UMTS PA from Fojutowo Microwave Meetup
The amplifier board comes from a commercial 2100-band cell amplifier. I'm planning to retune it to 2400 MHz for QO-100 DATV operation.
Board layout
- Signal input
- MW6S004NT1
- RF N-MOSFET
- wideband matching (datasheet says "up to 2000 MHz" - lol)
- 18 dB gain, P1dB >+37dBm (nice), 28V
- IL05AL(R)2140AAE
- 2.14G isolator "2G14 P4N"
- 0.6dB loss between 2110-2170 MHz
- 10W FWD max
- 2.4 GHz attenuation? I guess not more than 3dB & poor isolation
- X3C21P1-05S
- 5dB splitter for 2000-2300 MHz
- maybe wideband just enough - DS ends on 2300 MHz
- ideal replacement: X3C25P1-05S (2300-2700 MHz) - 4€ @ Mouser
- why 5 dB? need to figure out the layout (Doherty amplifier..?)
- AT2S15N / AFT20S015N
- 1800-2700 MHz, N-MOSFET
- ~17 dB gain @ 2100 MHz, P1dB 16W, 28V
- AFT21S240-12S
- N-LDMOS
- internal matching 2110-2170 MHz
- ~20dB gain, P1dB 230W, 28V
- datasheet page 7, 2400 MHz: >20dB gain, -10dB IRL
- [SKYFR-000819]
- 0.5 kW 2110-2170 MHz isolator
- to be replaced with a piece of semirigid coax?
Steps
- remove one of the small silver filters and measure its frequency response
- supply bias voltage to the respective stages, possibly by reusing the AMC7812 chip instead of homebrewing own bias solution
- read AMC7812 datasheet, check if it has some kind of internal memory
- identify left-hand-side connectors (where's the main VCC? top left?)