High Power Active Microwave Devices Face Many Hazards

Parent Category: 2015 HFE

By Tim Galla

Solid state RF and microwave amplifiers are increasing in power and power density. This greater threshold of power in a smaller area is raising the bar for other active components power handling capability. A challenge to maintain at high RF and microwave power levels is to handle higher powers as fast switching speed, good linearity, and wide bandwidth performance. Engineers working with high power systems in the telecommunications or aerospace/defense are accustomed to accounting for the many design constraints of high power at high frequencies. Test and measurement and research engineers are more often encountering high power levels to best handle developing materials and modeling high power technologies.

For any application calling for high levels of RF and microwave power, there are many design constraints and specialized technologies suited to high power systems. The design constraints associated with high power regime technologies are mostly based on preventing material damage, heating, and potential component failure. Knowledge of the limitations of each active device category is necessary during system and device design to avoid overpower and stress damage leading to failure and early aging.

Greater Complexity

Unlike passive devices, active devices have an additional level of complexity and high power challenges. One of these challenges is that active devices have many more states and operation modes than passive devices. Generally, this leads to active devices being much more, if not impossible, to fully characterize. Often, datasheets will only offer rough details on a preferred mode of operation, though it is implied that the device will operate similarly in other modes. The modulation method, either continuous wave (CW), pulsed, or complex modulation schemes must also be taken into account and the time dependent power and voltage characteristics of the signals will also impact device performance. A designer will have to be much more prudent in active device selection and prototype testing to ensure that the device truly does meet system requirements, especially under extreme power, temperature, and environmental parameters.

1511 HFE highPower01

Figure 1 • A research laboratory from Caltech emulated power amplifier device failure by using a high powered laser to test self-healing technologies. 

Source: http://phys.org/news/2013-03-caltech-self-healing-electronic-chips.html

Active devices also have power/bias, control, and often, sensors as a necessary part of device operation. With the addition of any components, the increase chance for high power devices to suffer from failures associated with the accessory electronics. Also, active device performance specifications, such as linearity, noise, and accuracy, are dependent upon these accessory electronics. In an high power environment, small degrees of nonoptimal compliance can dramatically impact downstream performance and induce failures. A savvy engineer will have thoroughly scrutinized the complete active device system during prototype and testing phase to uncover any common or specific device anomalies or failure modes.

High Power Amplifiers

Often considered the primary component of a high power RF and microwave system, the high power amplifier (HPA) essentially converts DC or AC power into a higher powered version of a stimulus signal. This means that a small stimulating signal on the range of microwatts or milliwatts can have the effect of kilowatts at the output of an HPA. Additionally, these signals are generally used to transfer information or as a radar/imaging technique were detailed knowledge of the output response is critical for accurate performance. Another important feature is that HPAs are commonly driven into the nonlinear region of operation, making predictions of performance challenging.

1511 HFE highPower02

Figure 2 • As the various materials used in the construction of RF and microwave devices each have different power and temperature handling capability, the “weakest” material or component is the limiting factor in a design.

Source: http://llis.nasa.gov/lesson/1796

These factors bring up the challenge of properly modeling, simulating, and measuring an HPA’s performance. As the power levels are so intense, there are very few instruments that can directly measure an HPA’s true performance. Also, tiny inaccuracies in measurement can lead to enormous impacts in high power systems. However, devices such as network analyzers, can contribute uncertainty, which could be large compared to the magnitude of the reflection coefficient. This is especially and issue with HPAs, as the real part of the input/output impedance can be below 1Ω. With a network analyzer with error on the order of 1Ω, this could lead to substantially different expectations and reality of device operation.

For example, an HPA’s input and output impedance is determined by many factors. The primary contributing factors that can be controlled with an HPA are the frequency and operating conditions. Of which, the temperature, mode of operation, bias conditions, age, and modulation content of the signal all contribute. All of these conditions and the various states these conditions have, create an extremely complex web of the complete system states. This makes HPAs largely impossible to fully characterize.

Moreover, there are certain device parameters, such as input and output impedance, that depend upon these characteristics. Commonly, HPAs are tuned during operation to ensure optimal impedance match. This optimal match will only be in place for a certain set of conditions, and will be nonoptimal as the conditions change. Understandably, this leads to a wide range of potential impedance, reflection, and VSWR behavior.

There are a few modern solutions to this challenge. Automated tuners can be used to measure HPA states much more rapidly, producing a more detailed understanding of the devices behavior under many modes/conditions. Enhanced nonlinear simulators are also increasing in their ability to accurate describe HPA device performance. With more accurate and detailed measurements alongside heightened nonlinear simulator performance, an HPA can be more accurately predicted and better tuned to be optimal under a wide range of conditions. Obviously, these methods still hinge on an engineer’s detailed knowledge of the environment in which the HPA will be operating and the desired performance constraints.

Switches and Switch Matrices

Switches are often employed in high power applications to avoid the need for having many HPAs. With modern radar, and even telecommunications, antenna arrays require a high density of switches, or switch matrices, to operate efficiently. In a high power scenario, the factors of greatest concern, are the amount of power the switch can pass or block, and the amount of signal degradation the switch will contribute. These factors are significant concerns in switch matrices, where a signal may pass through many switches before routing through the switch matrix output.

1511 HFE highPower03

Figure 3 •The degradation or failure of an interconnect component can lead to undesired device performance, as the impedance may have an effect on the tuning or configuration of the active device.

Source: http://www.microwaves101.com/mortuary/992-microwave-mortuary-2010

Generally used as the figure of merit for linear switch operation, the 1dB compression point, details the switches power limit before significant nonlinear performance begins. Most engineers will spec a switch with a 1dB compression point 10% to 10% beyond their predicted operational requirements to ensure linear modes of operation. If complex modulation schemes are being implemented, the signal integrity constraints may force the need for greater linearity headroom. 

1511 HFE highPower04

Figure 4 • Often, high power microwave electronics are housed in larger metal structures to aid in heat dissipation and thermal capacity.

Source: http://lss.fnal.gov/archive/2005/conf/fermilab-conf-05-107-td.pdf

Considerations should also be made for switch aging. Switch aging can be affected by time, power levels used, heat, shock/vibration, and environmental exposure. Aging may reduce the linear operation capability, power handling, insertion loss, and switching speed. This is the case for both mechanical and solid state switches.

In cases that the switch may be operated when RF power is at the input, the hot switching capability of the device needs to be evaluated. Under hotswitching conditions, the internal stresses on the transistors or mechanical contacts are dramatically increased. Voltage and current transients under hot switching can lead to carrier migration in solid states switches and arcing that causes debris build up on mechanical contacts. Thus, lifetime of switches can be reduced significantly this way. Also, the voltage and current transients during hot switching could exceed the power handling parameters and permanently damage the device if care isn’t applied.

1511 HFE highPower05

Figure 5 • Under high stress electrical conditions, certain effects, such as electrically welding metals and metallic sputtering, can arise.

Source: https://www.bicsi.org/uploadedfiles/bicsi_conferences/fall/2012/presentations/CONCSES_4C.pdf

Specifically for mechanical switches, environmental conditions could lead to moisture or particulate ingress into the switch body. Any interference with the switch contacts can enhance the aging effect and lead to premature failure. There are many types of sealing technology, though not all may be ideal for a specific application. Additionally, mechanical switches may build up heat when performing a high number of switch operations and carrying significant RF power. Often under these extreme conditions, a switches datasheet nominal performance may not be realizable. Ensuring proper environmental protections and thermal management could extend the switch’s life and maintain the best switch performance.

High Power Phase Shifters

Phase shifters are often used to aid in the combining of multiple HPA outputs or in coordinating phased array antenna inputs. As phase shifters operate at the output of HPAs, both solid state and mechanical phase shifters may need to deal with substantial power outputs under a wide range of signal modes. In over power conditions, the insertion loss and VSWR of a phase shifter may increase substantially. Amplitude balance and minimal reflections are also desired parameters of phase shifters, and having a phase shifter specified for the power operation thresholds of the design is especially critical in concern to these parameters.

The power handling of an electronic phase shifter is a function of the diodes reverse bias and the size of the diodes. Greater power levels often reduce the phase shifting range of solid state phase shifters. For mechanical phase shifters, the devices are limited by the resonant behavior of the L-C network. If the Q of the network is extremely high, the power handling capability of the mechanical phase shifter will be higher. Though, the resistive components of the network will result in insertion loss and heat dissipation primarily within the inductors of the phase shifter. Even with high Q inductors, depending upon the power levels, thermal management technologies, such as heatsinks, may be necessary to keep the inductors operating within nominal temperatures. As phase shifters tend to be sensitive to temperature and environmental factors, proper protections and thermal management may be necessary to maintain nominal phase shifter operation.

 Surge Suppressors

Especially with high power telecommunications and fixed radar applications, voltage and current transients due to adverse weather are unavoidable concerns. Lightning strikes often hit telecommunications towers at a rate of tens of times a year. This necessitates protections from electronic equipment from the extreme power and temperatures associated with lightning strikes. This is evident as the average peak power of a lighting strike is approximately one trillion watts.

Fortunately, the tower and enclosure designs are often designed to dissipate the majority of the electrical surge from a lightning strike. However, a significant amount of electrical energy may travel along the ground shielding of coaxial cables or the outer body of a waveguide. Unfortunately, these high performance RF transmission lines and waveguides may conduct surge currents directly into the sensitive RF electronics. On the other hand, the majority of energy of a lightning surge current is below 1 MHz. This means that surging electronics may be used to shunt the surge power without damaging the sensitive RF circuitry.

Surge suppressors are conveniently designed to do just that, without conflicting with the normal RF operation. Often found between base station enclosures and the coaxial cabling traveling to a communications tower, RF surge suppressors use surge activated breakdown devices. Types of these devices include, thyristor diodes, metal-oxide varistors (MOVs), or gas discharge tubes (GDTs), to shunt the surge energy under a high current event to a grounded plug. These devices aren’t the perfect defenders, as they do have a lifetime and can be overwhelmed by an extremely high surge event, possibly leading to cascade failures. Several surge suppressors are often used in series in different parts of the signal path to ensure redundant protection to sensitive electronics and prevent system downtime.

Conclusion

In addition to the design challenges associated with predicted high power RF and microwave active device performance, environmental factors heap greater constraints and considerations on top of an already complex task. Thus, detailed knowledge of the environmental factors where the high power system will be deployed make a significant impact on the success or failure of a system. When designing for high power applications, failure modes and device damage often drive system constraints, even before meeting necessary operational parameters. Also, each high power RF device must be carefully selected and proven to perform before being included in a high power system. 

References & Resources

http://llis.nasa.gov/lesson/1796

https://www.craneae.com/Products/Microwave/ApplicationNotes/PhaseShifters.pdf

http://www.microwavejournal.com/ext/resources/whitepapers/2012/Thermal-Management-of-RF-and-Digital-Electronic-Devices-Feb-22-2012-Rev5.pdf?1378753220

http://accelconf.web.cern.ch/accelconf/p05/papers/wpat050.pdf

http://www.freescale.com/files/rf_if/doc/app_note/AN1526.pdf

Microwave Engineering 4th Edition, David M. Pozar, John Wiley & Sons, Inc, 2012.

About the Author

1511 HFE highPower06
Tim Galla is Pasternack’s Product Manager for Active RF Components. Mr. Galla brings 25+ years of product development, applications engineering and business development expertise to his position. Prior to joining Pasternack, he held the position of Business Development Manager and Program Manager of RF/microwave components at Mercury Systems. Mr. Galla has also worked for several other industry leaders including Watkins-Johnson, Stellex Microwave Systems, Teledyne Microwave, Tyco Electronics and M/A-COM. He holds a degree in Electrical Engineering from a leading California University with an emphasis in advanced mathematics.