation vs Power Supply Voltage DS005065-57 DC Gain Deviation vs Temperature DS005065-59 DC Gain Deviation vs Clock Frequency DS005065-61 DC Gain Deviation vs Temperature DS005065-62 7 DS005065-60 DS005065-63 www.national.com Typical Performance Characteristics DC Gain Deviation vs Power Supply Voltage (Continued) DC Gain Deviation vs Clock Frequency DS005065-64 DS005065-65 Crosstalk Test Circuits From Filter to Op Amps DS005065-10 From Either Op Amp to Filter Output DS005065-11 Pin Descriptions (Pin Numbers) Pin Description FILTER OUT (3) The output of the lowpass filter. It will typically sink 0.9 mA and source 3 mA and swing to within 1V of each supply rail. FILTER IN (8) The input to the lowpass filter. To minimize gain errors the source impedance that drives this input should be less than 2k (see section 1.4). For single supply operation the input signal must be biased to mid-supply or AC coupled. www.national.com 8 Pin Description VOSADJ (7) This pin is used to adjust the DC offset of the filter output; if not used it must be tied to the AGND potential. (See section 1.3) Pin Descriptions (Pin Numbers) (Continued) Pin Description AGND (5) The analog ground pin. This pin sets the DC bias level for the filter section and the non-inverting input of Op-Amp #1 and must be tied to the system ground for split supply operation or to mid-supply for single supply operation (see section 1.2). When tied to mid-supply this pin should be well bypassed. VO1 (4), INV1 (13) VO1 is the output and INV1 is the inverting input of Op-Amp #1. The non-inverting input of this Op-Amp is internally connected to the AGND pin. VO2 (2), INV2 (14), NINV2 (1) VO2 is the output, INV2 is the inverting input, and NINV2 is the non-inverting input of Op-Amp #2. V+(6), V-(10) The positive and negative supply pins. The total power supply range is 5V to 14V. Decoupling these pins with 0.1 F capacitors is highly recommended. CLK IN (9) A CMOS Schmitt-trigger input to be used with an external CMOS logic level clock. Also used for self-clocking Schmitt-trigger oscillator (see section 1.1). CLK R (11) A TTL logic level clock input when in split supply operation ( 2.5V to 7V) and L. Sh tied to system ground. This pin becomes a low impedance output when L. Sh is tied to V-. Also used in conjunction with the CLK IN pin for a self clocking Schmitt-trigger oscillator (see section 1.1). L. Sh (12) Level shift pin, selects the logic threshold levels for the desired clock. When tied to V- it enables an internal tri-state (R) buffer stage between the Schmitt trigger and the internal clock level shift stage thus enabling the CLK IN Schmitt-trigger input and making the CLK R pin a low impedance output. When the voltage level at this input exceeds [25%(V+ - V-) + V-] the internal tri-state buffer is disabled allowing the CLK R pin to become the clock input for the internal clock level shift stage. The CLK R threshold level is now 2V above the voltage applied to the L. Sh pin. Driving the CLK R pin with TTL logic levels can be accomplished through the use of split supplies and by tying the L. Sh pin to system ground. 9 www.national.com 1.0 MF6 Application Hints back capacitors in the integrators. The higher the clock to cutoff frequency ratio (or the sampling rate) the closer this approximation is to the theoretical Butterworth response. The MF6 is available in fCLK/fc ratios of 50:1 (MF6-50) or 100:1 (MF6-100). The MF6 is comprised of a non-inverting unity gain lowpass sixth order Butterworth switched capacitor filter section and two undedicated CMOS Op-Amps. The switched capacitor topology makes the cutoff frequency (where the gain drops 3.01 dB below the DC gain) a direct ratio (100:1 or 50:1) of the clock frequency supplied to the lowpass filter. Internal integrator time constants set the filter's cutoff frequency. The resistive element of these integrators is actually a capacitor which is "switched" at the clock frequency (for a detailed discussion see Input Impedance Section). Varying the clock frequency changes the value of this resistive element and thus the time constant of the integrators. The clock to cutoff frequency ratio (fCLK/fc) is set by the ratio of the input and feed- 1.1 CLOCK INPUTS The MF6 has a Schmitt-trigger inverting buffer which can be used to construct a simple R/C oscillator. The oscillator's frequency is dependent on the buffer's threshold levels as well as on the resistor/capacitor tolerance (see Figure 1). DS005065-12 FIGURE 1. Schmitt Trigger R/C Oscillator www.national.com 10 1.0 MF6 Application Hints (Continued) DS005065-4 FIGURE 3. Dual Supply Operation MF6 Driven with TTL Logic Level Clock DS005065-3 (VIH FIGURE 2. Dual Supply Operation MF6 Driven with CMOS Logic Level Clock 0.8 VCC and VIL 0.2 VCC where VCC = V+ - V-) 11 www.national.com 1.0 MF6 Application Hints (Continued) DS005065-14 a) Resistor Biasing of AGND DS005065-15 b) Using Op-Amp 2 to Buffer AGND FIGURE 4. Single Supply Operation www.national.com 12 1.0 MF6 Application Hints (Continued) DS005065-16 DS005065-17 FIGURE 5. VOS Adjust Schemes 1.4 INPUT IMPEDANCE Schmitt-trigger threshold voltage levels can change significantly causing the R/C oscillator's frequency to vary greatly from part to part. Where accuracy in fc is required an external clock can be used to drive the CLK R input of the MF6. This input is TTL logic level compatible and also presents a very light load to the external clock source ( z2 A) with split supplies and L. Sh tied to system ground. The logic level is programmed by the voltage applied to level shift (L. Sh) pin (See the Pin description for L. Sh pin). DS005065-18 a) Equivalent Circuit for MF6 Filter Input 1.2 POWER SUPPLY BIASING The MF6 can be biased from a single supply or dual split supplies. The split supply mode shown in Figure 2 and Figure 3 is the most flexible and easiest to implement. As discussed earlier split supplies, 5V to 7V, will enable the use of TTL or CMOS clock logic levels. Figure 4 shows two schemes for single supply biasing. In this mode only CMOS clock logic levels can be used. 1.3 OFFSET ADJUST The VosADJ pin is used in adjusting the output offset level of the filter section. If this pin is not used it must be tied to the analog ground (AGND) level, either mid-supply for single ended supply operation or ground for split supply operation. This pin sets the zero reference for the output of the filter. The implementation of this pin can be seen in Figure 5. In Figure 5a, DC offset is adjusted using a potentiometer; in Figure 5b, the Op-Amp integrator circuit keeps the average DC output level at AGND. The circuit in Figure 5b is therefore appropriate only for AC-coupled signals and signals biased at AGND. DS005065-19 b) Actual Circuit for MF6 Filter Input FIGURE 6. MF6 Filter Input The MF6 lowpass filter input (FILTER IN pin) is not a high impedance buffer input. This input is a switched capacitor resistor equivalent, and its effective impedance is inversely proportional to the clock frequency. The equivalent circuit of the input to the filter can be seen in Figure 6. The input capacitor charges to the input voltage (Vin) during one half of the clock period, during the second half the charge is transferred to the feedback capacitor. The total transfer of charge in one clock cycle is therefore Q = CinVin, and since current is defined as the flow of charge per unit time the average input current becomes Iin = Q/T 13 www.national.com 1.0 MF6 Application Hints (Continued) (where T equals one clock period) or The propagation delay in the logic and the settling time required to acquire a new voltage level on the capacitors increases as the MF6 power supply voltage decreases. This causes a shift in the fCLK/fc ratio which will become noticeable when the clock frequency exceeds 250 kHz. The amplitude characteristic will stay within tolerance until fCLK exceeds 500 kHz and will peak at about 0.5 dB at the corner frequency with a 1 MHz clock. The response of the MF6 is still a reasonable approximation of the ideal Butterworth lowpass characteristic as can be seen in Figures 7, 8, 9, 10. The equivalent input resistor (Rin) then can be defined as The input capacitor is 2 pF for the MF6-50 and 1 pF for the MF6-100, so for the MF6-100 and for the MF6-50. As shown in the above equations for a given cutoff frequency (fc) the input impedance remains the same for the MF6-50 and the MF6-100. The higher the clock to center frequency ratio, the greater equivalent input resistance for a given clock frequency. As the cutoff frequency increases the equivalent input impedance decreases. This input resistance will form a voltage divider with the source impedance (Rsource). Since Rin is inversely proportional to the cutoff frequency, operation at higher cutoff frequencies will be more likely to load the input signal which would appear as an overall decrease in gain to the output of the filter. Since the filter's ideal gain is unity its overall gain is given by: DS005065-20 FIGURE 7. MF6-100 5V Supplies Amplitude Response If the MF6-50 or the MF6-100 were set up for a cutoff frequency of 10 kHz the input impedance would be: DS005065-21 FIGURE 8. MF6-50 5V Supplies Amplitude Response In this example with a source impedance of 10k the overall gain, if the MF6 had an ideal gain of 1 or 0 dB, would be: Since the maximum overall gain error for the MF6 is 0.3 dB with a Rs 2 k the actual gain error for this case would be +0.21 dB to -0.39 dB. 1.5 CUTOFF FREQUENCY RANGE The filter's cutoff frequency (fc) has a lower limit caused by leakage currents through the internal switches discharging the stored charge on the capacitors. At lower clock frequencies these leakage currents can cause millivolts of error, for example: www.national.com DS005065-22 FIGURE 9. MF6-100 2.5V Supplies Amplitude Response 14 1.0 MF6 Application Hints (Continued) DS005065-23 FIGURE 10. MF6-50 2.5V Supplies Amplitude Response DS005065-24 FIGURE 11. Design Example Magnitude Response Specification Where the Response of the Filter Design Must Fall Within the Shaded Area of the Specification 2.0 Designing with the MF6 Given any lowpass filter specification two equations will come in handy in trying to determine whether the MF6 will do the job. The first equation determines the order of the lowpass filter required: Since the MF6's cutoff frequency fc, which corresponds to a gain attenuation of -3.01 dB, was not specified in this example it needs to be calculated. Solving equation 2 where f = fc as follows: where n is the order of the filter, Amin is the minimum stopband attenuation (in dB) desired at frequency fs, and Amax is the passband ripple or attenuation (in dB) at frequency fb. If the result of this equation is greater than 6, then more than a single MF6 is required. The attenuation at any frequency can be found by the following equation: Attn(f) = 10 log [1 + (100.1A max-1) (f/fb)2n] dB (2) where n = 6 (the order of the filter). To implement this example for the MF6-50 the clock frequency will have to be set to fCLK = 50(1.116 kHz) = 55.8 kHz or for the MF6-100 fCLK = 100(1.116 kHz) = 111.6 kHz. 2.2 CASCADING MF6s In the case where a steeper stopband attenuation rate is required two MF6's can be cascaded (Figure 12) yielding a 12th order slope of 72 dB per octave. Because the MF6 is a Butterworth filter and therefore has no ripple in its passband, when MF6s are cascaded the resulting filter also has no ripple in its passband. Likewise the DC and passband gains will remain at 1V/V. The resulting response is shown in Figures 13, 14. In determining whether the cascaded MF6s will yield a filter that will meet a particular amplitude response specification, as above, equations 3 and 4 can be used, shown below. 2.1 A LOWPASS DESIGN EXAMPLE Suppose the amplitude response specification in Figure 11 is given. Can the MF6 be used? The order of the Butterworth approximation will have to be determined using eq. 1: Amin = 30 dB, Amax = 1.0 dB, fs = 2 kHz, and fb = 1 kHz Since n can only take on integer values, n = 6. Therefore the MF6 can be used. In general, if n is 6 or less a single MF6 stage can be utilized. Likewise, the attenuation at fs can be found using equation 2 with the above values and n = 6 giving: Atten (2 kHz) = 10 log [ 1 + (100.1 - 1) (2 kHz/1 kHz)12] = 30.26 dB where n = 6 (the order of each filter). Equation 3 will determine whether the order of the filter is adequate (n 6) while equation 4 can determine if the required stopband attenuation is met and what actual cutoff frequency (fc) is required to obtain the particular frequency response desired. The design procedure would be identical to the one shown in section 2.1. This result also meets the design specification given in Figure 11 again verifying that a single MF6 section will be adequate. 15 www.national.com 2.0 Designing with the MF6 Since R1 does not equal R2 there will be a gain inequality above and below the notch frequency. At frequencies below the notch frequency (f << fn), the signal through the filter has a gain of one and is non-inverting. Summing this with the input signal through the Op-Amp yields an overall gain of two or +6 dB. For f >> fn, the signal at the output of the filter is greatly attenuated thus only the input signal will appear at the output of the Op-Amp. With R3 = R1 = 1.014 R2 the overall gain is 0.986 or -0.12 dB at frequencies above the notch. (Continued) 2.3 IMPLEMENTING A "NOTCH" FILTER WITH THE MF6 A "notch" filter with 60 dB of attenuation can be obtained by using one of the Op-Amps, available in the MF6, and three external resistors. The circuit and amplitude response are shown in Figures 15, 16. The frequency where the "notch" will occur is equal to the frequency at which the output signal of the MF6 will have the same magnitude but be 180 degrees out of phase with its input signal. For a sixth order Butterworth filter 180 phase shift occurs where f = fn = 0.742 fc. The attenuation at this frequency is 0.12 dB which must be compensated for by making R1 = 1.014 x R2. DS005065-25 FIGURE 12. Cascading Two MF6s DS005065-27 FIGURE 14. Phase Response of Two Cascaded MF6-50s DS005065-26 FIGURE 13. One MF6-50 vs. Two MF6-50s Cascaded www.national.com 16 2.0 Designing with the MF6 (Continued) DS005065-28 FIGURE 15. "Notch" Filter DS005065-29 FIGURE 16. MF6-50 "Notch" Filter Amplitude Response The transient response of the MF6 seen in Figure 18 is also dependent on the fc and thus the fCLK applied to the filter. The MF6 responds as a classical sixth order Butterworth lowpass filter. 2.4 CHANGING CLOCK FREQUENCY INSTANTANEOUSLY The MF6 will respond favorably to a sudden change in clock frequency. Distortion in the output signal occurs at the transition of the clock frequency and lasts approximately three cutoff frequency (fc) cycles. As shown in Figure 17, if the control signal is low the MF6-50 has a 100 kHz clock making fc = 2 kHz; when this signal goes high the clock frequency changes to 50 kHz yielding 1 kHz fc. 17 www.national.com 2.0 Designing with the MF6 passband of the filter and of large enough amplitude it can cause problems. Therefore if frequency components in the input signal exceed fCLK/2 they must be attenuated before being applied to the MF6 input. The necessary amount of attenuation will vary depending on system requirements. In critical applications the signal components above fCLK/2 will have to be attenuated at least to the filter's residual noise level. An example circuit is shown in Figure 20 using one of the uncommitted Op-Amps available in the MF6. (Continued) DS005065-30 fIN = 1.5 kHz (scope time base = 2 ms/div) FIGURE 17. MF6-50 Abrupt Clock Frequency Change 2.5 ALIASING CONSIDERATIONS Aliasing effects have to be taken into consideration when input signal frequencies exceed half the sampling rate. For the MF6 this equals half the clock frequency (fCLK). When the input signal contains a component at a frequency higher than half the clock frequency, as in Figure 19a, that component will be "reflected" about fCLK/2 into the frequency range below fCLK/2 as in Figure 19b. If this component is within the DS005065-31 FIGURE 18. MF6-50 Step Input Response, Vertical = 2V/div., Horizontal = 1 ms/div., fCLK = 100 kHz DS005065-37 (a) Input Signal Spectrum DS005065-38 (b) Output Signal Spectrum. Note that the input signal at fs/2 + f causes an output signal to appear at fs/2 - f. FIGURE 19. The phenomenon of aliasing in sampled-data systems. An input signal whose frequecy is greater than one-half the sampling frequency will cause an output to appear at a frequency lower than one-half the sampling frequency. In the MF6, fsfCLK. www.national.com 18 2.0 Designing with the MF6 (Continued) DS005065-34 Note: The parallel combination of R4 (if used), R1 and R2 should be 10 k in order not to load Op-Amp #2. FIGURE 20. Second Order Butterworth Anti-Aliasing Filter Using Uncommitted Op-Amp #2 19 www.national.com MF6 6th Order Switched Capacitor Butterworth Lowpass Filter Physical Dimensions inches (millimeters) unless otherwise noted Small Outline Wide Body (M) Order Number MF6CWM-50 or MF6CWM-100 NS Package Number M14B LIFE SUPPORT POLICY NATIONAL'S PRODUCTS ARE NOT AUTHORIZED FOR USE AS CRITICAL COMPONENTS IN LIFE SUPPORT DEVICES OR SYSTEMS WITHOUT THE EXPRESS WRITTEN APPROVAL OF THE PRESIDENT AND GENERAL COUNSEL OF NATIONAL SEMICONDUCTOR CORPORATION. As used herein: 1. Life support devices or systems are devices or systems which, (a) are intended for surgical implant into the body, or (b) support or sustain life, and whose failure to perform when properly used in accordance with instructions for use provided in the labeling, can be reasonably expected to result in a significant injury to the user. National Semiconductor Corporation Americas Tel: 1-800-272-9959 Fax: 1-800-737-7018 Email: support@nsc.com www.national.com National Semiconductor Europe Fax: +49 (0) 1 80-530 85 86 Email: europe.support@nsc.com Deutsch Tel: +49 (0) 1 80-530 85 85 English Tel: +49 (0) 1 80-532 78 32 Francais Tel: +49 (0) 1 80-532 93 58 Italiano Tel: +49 (0) 1 80-534 16 80 2. A critical component is any component of a life support device or system whose failure to perform can be reasonably expected to cause the failure of the life support device or system, or to affect its safety or effectiveness. National Semiconductor Asia Pacific Customer Response Group Tel: 65-2544466 Fax: 65-2504466 Email: sea.support@nsc.com National Semiconductor Japan Ltd. Tel: 81-3-5639-7560 Fax: 81-3-5639-7507 National does not assume any responsibility for use of any circuitry described, no circuit patent licenses are implied and National reserves the right at any time without notice to change said circuitry and specifications.