Abstract

Solid State Power Amplifiers (SSPAs) incorporating GaN transistors provide compact and efficient sourc-es of microwave power. CPI has developed a 1.5 kW X-band SSPA, CPI model VSX3622, for radar applica-tions. The VSX3622 SSPA combines the power from two VSX3614 SSPAs, which each operate at nominal-ly 1 kW, saturated. This paper will present details of the design and performance data for both amplifiers.

Background

The Beverly Microwave Division of Communications & Power Industries LLC (CPI BMD) has been manu-facturing microwave and radar components for more than 60 years. Communications & Power Industries LLC (CPI) is a global company with a world-class global network of service centers. CPI BMD devel-ops, manufactures, and repairs radar components and systems having full compliance to military stan-dards. Our design and manufacturing processes are geared for military as well as high-reliability commer-cial workmanship. CPI BMD is an ISO 9001/AS9100 certified manufacturer.

VSX3614 SSPA Design

For a given output power, size, weight, and efficiency are critically important for mobile applications. Solid state power amplifiers incorporating GaN devices have quickly found a home in mobile applications due to the high power density and efficiency of the GaN transistors. CPI has capitalized on these attributes for the development of the VSX3614, the VSX3630, and the VSX3622 X-Band SSPAs for radar applications.

Figure 1 VSX3614 Solid State Power Amplifier

The VSX3614 design has been optimized for duty cycles to 10% duty and to allow multiple amplifiers to be combined efficiently using waveguide combiners to produce a multi kilowatt transmitter. The VSX3630 design has been optimized for duty cycles up to 20%.

Figure 2 VSX3630 Solid State Power Amplifier

VSX3614 Electrical Design

The VSX3614 design is based on a 100 watt pow-er device. The system block diagram is shown in Figure 3.

The VSX3622, a 1.5 kW X-Band GaN Power Amplifier for Radar Application

George Solomon, Dave Riffelmacher, Matt Boucher, Mike Tracy, Brian Carlson, Todd Treado

Communications & Power Industries LLC, Beverly Microwave Division

The design consists of a two stage pre- driver with power split two ways and then amplified. Each sec-ondary driver feeds a 6-way radial power divider and the 12 power stages. The signals are then recom-bined in two 6-way isolated, radial combiners. The isolated radial in-phase combining structure is used to sum the powers from individual transistors while maintaining isolation between adjacent devices. The combiners provide greater than 20-dB return loss for the transistors.

The 12 power devices are mounted directly to the heat exchanger to provide the best thermal interface with the lowest thermal resistance and optimum heat spreading. Figure 4 shows a computational simu-lation of the temperature profile of the VSX3614 at the transistor to baseplate interface. The maximum temperature rise is less than 3 °C.

Figure 4 Thermal profile of VSX3614 SSPA

The SSPA’s cascaded gain and output power at Psat is shown in Figure 5.

Figure 5 Cascaded gain and output power for VSX3614 SSPA

The CPI-proprietary combiners enable an overall am-plifier efficiency of greater than 15% in the VSX3614 SSPA, where overall amplifier efficiency is defined as the ratio of RF output power to DC input power. The RF bandwidth is greater than 20%

The VSX3614 SSPA is O-ring sealed and internally temperature compensated. Packaged GaN FETS and MMICS ensure high reliability under extreme en-vironmental conditions. Table 1 summarizes the data for the VSX3614 SSPA.

Node

C P , C G A I N ( d B m )

0 7

14 21 28 35 42 49 56 63 70

Node 1 22 31 3 9 8 20 15 54 55 34 74 77 14 2

X-Band SSPA Cascaded Power and Gain

Coupler 1_1 Coupler 1_2 -180° 0° WG_COM B_1

Radi al_2 RFAmp_12 Isol ator_7 Split3 RFAmp_14 Isol ator_14 -180° 0°

Split20_1 RFAmp_Dri ve_2 Isol ator_16 RFAMP_DRIVE_1 Isol ator_15 CP CGAIN

SSPA

SSPA OUTPUT POWER Node 2, 60.475 dBm

Node 2, 45.21 dB

Figure 3 Schematic design of VSX3614 and VSX3630 SSPAs

ZO=50REV_PO_DET

ZO=50FWD_POW_DET

ZO=50Port_2

Radial_2

RFAmp_10

RFAmp_9

RFAmp_8

RFAmp_11

RFAmp_12

RFAmp_7

RFAmp_6

RFAmp_5

RFAmp_4

RFAmp_3

RFAmp_2

RFAmp_1

Radial_1

-180°

0°

WG_COMB_

Coupler1_1 Coupler1_2

Isolator_7

Isolator_8

Isolator_9

Isolator_10

Isolator_11

Isolator_12

Isolator_4

Isolator_5

Isolator_1

Isolator_2

Isolator_3

Isolator_6 Split6_1

RFAmp_1

Isolator_14

RFAmp_1

Isolator_13

-

0°

Split20_1 CW CWSource_1 RFAmp_Drive_2

Isolator_15 Isolator_16 RFAMP_DRIVE_

Split3

Table 1 Key Parameters for VSX3614 SSPA

Output power as a function of frequency and tem-perature is plotted in Figure 6 for the VSX3614 SSPA. This data was taken at 100 µs pulse widths at 10% duty.

Figure 6 VSX 3614 output power at room temperature

VSX3614 SSPA Life Test

CPI tests the GaN transistors and the SSPA design under pulsed RF conditions to validate the robustness of the devices and amplifier design. A 1000+ hour life test was conducted on the VSX3614 amplifier while operating at 100 µs pulse width and 5% duty factor. The VSX3614 SSPA was operated, at ambient temperature, in a rack assembly that mimicked the cooling air flow of a system configuration. The RF output was monitored for peak power using a USB pulsed-power sensor and the phase stability was monitored using a quadrature-detector-type phase bridge. The power and phase data was automatically logged every 5 minutes for the duration of the life test. Power supply voltages and currents and the ambient temperatures were recorded periodically.

CPI’s 1.5 kW Power Amplifier, the VSX3622

Figure 7 VSX 3622 is two VSX3614 SSPAs combined with fan box and slots for power supply

The VSX3622 amplifier was designed for mobile, air-cooled applications. As such, the overall size and weight and efficiency are driven by the choice of the power combiners. The generation of the 1.5 kW of output power from the coherent addition of two lower power SSPA bricks. The output combiner used for this amplifier is a half-height WR90 magic T with a load port for combining isolation. The system will also include a waveguide isolator, forward and reverse power samplers in the system packaging.

Table 2 Key Parameters for VSX3622 SSPA

CPI conducted a temperature test on the VSX3622 SSPA. The pair of amplifiers comprising the VSX3622 SSPA were mounted on a chassis and cooled with a fan mounted in the box at the end of the enclosure. As shown in Figure 7, the two adjacent slots will house the system power supply cooled by the same fan. The thermal profile shown in Figure 8 depicts the chamber ambient air in blue and the output from four

Frequency Range 7.6 to 9.6 GHzPeak RF Power 1 kW, saturatedPulse Width 0.2 to 100 microsecondSmall Signal Gain 50 dB small signal,45 dB

at nominal output power

Duty Cycle 10%Pulse Droop 0.5 dBOutput Power Flatness 1 dB, over selected

bandwidthsHarmonic Output -40 dBc maximumInter Pulse NoisePower Density

-165 dBc/Hz maximum

Prime Power 42 VDC at 13 AmpsWeight 11 pounds, including

air heat exchanger

Frequency Range 7.6 to 9.6 GHzPeak RF Power 1.5 kW, saturatedPulse Width 0.2 to 100 microsecondSmall Signal Gain 50 dB small signal, 45 dB

nominal output power

Duty Cycle 10%Pulse Droop 0.5 dBOutput Power Flatness 1 dB, over selected

bandwidthsHarmonic Output -40 dBc maximumInter Pulse NoisePower Density

-165 dBc/Hz maximum

Prime Power 42 VDC at 25 Amps@ 10% Duty Cycle

VSX3614 Saturated Power

55

56

57

58

59

60

61

62

63

64

65

7.6 8.1 8.6 9.1 9.6

Out

put P

ower

dBm

Frequency GHz

VSX 3614 Output Power

thermocouples; two mounted on the body of the units and the other two mounted in the output air plenum. The graph shows that the unit operating at 100 µs pulse width and at 7% duty has a temperature rise of 15 -18 °C above ambient air temperature.

Figure 8 Temperature profile for VSX 3622

Data was measured and recorded every minute in the profile while frequency swept data was measured periodically. The RF output power was plotted across frequency and temperature.

Figure 9 Power vs. frequency and temperature for VSX3622

Power data from the pulsed power meter was mea-sured at the 5% and 95% portion of the 100 µs pulse to record pulse amplitude droop. The RF output power droop is plotted as a function of frequency and temperature in Figure 10.

Figure 10 Pulse amplitude droop vs. temperature for VSX3622

Output power versus input power and frequency was measured at the temperature extremes and 25 °C.

Figure 11 Output power vs. frequency and input power at 25 °C for VSX3622

Figure 12 Output power vs. frequency and input power at -40 °C for VSX3622

−40 −20 0 20 40 600.0

0.2

0.4

0.6

0.8

1.0

[dB

]

7600.0MHz8200.0MHz8900.0MHz9600.0MHz

Pulse Droop vs Temperature @ Frequency

Temperature [C]

0 1 2 3 4 5 6 7−40

−20

0

20

40

60

80

[C]

Temperature vs Time

Time [Hours]

7600 7800 8000 8200 8400 8600 8800 9000 9200 9400 960060.0

60.5

61.0

61.5

62.0

62.5

63.0

63.5

64.0

64.5

[dB

m]

-40.0C-30.0C-20.0C-10.0C0.0C10.0C20.0C30.0C40.0C50.0C55.0C

RF Power Out vs Frequency @ Temperature

Frequency [MHz]

0

500

1,000

1,500

2,000

2,500

7.6 7.7 7.8 7.9 8.0 8.1 8.2 8.3 8.4 8.5 8.6 8.7 8.8 8.9 9.0 9.1 9.2 9.3 9.4 9.5 9.6

Wat

ts

Frequency, GHz

-6dBm

-5dBm

-4dBm

-3dBm

-2dBm

-1dBm

0 dBm

VSX 3622 X Band SSPA at 25°C

0

500

1,000

1,500

2,000

2,500

7.6 7.7 7.8 7.9 8.0 8.1 8.2 8.3 8.4 8.5 8.6 8.7 8.8 8.9 9.0 9.1 9.2 9.3 9.4 9.5 9.6

Wat

ts

Frequency, GHz

VSX 3622 X Band SSPA at -40oC

-6dBm

-5dBm

-4dBm

-3dBm

-2dBm

-1dBm

0 dBm

Figure 13 Output power vs. frequency and input power at 55 °C for VSX3622

The data measured at 55 °C indicates the design could use more drive power at the higher end of the band to fully saturate the output devices.

Pulsed waveforms are shown in Figures 14 and 15 at an operating frequency of 9.0 GHz. Figure 14 shows the output power of a 10 µs pulse at a pulse repeti-tion frequency of 1 kHz. Figure 15 shows the output power of a 100 µs pulse at a pulse repetition frequen-cy of 1 kHz. Power is measured with a pulsed pow-er meter. These waveforms were taken at ambient temperature.

Figure 14 Power at 1 µs pulse width and 1 kHz pulse repetition frequency.

Figure 15 Power at 100 µs pulse width and 1 kHz pulse repetition frequency

In Figure 16 the pulse phase droop data is plotted. Figure 17 shows the AM and PM noise data.

Figure 16 Pulse droop and phase data as a function of time across a 100 µs pulse for VSX3622

Figure 17 AM and PM noise for VSX3622

Figure 18 Four-amplifier module providing 2.5 kW at X-band

Summary

CPI has developed and extensively tested the VSX3614, the VSX3630 and the VSX3622 SSPAs. CPI has demonstrated efficient and compact combin-ing of multiple amplifiers at X-band. These amplifiers extend CPI’s proud heritage of high-power, high-reli-ability RF transmitters into a new technology regime. CPI’s GaN SSPAs can be readily combined into amplifiers with other form factors for power levels from 1 kW to 20 kW in a cost-effective manner at frequency ranges from L-band to X-band. Figure 18 shows one such form factor with four SSPAs power combined to generate 2.5 kW at X-band.

0 20 40 60 80 100−0.05

0.000.050.100.150.200.250.300.35

0 20 40 60 80 100−1.0

−0.5

0.0

0.5

1.0

1.5

2.0

2.5

5 to 99% Pulse Top Data

Time [uS]

Phas

e [d

eg]

I & Q

[V]

−160

−150

−140

−130

−120

−110

−100

−90AMPM

AM and PM Noise

Combined

Frequency [Hz]

[dB

c/H

z]

0

500

1,000

1,500

2,000

2,500

7.6 7.7 7.8 7.9 8.0 8.1 8.2 8.3 8.4 8.5 8.6 8.7 8.8 8.9 9.0 9.1 9.2 9.3 9.4 9.5 9.6

Wat

ts

Frequency, GHz

VSX 3622 X Band SSPA at +55oC

-6dBm

-5dBm

-4dBm

-3dBm

-2dBm

-1dBm

0 dBm