AMMC-6650
DC–40 GHz Variable Attenuator
Data Sheet
Description
The AMMC-6650 is a voltage controlled variable attenua-
tor designed to operate from DC-40 GHz. It is fabricated
using Avago Technologies enhancement mode pHEMT
MMIC process with backside ground vias, and gate lengths
of approximately 0.25um. The distributed topology of the
AMMC-6650 facilitates broadband operation by absorbing
parasitic e ects of its series and shunt FETs. An on-chip DC
reference circuit may be used to maintain optimum VSWR
for any attenuation setting or to provide more linear
attenuation versus voltage response.
Simpli ed Schematic
Features
• Wide Frequency Range DC-40 GHz
• Attenuation Range 20dB
• Single Positive Bias Supply
• Unconditionally Stable
Applications
• Microwave Radio Systems
• Satellite VSAT, DBS Up / Down Link
• LMDS & Pt – Pt mmW Long Haul
• Broadband Wireless Access (including 802.16 and
802.20 WiMax)
• WLL and MMDS loops
Attention: Observe precautions for
handling electrostatic sensitive devices.
ESD Machine Model = 80 V
ESD Human Body Model = 400 V
Refer to Avago Application Note A004R:
Electrostatic Discharge, Damage and Control.
Chip Size: 1530 μm x 660 µm (61.2 x 26.4 mils)
Chip Size Tolerance: ±10 µm (±0.4 mils)
Chip Thickness: 100 ± 10 µm (4 ±0.4 mils)
Pad Dimensions: 80 x 120 µm (3.2 x 4.8 mils)
DCin DCout V2
V1
RFoutRFin
variable
attenuator
DC
reference circuit
2
Table 1. Absolute Maximum Ratings
Symbol Parameters and Test Conditions Unit Minimum Maximum
V1Voltage to Control VSWR V 0 1.6
V2Voltage to Control Attenuation V 0 1.6
Pin RF Input Power dBm - 17
Tch Operating Channel Temperature °C - +150
Tstg Storage Temperature °C -65 +150
Tmax Maximum Assembly Temperature °C +300
for 60 seconds
Notes:
Operation in excess of any one of these conditions may result in permanent damage to this device.
The absolute maximum ratings for V1, V2 and Pin were determined at an ambient temperature of 25ºC unless noted otherwise.
Table 2. DC Speci cations
Symbol Parameters Test Conditions Unit Min Typical Max
Ic_V1_ref V1 Control Current (Min Attenuation) V1=1.5 V, V2=0 V mA - 1.93 2.0
Ic_V2_ref V2 Control Current (Min Attenuation) V1=1.5 V, V2=0 V uA - 0.8 2.5
Ic_V1_max V1 Control Current (Max Attenuation) V1=0V, V2=1.25 V uA - 1.1 2.5
Ic_V2_max V2 Control Current (Max Attenuation) V1=0 V, V2=1.25 V mA - 1.41 1.5
Notes:
Ambient temperature TA = 25°C
Table 3. RF Speci cations (TA = 25°C, Z0 = 50 Ω)
Symbol Parameters and Test Conditions Units Freq. [GHz] Minimum Typical Maximum
Minimum Attenuation
(Reference State)
|S21|
V1 = 1.5 V
V2 = 0.0 V
dB 2 1.1 2.0
20 1.7 2.5
33 2.6 4.0
40 3.1 5.0
Maximum Attenuation |S21|
V1 = 0.0 V
V2 = 1.25 V
dB 2 24.0 26.4
20 24.5 28.1
33 26.0 32.7
40 27.0 35.7
Return Loss (In/Out)
at Reference State
V1=1.5 V, V2=0.0 V dB <40 10
Return Loss (In/Out)
at Max. Attenuation
V1=0.0 V, V2=1.25 V dB <40 10
Notes:
Data obtained from on-wafer measurements
3
Typical Distribution Charts
Figure 3d. Min Attenuation @ 33GHz, Nominal=2.6, USL=4.0 Figure 4d. Min Attenuation @ 40GHz, Nominal=3.1, USL=5.0
Figure 5d. Max Attenuation @ 2GHz, LSL=24.0, Nominal=26.4 Figure 6d. Max Attenuation @ 20GHz, LSL=24.5, Nominal=24.5
Figure 1d. Min Attenuation @ 2GHz, Nominal=1.1, USL=2.0 Figure 2d. Min Attenuation @ 20GHz, Nominal=1.7, USL=2.5
Figure 7d. Max Attenuation @ 33GHz, LSL=26.0, Nominal=32.7 Figure 8d. Max Attenuation @ 40GHz, LSL=27.0, Nominal=35.7
Notes:
1. All data from on-wafer measurements
2. Distribution data based on 5000 part sample from two wafer lots tested during initial characterization. Future wafers may have nominal values
anywhere between upper and lower limit
4
0
10
20
30
40
50
0 1020304050
Frequency (GHz)
Frequency (GHz)
Frequency (GHz)
Attenuation (dB)
Min
Min+2dB
Min+4dB
Min+6dB
Min+8dB
Min+12dB
Min+16dB
Min+20dB
Max -40
-30
-20
-10
0
0 102030 4050
Input Return Loss (dB)
Min
Max
-40
-30
-20
-10
0
0 10 20 30 40 50
Output Return Loss (dB)
0
5
10
15
20
25
30
0 5 10 15 20 25 30
Attenuation (dB)
Attenuation (dB)
IIP3 (dBm)
0
5
10
15
20
25
0 5 10 15 20 25 30
Attenuation (dB)
IIP3 (dBm)
IIP3 (dBm)
0
5
10
15
20
25
0 5 10 15 20 25 30
Min
Max
Typical Performance (TA = 25°C, Zin = Zout = 50 Ω)
Figure 1. Attenuation vs Frequency Figure 2. Input Return Loss vs Frequency
Figure 3. Output Return Loss vs Frequency Figure 4. IIP3 vs Attenuation at 2 GHz (note 2)
Figure 5. IIP3 vs Attenuation at 12 GHz (note 2) Figure 6. IIP3 vs Attenuation at 22 GHz (note 2)
5
0
5
10
15
20
25
30-10 -5 0 5 10
Input Power (dBm) Input Power (dBm)
Attenuation (dB)
Input Power (dBm)
Attenuation (dB)
Attenuation (dB)
Input Power (dBm)
Attenuation (dB)
Min
Min+2dB
Min+4dB
Min+6dB
Min+8dB
Min+12dB
Min+16dB
Min+20dB
Max
0
5
10
15
20
25
30
35-10 -5 0 5 10
Min
Min+2dB
Min+4dB
Min+6dB
Min+8dB
Min+12dB
Min+16dB
Min+20dB
Max
0
5
10
15
20
25
30
35-10 -5 0 5 10
Min
Min+2dB
Min+4dB
Min+6dB
Min+8dB
Min+12dB
Min+16dB
Min+20dB
Max
0
5
10
15
20
25
30
35-10 -5 0 5 10
Min
Min+2dB
Min+4dB
Min+6dB
Min+8dB
Min+12dB
Min+16dB
Min+20dB
Max
0
2
4
6
8
10
0 1020304050
Frequency (GHz)
Attenuation (dB)
Attenuation (dB)
Frequency (GHz)
20
25
30
35
40 01020304050
-40C
25C
85C
-40C
25C
85C
Figure 7. Attenuation vs Input Power at 2 GHz Figure 8. Attenuation vs Input Power at 12 GHz
Figure 9. Attenuation vs Input Power at 22 GHz Figure 10. Attenuation vs Input Power at 32 GHz
Figure 11. Attenuation vs Frequency (Min Attenuation) Figure 12. Attenuation vs Frequency (Max Attenuation)
Notes:
1. All tests done on a AMMC-6650 mounted on a PCB equipped with RF connectors and an op-amp driver shown in Figure 14.
2. IIP3 measured with two input signals with frequency di erence of 10 MHz, each input signal at -10 dBm
3. All attenuation settings were done at 2GHz
6
AMMC-6650 Typical Scattering Parameters at Minimum Attenuation
(Measured on-wafer, Tc = 25°C, Zo = 50ohm, V1 = 1.5V, V2 = 0V)
Freq
GHz
S11 S21 S12 S22
dB Mag Phase dB Mag Phase dB Mag Phase dB Mag Phase
0.5 -25.969 0.050 -39.087 -1.135 0.878 -5.786 -1.147 0.876 -5.788 -26.143 0.049 -39.150
1.0 -26.108 0.050 -66.653 -1.103 0.881 -10.300 -1.176 0.873 -10.283 -25.498 0.053 -60.168
2.0 -22.757 0.073 -93.483 -1.132 0.878 -19.219 -1.230 0.868 -19.238 -22.662 0.074 -85.472
3.0 -20.131 0.099 -108.015 -1.242 0.867 -28.116 -1.341 0.857 -28.094 -20.202 0.098 -102.098
4.0 -18.373 0.121 -120.271 -1.313 0.860 -36.932 -1.428 0.848 -36.892 -18.533 0.118 -115.389
5.0 -16.948 0.142 -131.540 -1.359 0.855 -45.539 -1.470 0.844 -45.485 -17.310 0.136 -127.577
6.0 -16.016 0.158 -140.873 -1.420 0.849 -54.184 -1.544 0.837 -53.995 -16.346 0.152 -137.375
7.0 -15.340 0.171 -149.734 -1.464 0.845 -62.729 -1.595 0.832 -62.548 -15.740 0.163 -146.531
8.0 -14.813 0.182 -157.525 -1.522 0.839 -71.279 -1.653 0.827 -71.025 -15.254 0.173 -155.038
9.0 -14.689 0.184 -165.279 -1.551 0.837 -79.808 -1.702 0.822 -79.509 -15.026 0.177 -163.282
10.0 -14.466 0.189 -172.907 -1.616 0.830 -88.399 -1.746 0.818 -88.014 -14.943 0.179 -171.767
11.0 -14.737 0.183 -179.967 -1.643 0.828 -96.805 -1.788 0.814 -96.516 -15.006 0.178 179.931
12.0 -14.851 0.181 172.450 -1.678 0.824 -105.447 -1.819 0.811 -105.110 -15.360 0.171 172.348
13.0 -15.386 0.170 165.587 -1.699 0.822 -114.080 -1.838 0.809 -113.678 -15.831 0.162 164.283
14.0 -15.918 0.160 158.241 -1.733 0.819 -122.788 -1.850 0.808 -122.395 -16.415 0.151 156.596
15.0 -16.809 0.144 151.016 -1.753 0.817 -131.503 -1.870 0.806 -131.171 -17.355 0.136 148.580
16.0 -17.890 0.128 143.458 -1.781 0.815 -140.327 -1.889 0.805 -139.898 -18.621 0.117 141.079
17.0 -19.372 0.108 135.142 -1.801 0.813 -149.250 -1.896 0.804 -148.860 -20.175 0.098 134.038
18.0 -21.140 0.088 127.266 -1.806 0.812 -158.422 -1.917 0.802 -158.009 -22.192 0.078 125.365
19.0 -23.388 0.068 116.287 -1.836 0.810 -167.679 -1.934 0.800 -167.132 -24.928 0.057 118.442
20.0 -27.432 0.043 99.261 -1.867 0.807 -176.938 -1.961 0.798 -176.522 -28.730 0.037 107.600
21.0 -32.956 0.023 59.397 -1.920 0.802 173.528 -2.023 0.792 174.005 -35.239 0.017 81.055
22.0 -31.568 0.026 -19.082 -2.000 0.794 163.905 -2.096 0.786 164.364 -37.589 0.013 -23.760
23.0 -25.613 0.052 -53.962 -2.121 0.783 154.170 -2.209 0.775 154.655 -29.473 0.034 -61.826
24.0 -21.577 0.083 -71.482 -2.230 0.774 144.556 -2.348 0.763 145.178 -24.437 0.060 -73.246
25.0 -18.380 0.121 -85.747 -2.348 0.763 135.424 -2.448 0.754 136.043 -20.896 0.090 -85.749
26.0 -16.461 0.150 -100.632 -2.387 0.760 126.769 -2.498 0.750 127.328 -18.496 0.119 -98.530
27.0 -15.006 0.178 -114.327 -2.450 0.754 118.757 -2.553 0.745 119.173 -16.755 0.145 -110.996
28.0 -14.352 0.192 -127.726 -2.494 0.750 110.138 -2.597 0.742 110.632 -15.885 0.161 -123.634
29.0 -14.239 0.194 -137.892 -2.585 0.743 101.101 -2.638 0.738 101.581 -15.473 0.168 -134.550
30.0 -14.485 0.189 -147.053 -2.649 0.737 91.495 -2.681 0.734 91.910 -15.650 0.165 -144.865
31.0 -14.535 0.188 -155.460 -2.703 0.733 82.323 -2.726 0.731 82.845 -15.735 0.163 -151.713
32.0 -15.001 0.178 -164.229 -2.737 0.730 73.207 -2.773 0.727 73.808 -16.165 0.156 -160.229
33.0 -16.021 0.158 -171.696 -2.784 0.726 63.269 -2.832 0.722 63.902 -17.215 0.138 -168.812
34.0 -17.735 0.130 -177.349 -2.833 0.722 53.098 -2.873 0.718 53.676 -18.854 0.114 -174.738
35.0 -20.087 0.099 -178.011 -2.832 0.722 42.787 -2.922 0.714 43.305 -20.964 0.090 -173.111
36.0 -22.476 0.075 -166.294 -2.949 0.712 31.993 -2.958 0.711 32.694 -23.427 0.067 -162.318
37.0 -23.479 0.067 -143.625 -2.966 0.711 21.034 -3.013 0.707 21.810 -24.082 0.063 -137.349
38.0 -21.463 0.085 -122.928 -3.009 0.707 9.877 -3.088 0.701 10.556 -21.587 0.083 -117.676
39.0 -18.570 0.118 -116.928 -3.119 0.698 -1.475 -3.194 0.692 -0.735 -18.599 0.118 -113.548
40.0 -16.375 0.152 -118.715 -3.243 0.688 -13.001 -3.317 0.683 -12.195 -16.283 0.153 -115.929
41.0 -14.563 0.187 -123.769 -3.424 0.674 -24.858 -3.491 0.669 -24.054 -14.316 0.192 -121.984
42.0 -13.116 0.221 -130.906 -3.690 0.654 -36.766 -3.756 0.649 -36.082 -12.686 0.232 -130.382
43.0 -12.013 0.251 -139.354 -4.013 0.630 -48.669 -4.088 0.625 -48.156 -11.509 0.266 -139.456
44.0 -11.264 0.273 -148.870 -4.342 0.607 -60.453 -4.425 0.601 -59.744 -10.672 0.293 -149.803
45.0 -10.800 0.288 -158.844 -4.784 0.577 -72.706 -4.816 0.574 -71.850 -9.962 0.318 -160.722
46.0 -10.586 0.296 -169.457 -5.259 0.546 -84.156 -5.296 0.544 -83.532 -9.476 0.336 -171.880
47.0 -10.592 0.295 -179.737 -5.672 0.521 -95.670 -5.720 0.518 -95.070 -9.224 0.346 176.644
48.0 -10.779 0.289 170.426 -6.145 0.493 -107.654 -6.224 0.488 -107.004 -9.136 0.349 164.963
49.0 -11.283 0.273 162.115 -6.762 0.459 -119.116 -6.798 0.457 -118.459 -9.000 0.355 152.822
7
AMMC-6650 Typical Scattering Parameters at Maximum Attenuation
(Measured on-wafer, Tc = 25°C, Zo = 50ohm, V1 = 0V, V2 = 1.25V)
Freq
GHz
S11 S21 S12 S22
dB Mag Phase dB Mag Phase dB Mag Phase dB Mag Phase
0.5 -22.853 0.072 -12.256 -24.466 0.060 -3.791 -24.539 0.059 -3.663 -22.627 0.074 -12.664
1.0 -22.534 0.075 -21.574 -24.510 0.060 -6.688 -24.568 0.059 -6.693 -22.545 0.075 -21.431
2.0 -21.906 0.080 -38.524 -24.539 0.059 -12.555 -24.524 0.059 -12.714 -21.895 0.080 -38.026
3.0 -20.867 0.091 -53.432 -24.510 0.060 -18.549 -24.481 0.060 -18.529 -21.042 0.089 -53.608
4.0 -19.862 0.102 -66.954 -24.451 0.060 -24.608 -24.451 0.060 -24.729 -20.140 0.098 -66.814
5.0 -19.023 0.112 -78.640 -24.308 0.061 -30.768 -24.351 0.061 -30.758 -19.228 0.109 -78.878
6.0 -18.209 0.123 -89.031 -24.166 0.062 -37.148 -24.194 0.062 -37.219 -18.387 0.120 -89.054
7.0 -17.445 0.134 -98.870 -24.013 0.063 -44.026 -23.986 0.063 -43.803 -17.661 0.131 -98.201
8.0 -16.803 0.145 -107.012 -23.849 0.064 -51.085 -23.822 0.064 -50.879 -17.046 0.141 -106.952
9.0 -16.346 0.152 -114.618 -23.688 0.065 -58.509 -23.675 0.066 -58.452 -16.496 0.150 -114.984
10.0 -15.890 0.161 -121.500 -23.531 0.067 -66.183 -23.531 0.067 -66.106 -16.110 0.157 -122.545
11.0 -15.494 0.168 -127.928 -23.375 0.068 -74.133 -23.363 0.068 -74.120 -15.751 0.163 -129.664
12.0 -15.315 0.172 -134.558 -23.248 0.069 -82.421 -23.210 0.069 -82.298 -15.509 0.168 -136.450
13.0 -15.169 0.174 -140.662 -23.135 0.070 -91.000 -23.073 0.070 -90.910 -15.340 0.171 -142.742
14.0 -15.011 0.178 -146.591 -23.061 0.070 -100.015 -23.012 0.071 -99.961 -15.274 0.172 -149.243
15.0 -15.045 0.177 -151.665 -22.987 0.071 -109.481 -22.963 0.071 -109.601 -15.264 0.173 -154.910
16.0 -15.209 0.174 -156.830 -22.902 0.072 -119.366 -22.914 0.072 -119.416 -15.371 0.170 -160.368
17.0 -15.325 0.171 -162.011 -22.950 0.071 -130.119 -22.902 0.072 -130.079 -15.540 0.167 -165.316
18.0 -15.453 0.169 -166.083 -23.135 0.070 -141.644 -23.073 0.070 -141.663 -15.783 0.163 -169.305
19.0 -15.708 0.164 -170.398 -23.544 0.067 -153.417 -23.453 0.067 -153.207 -15.928 0.160 -172.863
20.0 -15.972 0.159 -174.733 -24.027 0.063 -164.758 -23.849 0.064 -164.846 -15.890 0.161 -176.478
21.0 -16.143 0.156 -178.859 -24.423 0.060 -174.164 -24.408 0.060 -174.163 -15.896 0.160 178.673
22.0 -16.472 0.150 176.398 -24.913 0.057 178.184 -24.852 0.057 178.157 -16.027 0.158 173.400
23.0 -16.973 0.142 170.743 -25.368 0.054 172.250 -25.272 0.055 172.286 -16.160 0.156 167.414
24.0 -17.530 0.133 166.527 -25.849 0.051 167.104 -25.730 0.052 166.986 -16.478 0.150 161.938
25.0 -18.041 0.125 164.230 -26.214 0.049 164.229 -26.249 0.049 163.911 -16.839 0.144 156.076
26.0 -18.651 0.117 159.564 -26.840 0.046 159.223 -27.111 0.044 159.007 -17.171 0.139 149.903
27.0 -19.356 0.108 154.880 -27.351 0.043 153.963 -27.556 0.042 153.449 -17.582 0.132 143.328
28.0 -19.676 0.104 149.943 -28.046 0.040 146.145 -27.766 0.041 145.705 -17.951 0.127 136.221
29.0 -20.537 0.094 144.531 -28.179 0.039 137.226 -28.134 0.039 137.096 -18.540 0.118 128.334
30.0 -21.650 0.083 135.232 -28.382 0.038 127.342 -28.382 0.038 126.543 -19.188 0.110 118.732
31.0 -22.793 0.073 126.384 -28.898 0.036 120.275 -28.826 0.036 120.255 -20.035 0.100 109.204
32.0 -24.013 0.063 118.078 -28.995 0.036 112.761 -28.947 0.036 112.587 -20.612 0.093 99.496
33.0 -25.482 0.053 103.602 -29.168 0.035 103.999 -29.143 0.035 103.186 -21.300 0.086 86.280
34.0 -26.916 0.045 81.551 -29.551 0.033 95.536 -29.499 0.034 95.099 -21.598 0.083 71.770
35.0 -27.639 0.042 58.000 -29.924 0.032 86.633 -29.924 0.032 86.302 -21.820 0.081 57.472
36.0 -27.171 0.044 33.749 -30.257 0.031 77.849 -30.286 0.031 76.950 -21.982 0.080 41.425
37.0 -25.866 0.051 12.352 -30.842 0.029 69.338 -30.663 0.029 70.386 -21.660 0.083 26.238
38.0 -23.917 0.064 -1.642 -31.150 0.028 61.281 -31.437 0.027 59.108 -20.983 0.089 12.205
39.0 -22.114 0.078 -15.994 -31.768 0.026 53.207 -31.835 0.026 53.814 -20.436 0.095 0.368
40.0 -20.391 0.096 -25.159 -32.146 0.025 45.630 -32.146 0.025 43.354 -19.643 0.104 -11.716
41.0 -19.220 0.109 -32.093 -32.542 0.024 36.224 -32.653 0.023 34.798 -18.771 0.115 -21.615
42.0 -17.951 0.127 -40.338 -33.191 0.022 27.538 -33.231 0.022 27.350 -18.223 0.123 -29.792
43.0 -17.133 0.139 -46.939 -33.893 0.020 18.674 -33.723 0.021 17.243 -17.835 0.128 -37.995
44.0 -16.397 0.151 -52.696 -34.379 0.019 10.281 -34.379 0.019 11.852 -17.278 0.137 -44.159
45.0 -15.762 0.163 -56.989 -35.090 0.018 -0.062 -35.289 0.017 -2.928 -16.905 0.143 -47.672
46.0 -15.045 0.177 -58.021 -35.863 0.016 -10.034 -35.972 0.016 -7.633 -16.671 0.147 -49.016
47.0 -14.289 0.193 -58.820 -36.595 0.015 -19.609 -36.536 0.015 -18.594 -15.923 0.160 -49.905
48.0 -13.457 0.212 -62.285 -37.856 0.013 -29.064 -37.458 0.013 -28.775 -14.685 0.184 -51.045
49.0 -12.027 0.250 -67.461 -38.344 0.012 -41.406 -37.924 0.013 -37.826 -13.568 0.210 -54.396
8
Attenuation is controlled by applying voltage to pins V1
and V2 as shown in Figure 13.
Figure 13. Bias voltage connections
Figure 14. AMMC-6650 and the op-amp driver circuit
Biasing considerations For the minimum attenuation, V1 is set to 1.5 V and V2 is
set to 0 V. The 1.5 V applied to the V1 pin biases the series
FETs to a full “on” state, while the 0 V applied to the V2
pin keeps the shunt FETs in an “o ” or “open” state; thus
creating the lumped element 50Ω transmission line
e ect. The V2 voltage swing from 0 V to 1.25 V increases
the level of attenuation. The V1 voltage swing from 1.5 V
to 0 V e ectively optimizes the input and output match at
higher attenuation levels. The AMMC-6650 can be driven
by two complementary voltage ramps placed on V1 and
V2. Careful adjustment of the two control lines over a rela-
tively small voltage ranges are required to set the attenu-
ation and optimize VSWR.
The on-chip DC reference circuit can be used to optimize
VSWR for any attenuation setting, improve voltage versus
attenuation linearity and range, and provide temperature
compensation.
The on-chip DC reference circuit is a non-distributed “T” at-
tenuator designed to operate in a 500Ω system and track
the control voltage versus attenuation characteristics of
the RF attenuator. A simpli ed schematic of the AMMC-
6650 together with an op-amp driver that utilizes the DC
reference circuit is shown in Figure 14.
RFin
V1
RFout
+
_
500
R
REF
(620)
R
S
500
VREF
VCONTROL
R
L
(500)
OP AMP 1
A
B
C1
DCin
+
_
R1 (10K)
R2 (100)
OP AMP 2
C
D
V2DCout
9
OP AMP 1 insures that the attenuator maintains a good
input and output match to 50Ω, while OP AMP 2 increases
the usable control voltage range versus using only direct
voltage ramps for V1 and V2 and improves over tempera-
ture operation.
If optimum VSWR is all that is required, OP AMP 2 may be
eliminated however, RL must remain connected to the
DCout
pad of the AMMC-6650 and the control voltage can be
applied directly to V2.
CAUTION: Low voltage op-amps must be used so as not to
exceed the maximum limit of V1 and V2 control voltages.
As shown, a voltage reference (VREF) is fed to the reference
circuit DCin pad via a 500Ω resistor, creating a 500Ω source.
The reference circuit termination RL, is connected to the
DCout pad and ideally is also equal to 500Ω. This voltage
is controlled in parallel with the RF attenuator. The chosen
value of VREF must be low enough to avoid modifying the
FET biasing and lower than the turn-on voltage of the ESD
protection diode but high enough such that the attenuat-
ed voltage at OP AMP 2 is usable compared to input o sets
etc. The optimum value for the positive reference voltage
is approximately 0.1 to 0.4 V.
At equilibrium, the voltages at nodes A and B of the OP AMP 1
must be equal which implies that the input impedance
to the DC reference circuit is equal to RREF. When V2 is
changed to a lower value, the voltage at node A becomes
greater than that of node B. This voltage di erence causes
the output voltage of op OP AMP 1 to move toward its
positive rail until equilibrium is once again established.
When V2 is changed to a higher value the voltage at node
A becomes less than that of node B and the output voltage
of OP AMP 1 will swing toward its negative rail until equi-
librium is reached. If the reference circuit precisely tracks
the RF circuit, the voltage output of OP AMP 1 at equilibrium
insures that the RF circuit is matched to 50Ω.
If attenuation linearity is required, OP AMP 2 is included
as shown in Figure 14 and a positive control voltage is
applied to VCONTROL.
At equilibrium, voltages at nodes C and D are equal.
When VCONTROL is changed, the output of OP AMP 2 adjusts
to a value that forces the voltage at node C to equal the
voltage at node D. Therefore, the output voltage of the
DC reference circuit is proportional to VCONTROL. The input
voltage to the reference circuit is being held constant and
the log(VCONTROL) is proportional to the reference circuit
attenuation 20log(DCout/DCin).
If the FET parameters of the DC reference circuit track the
FET parameters of the RF circuit, the voltage output of the
RF circuit is also proportional to the control voltage. This
translates to a linear relationship between the attenuation
(in dB) and the log(VCONTROL).
Two RF attenuation vs voltage curves corresponding
to di erent values of VREF are shown in Figure 15. These
curves were obtained by using the driver circuit shown in
Figure 14 and the VREF values 0.1 V and 0.4 V.
Values for RL, R1 and R2 were 500Ω, 10 kΩ and 100Ω re-
spectively. Control voltage ranged from 4.5 V to 0 V.
Because the FETs in the DC circuit are not identical to
those in the RF circuit, the DC circuit does not exactly track
the RF circuit. This results in attenuation vs. voltage curves
that are not exactly linear.
OP AMP 2 provides temperature compensation by adjusting
V2 in such a way as to keep voltage at point C equal to
that point D. If the attenuation changes over tempera-
ture, voltage at point C tries to change, but is corrected
by OP AMP 2.
Another way to improve performance of the attenuator
driver circuit is to adjust RL and RREF. If the reference circuit
precisely tracked the RF circuit and the ON resistance
of the FETs was zero ohms, then RL and RREF would be
exactly 500Ω. Due to the di erence in layout structures,
the reference circuit does not track the RF circuit precisely.
RL and RREF can be adjusted in order to compensate for
these di erences. Optimum values of RL and RREF have
been found to be between 500Ω and 650Ω.
For maximum dynamic range on the attenuation control
circuit, RL should be less than RREF by an amount equal
to the “ON resistance” of the reference circuit series FETs.
The “ON resistance” of the series FETs is about 95Ω total.
Therefore, the relationship between RL and RREF is as
follows:
RREF = RL + 95Ω
The voltage divider formed by R1 and R2 can be used
to adjust the sensitivity of the attenuator versus control
voltage. For the driver circuit shown in Figure 14, maximum
attenuation is always achieved by setting VCONTROL equal
to 0 V. Minimum attenuation is achieved when
Vcontrol ≈ x x Vref
or
Vcontrol ≈ x DCout
Therefore, an increase in the resistor ratio R1/R2 increases
the value of the control voltage required to produce
minimum attenuation.
R1 + R2
R2
⎛
⎜
⎝
⎛
⎜
⎝
RL
500 Ω + RL
⎛
⎜
⎝
⎛
⎜
⎝
R1
R2
⎛
⎜
⎝
⎛
⎜
⎝
1 +
For product information and a complete list of distributors, please go to our web site: www.avagotech.com
Avago, Avago Technologies, and the A logo are trademarks of Avago Technologies in the United States and other countries.
Data subject to change. Copyright © 2005-2008 Avago Technologies. All rights reserved.
AV02-1301EN - July 11, 2008
LMV932 (National Semiconductor) was used in the control
circuit that produced the results shown in Figure 15;
however, any low noise, low o set voltage op amp should
produce similar results. LMV932’s low supply voltage of
1.8 volts, limits the possibility of exceeding the 1.5 volt
absolute maximum of the AMMC-6650 V1 and V2 control
line inputs.
RF connections should be kept as short as reasonable to
minimize performance degradation due to undesirable
series inductance.
A single bond wire is normally su cient for signal con-
nections, however double bonding with 0.7mil gold wire
will reduce series inductance. Gold thermo-sonic wedge
bonding is the preferred method for wire attachment to
the bond pads. The recommended wire bond stage tem-
perature is 150°C +/- 2°C. Caution should be taken to not
exceed the Absolute Maximum Rating for assembly tem-
perature and time.
The chip is 100um thick and should be handled with care.
The MMIC has air bridges on the top surface and should be
carefully handled by the edges or with a custom collet, (do
not pick up the die with a vacuum on die center). Bonding
pads and chip backside metallization are gold.
This MMIC is also static sensitive and ESD precautions
should be taken.
Notes:
1. Ablebond 84-1 LMI silver epoxy is recommended
2. Eutectic attach is not recommended and may jeopardize reliability of
the device.
Bond Pad Dimensions and Locations
Part Number Ordering Information
Part Number
Devices Per
Container Container
AMMC-6650-W10 10 Gelpak
AMMC-6650-W50 50 Gelpak
Figure 15. Attenuation vs. Control Voltage, Frequency = 15 GHz
-50
-45
-40
-35
-30
-25
-20
-15
-10
-5
0
0.01 0.1 1 10
Control Voltage (Vin)
Attenuation (dB)
Vref=0.2V
Vref=0.1V
Assembly Techniques
The backside of the MMIC chip is RF ground. The chip
should be attached directly to the ground plane (e.g.
circuit carrier or heatsink) using electrically conductive
epoxy [1,2].
For best performance, the topside of the MMIC should be
brought up to the same height as the circuits surrounding
it. This can be accomplished by mounting a gold plated
metal shim (same length as the MMIC) under the chip. The
amount of epoxy used for the chip or shim attachment
should be just enough to provide a thin llet around the
bottom perimeter of the chip. The ground plane should
be free of any residue that may jeopardize electrical or
mechanical contact with the chip.
