. triquint semiconductor, inc ? phone 1-800-951-4401 ? fax: 408-577-6633 ? e-mail: info-sales@tqs.com ? web site: www.triquint.com april 2009 AH322 2w high linearity ingap hbt amplifier product features ? 400 ? 2700 mhz ? +33 dbm p1db ? +50 dbm output ip3 ? 13.4 db gain @ 2140 mhz ? 500 ma quiescent current ? +5 v single supply ? mttf > 100 years ? lead-free/rohs-compliant soic-8 package applications ? final stage amplifiers for repeaters ? high power amplifiers ? mobile infrastructure ? lte / wcdma / edge / cdma product description the AH322 is a high dynamic range driver amplifier in a low-cost surface mount package. the ingap/gaas hbt is able to achieve high performance for various narrowband- tuned application circuits with up to +50 oip3 and +33 dbm of compressed 1db power. it is housed in a lead- free/rohs-compliant soic-8 package. all devices are 100% rf and dc tested. the AH322 is targeted for use as a driver amplifier in wireless infrastructure where high linearity and medium power is required. the AH322 is ideal for the final stage of small repeaters or as driver stages for high power amplifiers. in addition, the amplifier can be used for a wi de variety of other applications within the 400 to 2700 mhz frequency band. functional diagram function pin no. iref 8 input 3 output / vcc 6, 7 vbias 1 gnd backside paddle gnd 2, 4, 5 specifications (1) parameter units min typ max operational bandwidth mhz 400 2700 test frequency mhz 2140 gain db 13.4 input r.l. db 14.7 output r.l. db 7.8 output p1db dbm +33 output ip3 (2) dbm +50 wcdma channel power (3) @ -50 dbc aclr dbm +24.1 noise figure db 4.8 vcc, vbias v +5 quiescent collector current (4) ma 500 iref ma 30 1. test conditions unless otherwise noted: 25oc, +5v vsupply, 2140 mhz, in tuned application circuit . 2. 3oip measured with two tones at an output power of +21 dbm / tone separated by 1 mhz, 940 mhz. oip3 measured with two tones at an output powe r of +24 dbm / tone separated by 1 mhz, 1960 mhz and 2140 mhz respectively. the suppression on the largest im3 product is used to calculate the oip3 using a 2:1 rule. 3. 3gpp wcdma, tm1+64dpch, 5 mhz offset, no clipp ing, par = 10.2 db @ 0.01% probability. 4. this corresponds to the quiescent current or op erating current under small-signal conditions into pins 6 and 7. absolute maximum rating parameter rating storage temperature -65 to +150 c rf input power, cw, 50 , t = 25oc input p 10 db device voltage, vcc, vbias +8 v device current 1400 ma device power 8 w thermal resistance, rth 18.6 c / w junction temperature, tj +200 c operation of this device above any of these paramet ers may cause permanent damage. typical performance parameter units typical frequency mhz 940 1960 2140 gain db 19.4 14.1 13.4 input return loss db 18 11.3 14.7 output return loss db 8.5 11.8 7.8 wcdma channel power (3) @ -50 dbc aclr dbm +23.6 +24.4 +24.1 output p1db dbm +33.0 +33.3 +33 output ip3 (2) dbm +47.6 +50.2 +50 noise figure db 8.5 4.5 4.8 vcc, vbias v +5 iref ma 30 quiescent collector current ma 600 500 500 ordering information part no. description AH322-s8g 2w high linearity ingap hbt amplifier (lead-free/rohs-compliant soic-8 pkg) AH322-s8pcb900 920 - 960 mhz evaluation board AH322-s8pcb1960 1930 - 1990 mhz evaluation board AH322-s8pcb2140 2110 - 2170 mhz evaluation board standard t/r size = 1000 pieces on a 7? reel. 1 2 3 4 8 7 6 5
. triquint semiconductor, inc ? phone 1-800-951-4401 ? fax: 408-577-6633 ? e-mail: info-sales@tqs.com ? web site: www.triquint.com april 2009 AH322 2w high linearity ingap hbt amplifier typical device data s-parameters (vcc = +5 v, i cq = 500 ma, t = 25 c, calibrated to device leads) 0 0.5 1 1.5 2 2.5 3 frequency (ghz) gain and maximum stable gain -5 0 5 10 15 20 25 30 35 40 45 2.14 ghz 7.19 db 2.14 ghz 22.7 db db(msg()) de_emebedded s_parameter db(|s(2,1)|) de_emebedded s_parameter 0 1.0 1.0 -1.0 10.0 1 0 . 0 - 1 0 . 0 5.0 5 . 0 - 5 . 0 2.0 2 . 0 - 2 . 0 3.0 3 . 0 - 3 . 0 4.0 4 . 0 - 4 . 0 0.2 0 . 2 - 0 . 2 0.4 0 . 4 - 0 . 4 0.6 0 . 6 - 0 . 6 0.8 0 . 8 - 0 . 8 s11 swp max 4ghz swp min 0.01ghz 2.14 ghz r 0.493722 x 0.825153 0 1.0 1.0 -1.0 10.0 1 0 . 0 - 1 0 . 0 5.0 5 . 0 - 5 . 0 2.0 2 . 0 - 2 . 0 3.0 3 . 0 - 3 . 0 4.0 4 . 0 - 4 . 0 0.2 0 . 2 - 0 . 2 0.4 0 . 4 - 0 . 4 0.6 0 . 6 - 0 . 6 0.8 0 . 8 - 0 . 8 s22 swp max 4ghz swp min 0.01ghz 2.14 ghz r 0.0838672 x 0.36526 notes: the gain for the unmatched device in 50 ohm system is shown as the trace in pink color, [db (s (2, 1)] . for a tuned circuit for a particular frequency, it is expected that actual ga in will be higher, up to the maximum stable gain. the maximum stable gain is shown in the blue line [db (gmax)]. the impedance plots are shown from 0.01 ? 4 ghz, wi th markers placed in 0.5 ghz increments. s-parameters (v cc = +5 v, i cq = 500 ma, t = 25 c, unmatched 50 ohm system, calibrated to device le ads) freq (mhz) s11 (db) s11 (ang) s21 (db) s21 (ang) s12 (db) s12 (ang) s22 (db) s22 (ang) 50 -0.74 -174.58 29.75 109.51 -43.47 25.51 -1.15 -1 35.3 100 -0.53 -179.31 24.21 98.59 -43.22 17.83 -1.22 -1 57.31 200 -0.45 176.71 18.46 89.55 -42.49 8.135 -1.19 -17 0.3 400 -0.44 170.07 12.77 80.82 -42.04 5.31 -1.22 -178 .39 600 -0.56 163.11 9.78 73.71 -41.41 10.52 -1.18 176. 07 700 -0.61 159.9 8.73 69.49 -41.21 11.31 -1.12 173.9 3 800 -0.64 156.03 7.94 65.68 -40.26 12.5 -1.17 171.6 9 1000 -0.78 147.66 6.8 56.95 -39.65 7.88 -1.22 166.8 2 1200 -0.87 138.49 6.11 46.99 -38.34 2.45 -1.26 162. 15 1400 -1.08 128.32 5.8 36.79 -37.99 -3.1 -1.33 157.2 9 1600 -1.4 117.39 5.83 25.05 -37.52 -14.57 -1.49 152 .31 1800 -1.94 106.19 6.17 10.83 -37.39 -27.07 -1.46 14 7.31 2000 -3.2 95.9 6.8 -7.89 -37.45 -42.22 -1.41 143.05 2200 -5.84 94.01 7.36 -33.75 -38.56 -69.38 -1.21 13 8.4 2400 -6.52 112.96 6.5 -64.88 -41.93 -115.31 -0.9 13 3.24 2600 -4.45 121.06 4.77 -92.83 -41.83 167.17 -0.4 12 6.56 2800 -2.44 117.78 2.24 -121.06 -38.13 103.07 -0.27 119.41 3000 -1.26 108.49 -1.12 -142.85 -34.99 62.15 -0.35 112.36 application circuit pc board layout circuit board material: top rf layer is .014? getek , ? r = 4.0, 4 total layers (0.062? thick) for mechanica l rigidity 1 oz copper, microstrip line details: width = .026? , spacing = .026? the silk screen markers ?a?, ?b?, ?c?, etc. and ?1? , ?2?, ?3?, etc. are used as placemarkers for the i nput and output tuning shunt capacitors ? c8, c5 and c2. the markers and vias are spaced in .050? increments.
. triquint semiconductor, inc ? phone 1-800-951-4401 ? fax: 408-577-6633 ? e-mail: info-sales@tqs.com ? web site: www.triquint.com april 2009 AH322 2w high linearity ingap hbt amplifier 824 - 894 mhz application circuit typical rf performance at 25 c frequency (mhz) units 824 848 894 gain db 19.7 19.7 19.7 input return loss db 16 16 13 output return loss db 7 8 12 output p1db dbm +33.0 +33 +32.6 channel power (1) (@-55 dbc is-95 cdma acpr) dbm +24.4 +24.4 +23.8 channel power (2) (@ -50 dbc wcdma aclr) dbm +23.7 +23.7 +23 output ip3 (3) (21 dbm / tone, 1mhz spacing) dbm +46.2 +46.3 +45.1 quiescent current, icq ma 600 vpd (4) v +5 vcc v +5 notes: 1. acpr test set-up: is-95 cdma, 9 channels fwd, 750 khz offset, 30 khz, meas bw, par = 9.7 db @ 0.01% prob. 2. aclr test set-up: 3gpp wcdma, tm164 dpch, 5mhz of fset no clipping, par = 10.34 db @ 0.01% probability. 3. oip3 is measured at 21 dbm / tone output power with 1 mhz spacing. 4. vpd is used as device power down voltage (low = rf off ). 5. the edge of l2 is placed at 265 mils from edge of a h322 rfout pin (12 o @ 850 mhz). 6. the edge of c2 is placed at 250 mils from edge of a h322 rfout pin (11 o @ 850 mhz). 7. the edge of c8 is placed at 25 mils from edge of ah 322 rfout pin (1 o @ 850 mhz). 8. do not exceed +5.5v supply or tvs diode d3 will be damaged. 9. zero ohm jumpers may be replaced with copper traces in the target application layout. 10. dnp implies do not place. small signal performance 15 16 17 18 19 20 800 820 840 860 880 900 frequency (mhz) g ain (d b ) -25 -20 -15 -10 -5 0 r e tu rn l o ss (d b ) gain s11 s22 aclr vs. output power 3gpp wcdma, tm1+64dpch, 5mhz offset freq., par = 1 0.34 % @ prob. -60 -55 -50 -45 -40 -35 -30 20 21 22 23 24 25 26 output power (dbm) ac lr (db c) 824 mhz 848 mhz 894 mhz acpr vs. output power is-95 cdma, 9 ch. fwd., 750 khz offset frequency, par = 9.7 db @ 0.01 % prob . -70 -65 -60 -55 -50 -45 -40 20 21 22 23 24 25 26 output power (dbm) a c pr (db c) 824 mhz 848 mhz 894 mhz oip3 vs. output power 1mhz spacing, 25c 35 38 41 44 47 50 18 19 20 21 22 23 24 25 26 output power / tone (dbm) oip3 (db m ) 824 mhz 848 mhz 894 mhz c8 c5 l2
. triquint semiconductor, inc ? phone 1-800-951-4401 ? fax: 408-577-6633 ? e-mail: info-sales@tqs.com ? web site: www.triquint.com april 2009 AH322 2w high linearity ingap hbt amplifier 920 - 960 mhz application circuit (AH322-s8pcb900) typical rf performance at 25 c frequency (mhz) units 920 940 960 gain db 19.2 19.4 19.2 input return loss db 16.6 18 15.3 output return loss db 7.8 8.5 9.4 output p1db dbm +33 +33 +33 channel power (1) (@-55 dbc is-95 cdma acpr) dbm +24.3 +24.4 +24.3 channel power (2) (@ -50 dbc wcdma aclr) dbm +23.5 +23.6 +23.5 output ip3 (3) (21 dbm / tone, 1mhz spacing) dbm +47.3 +47.6 +47.2 noise figure db 8.2 8.5 9 quiescent current, icq ma 600 vpd (4) v +5 vcc v +5 notes: 1. acpr test set-up: is-95 cdma, 9 channels fwd, 885 khz offset, 30 khz, meas bw, par = 9.7 db @ 0.01% prob. 2. aclr test set-up: 3gpp wcdma, tm164 dpch, 5mhz of fset no clipping, par = 10.34 db @ 0.01% probability. 3. oip3 is measured at 21 dbm / tone output power with 1 mhz spacing. 4. vpd is used as device power down voltage (low = rf off). 5. the edge of l2 is placed at 380 mils from the edge of AH322 rfout pin (19 o @ 940 mhz) 6. the edge of c2 is placed at 190 mils from the edge of AH322 rfout pin (9.5 o @ 940 mhz). 7. do not exceed +5.5v supply or tvs diode d3 will be damaged. 8. 0 jumpers may be replaced with copper traces in the target application layout. 9. dnp implies do not place. small signal performance 25 c 15 16 17 18 19 20 920 930 940 950 960 frequency (mhz) g a in (d b ) -25 -20 -15 -10 -5 0 r e tu rn l o s s (db ) s21 s11 s22 aclr vs. channel power 3gpp wcdma, tm164dpch, 5mhz offset, 940 mhz -60 -55 -50 -45 -40 -35 -30 17 18 19 20 21 22 23 24 25 26 27 output channel power (dbm) aclr (db m ) +25c -40c +85c aclr vs. channel power 3gpp wcdma, tm164dpch, 5mhz offset, 25c -60 -55 -50 -45 -40 -35 -30 17 18 19 20 21 22 23 24 25 26 27 output channel power (dbm) a clr (db c) 920 mhz 940 mhz 960 mhz acpr vs. channel power is-95cdma, 9 ch. fwd, 885 khz offset, 30 khz meas bw, 940 mhz -80 -75 -70 -65 -60 -55 -50 -45 -40 17 18 19 20 21 22 23 24 25 26 27 output channel power (dbm) acpr (dbc) +25c -40c +85c acpr vs. channel power is-95cdma, 9 ch. fwd, 885 khz offset, 30 khz meas bw, 25c -80 -75 -70 -65 -60 -55 -50 -45 -40 17 18 19 20 21 22 23 24 25 26 27 output channel power (dbm) ac pr (db c) 920 mhz 940 mhz 960 mhz gain vs. pout vs. temp freq. = 940 mhz 15 16 17 18 19 20 21 27 28 29 30 31 32 33 34 pout (dbm) g a in (d b ) 25c -40c +85c c8 c5 l2
. triquint semiconductor, inc ? phone 1-800-951-4401 ? fax: 408-577-6633 ? e-mail: info-sales@tqs.com ? web site: www.triquint.com april 2009 AH322 2w high linearity ingap hbt amplifier performance plots for AH322-s8pcb900 contd. oip3 vs. channel power freq. = 940, 941 mhz, 1mhz spacing 35 40 45 50 55 16 17 18 19 20 21 22 23 output power / tone (dbm) o ip3 (d bm ) 25c -40c 85c oip3 vs. channel power 1 mhz spacing, 25 c 35 40 45 50 55 16 17 18 19 20 21 22 23 output power / tone (dbm) o ip3 (db m ) 920 mhz 940 mhz 960 mhz noise figure vs. frequency 25c 0 2 4 6 8 10 920 930 940 950 960 frequency (mhz) n f (d b)
. triquint semiconductor, inc ? phone 1-800-951-4401 ? fax: 408-577-6633 ? e-mail: info-sales@tqs.com ? web site: www.triquint.com april 2009 AH322 2w high linearity ingap hbt amplifier 1930 - 1990 mhz application circuit (AH322-s8pcb1960 ) typical rf performance at 25 c frequency (mhz) units 1930 1960 1990 gain db 13.8 14.1 14.2 input return loss db 11.8 11.3 10.8 output return loss db 9 11.8 15.4 output p1db dbm +33.2 +33.3 +33.1 channel power (1) (@ -50 dbc wcdma aclr) dbm +23.9 +24.4 +23.7 output ip3 (2) (24 dbm / tone, 1mhz spacing) dbm +47.2 +50.2 +46.7 noise figure db 4.5 quiescent current, icq ma 500 vpd (4) v +5 vcc v +5 notes: 1. aclr test set-up: 3gpp wcdma, tm164 dpch, 5mhz of fset no clipping, par = 10.34 db @ 0.01% probability. 2. oip3 is measured at 24 dbm / tone output power with 1 mhz spacing. 3. the multilayer inductor l3 (82nh) is critical for linearity performance. 4. vpd is used as device power down voltage (low = rf off). 5. the edge of c5 is placed at 247 mils from the edge of AH322 rfout pin (11 o @ 1960 mhz). 6. do not exceed +5.5v supply or tvs diode d3 will be damaged. 7. 0 jumpers may be replaced with copper traces in the target application layout. 8. dnp implies do not place. small signal performance 25 c 10 11 12 13 14 15 1930 1940 1950 1960 1970 1980 1990 frequency (mhz) g a in ( d b ) -25 -20 -15 -10 -5 0 r e t u r n l o s s ( d b ) s21 s11 s22 aclr vs. channel power 3gpp wcdma, tm164dpch, 5mhz offset, 1960 mhz -65 -60 -55 -50 -45 -40 -35 20 21 22 23 24 25 26 27 output channel power (dbm) a c l r (d b c) +25c -40c +85c aclr vs. output power 3gpp wcdma, tm164dpch, 5mhz offset, 25c -65 -60 -55 -50 -45 -40 -35 20 21 22 23 24 25 26 27 output channel power (dbm) a c l r (d b c) 1930 mhz 1960 mhz 1990 mhz gain vs. pout vs. temp frequency = 1960 mhz 10 11 12 13 14 15 16 27 28 29 30 31 32 33 34 pout (dbm) g ain (d b ) +25c -40c +85c oip3 vs. output power 1 mhz spacing, 1960 mhz 35 40 45 50 55 20 21 22 23 24 25 26 27 output power / tone (dbm) o ip3 (dbm ) +25c -40c +85c oip3 vs channel power 1 mhz spacing, 25c 35 40 45 50 55 20 21 22 23 24 25 26 27 output power / tone (dbm) o ip 3 (db m ) 1930 mhz 1960 mhz 1990 mhz c8 c5
. triquint semiconductor, inc ? phone 1-800-951-4401 ? fax: 408-577-6633 ? e-mail: info-sales@tqs.com ? web site: www.triquint.com april 2009 AH322 2w high linearity ingap hbt amplifier 2110 - 2170 mhz application circuit (AH322-s8pcb2140 ) typical rf performance at 25 c frequency (mhz) units 2110 2140 2170 gain db 13.1 13.4 13.6 input return loss db 15 14.7 14.2 output return loss db 6.3 7.8 10 output p1db dbm +33 channel power (1) (@ -50 dbc wcdma aclr) dbm +24.1 +24.1 +23.8 output ip3 (2) (24 dbm / tone, 1mhz spacing) dbm +50.1 +50 +48.4 noise figure db 4.7 4.8 4.7 quiescent current, icq ma 500 vpd (4) v +5 vcc v +5 notes: 1. aclr test set-up: 3gpp wcdma, tm164 dpch, 5mhz of fset no clipping, par = 10.34 db @ 0.01% probability. 2. oip3 is measured at 24 dbm / tone output power with 1 mhz spacing. 3. the multilayer inductor l3 (82 nh) is critical for linearity performance.. 4. vpd is used as device power down voltage (low = rf off). 5. the edge of c5 is placed at 195 mils from the edge of AH322 rfout pin (22 o @ 2140 mhz). 6. the edge of c8 is placed at 0.5 mils from the edge of AH322 rfout pin (0 o @ 2140 mhz). 7. zero ohm jumpers may be replaced with copper traces in the target application layout. 8. dnp means do not place. small signal performance 25c 9 10 11 12 13 14 2110 2120 2130 2140 2150 2160 2170 frequency (mhz) s 2 1 (d b ) -25 -20 -15 -10 -5 0 r e tu r n l o s s (d b ) s21 s11 s22 aclr vs. channel power 3gpp wcdma, tm164dpch, 5mhz offset, 2140 mhz -65 -60 -55 -50 -45 -40 -35 20 21 22 23 24 25 26 27 output channel power (dbm) ac l r (d b c) +25c -40c +85c aclr vs. channel power 3gpp wcdma, tm164dpch, 5mhz offset, 25c -65 -60 -55 -50 -45 -40 -35 20 21 22 23 24 25 26 27 output channel power (dbm) a c lr (db c) 2110 mhz 2140 mhz 2170 mhz gain vs. pout vs. temp frequency = 2140 mhz 9 10 11 12 13 14 15 27 28 29 30 31 32 33 34 pout (dbm) g ain (db ) +25c -40c +85c oip3 vs. output power 1 mhz spacing, 25c 35 40 45 50 55 19 20 21 22 23 24 25 26 27 output power / tone (dbm) oip3 (dbm ) 2110 mhz 2140 mhz 2170 mhz oip3 vs channel power 1 mhz spacing, 25c 35 40 45 50 55 19 20 21 22 23 24 25 26 27 output power / tone (dbm) o ip3 (d bm ) +25c -40c +85c c8 c5
. triquint semiconductor, inc ? phone 1-800-951-4401 ? fax: 408-577-6633 ? e-mail: info-sales@tqs.com ? web site: www.triquint.com april 2009 AH322 2w high linearity ingap hbt amplifier mechanical information this package is lead-free/green/rohs-compliant. the plati ng material on the leads is nipdau. it is compatible with both lead-free (maximum 260 c reflow temperature) and lead (maximum 245 c reflow temperature) soldering processes. outline drawing mounting configuration / land pattern mounting config. notes 1. a heatsink underneath the area of the pcb for the mounted device is strictly required for proper thermal operation. damage to the device can occur without the use of one. 2. ground / thermal vias are critical for the prope r performance of this device. vias should use a .3 5mm (#80 / .0135?) diameter drill and have a final plated thru diameter of .25 mm (.0 10?). 3. add as much copper as possible to inner and ou ter layers near the part to ensure optimal thermal performance. 4. mounting screws can be added near the part to fasten the board to a heatsink. ensure that the gr ound / thermal via region contact the heatsink. 5. do not put solder mask on the backside of the pc board in the region where the board contacts the heatsink. 6. rf trace width depends upon the pc board mater ial and construction. 7. use 1 oz. copper minimum. 8. all dimensions are in millimeters (inches). ang les are in degrees. product marking the component will be marked with an ?AH322g? designator with an alphanumeric lot code on the top surface of the package. tape and reel specifications for this part are located on the website in the ?application notes? section. esd / msl information esd rating: class 1a value: passes 250v to < 500v test: human body model (hbm) standard: jedec standard jesd22-a114 esd rating: class iii value: passes 1000v min. test: charged device model (cdm) standard: jedec standard jesd22-c101 msl rating: level 3 at +260 c convection reflow standard: jedec standard j-std-020 functional diagram function pin no. iref 8 input 3 output / vcc 6, 7 vbias 1 gnd backside paddle gnd 2, 4, 5 1 2 3 4 8 7 6 5
|
