The right way to speed up RF device testing

Before RF testing can be performed, the DC operating conditions of these devices must be tested. For diodes, this includes forward voltage drop, feedback breakdown voltage, and junction leakage current. For transistors, this includes different junction breakdown voltages, junction leakage currents, collector or drain characteristics, etc. Choosing the right test instrument and setting it up properly can greatly speed up these test processes.

Instrument selection

Using various digital multimeters (DMMs), voltage sources, and current sources to implement the test will take up more rack space, require learning multiple command sets, and make system programming and maintenance more complicated. The trigger time becomes complicated, and the uncertainty of the trigger increases, while coordinating the operation of discrete instruments increases the communication traffic of the bus and reduces the test efficiency.

To solve these problems, the first step is to integrate several functions into one instrument. The source-measurement unit (SMU) integrates a precision voltage source, precision current source, voltmeter, and ammeter into one instrument, saving space and simplifying the operation between devices. The second step is to eliminate the communication delay between the instrument and the control computer.

Reduce communication overhead

随着仪器和计算机间的高速通讯成为可能，通过GPIB(IEEE-488总线)链接为测试的每个步骤提供命令和控制，使得测试系统自动化更为广泛。尽管这与以前相比有很大的进步，但还是具有明显的速度限制。首先，GPIB需要可观的通讯开销。GPIB用作实时测试的另外一个缺点是控制通常来自总线的另外一端-运行Windows操作系统的PC，Windows在通讯响应时具有显著的延时，并且不可预测，这使得在测试环境中使用PC作为唯一的控制器时，多个仪器的同步几乎是不可能的。

Figure 1: Measurement equipment setup for diode testing.

The solution to this problem is to preconfigure the instrument using GPIB and then let the instrument perform the test itself. Many modern instruments have source memory list programming capabilities that allow up to 100 complete test sequences to be set up and run without PC intervention. Each test can contain different instrument configurations and test conditions, which can include configuration of sources, measurements, conditional jumps, math functions, and pass/fail limit testing and storage functions. Some units can run in DC or pulse mode with different parameters and timing, making it possible to slow down more sensitive tests or speed up other tests to optimize the overall test time process.

When the instrument operates essentially autonomously, the role of the GPIB is to download the test program before testing and to upload the results to the PC after testing, neither of which interferes with the actual testing.

Instrument Trigger

To perform a simple current-voltage sweep (IV), the SMU outputs a series of voltages while measuring the corresponding currents. At each voltage level, the SMU first applies a voltage. Voltage changes in the circuit will induce a transient current, so setting an appropriate delay between stimulus and measurement is critical to test integrity. The instrument will automatically adjust the delay to produce the best results over a wide range. However, adding additional components to the test circuit, such as long cables, switch matrices, etc., will change the transient characteristics of the circuit. For high-resistance devices, longer test times are often necessary. In these cases, the user needs to define additional delays to maintain measurement integrity.

Diode Testing

Our first example includes a test instrument, a device handler, and a PC (Figure 1). Note how internal programming can be used to eliminate most GPIB communications to speed up testing.

Production testing of diodes includes a verification step to determine the polarity of the diode under test, followed by testing for forward voltage drop, reverse breakdown voltage, and leakage current.

Forward voltage drop is the voltage across a diode at some specified forward current. It is obtained by passing a specified current through the diode and then measuring the voltage across it. Reverse breakdown voltage (VRM or VBR) is the reverse voltage when the current suddenly increases infinitely. This is measured by applying a reverse current and measuring the voltage across the diode. The voltage read is compared to a specific minimum limit to determine whether the test passes or fails. Leakage current IR is sometimes also called reverse saturation current. IS is the current when a voltage less than the reverse breakdown voltage is applied to the diode. It is obtained by applying a specific reverse voltage and measuring the resulting current. Programs are written to set up the diode test in the memory location of the source/storage instrument, and then a trigger transmitted via the IEEE bus starts the execution. The instrument performs the operation according to the set program location in the memory without the intervention of the computer.

Figure 2: Two SMUs are typically used in transistor testing, the first between the HBT base and emitter, and the second between the emitter and collector.

RF power transistor test

Although there are many types of RF transistors, we will use the heterojunction bipolar transistor (HBT) as an example, but similar tests can be used for other devices. Since the transistor is a three-terminal device, two SMUs are usually required. Figure 2 shows two SMUs connected to the device, the first between the HBT base and emitter, and the second between the emitter and collector. To obtain the collector curve of the HBT, the base SMU is set to source current and measure voltage. After setting the first base current, the collector current is measured while sweeping the collector voltage. The base current is then increased by one step, and the collector voltage is swept again while measuring the collector current. This process is repeated until all collector IV curves are obtained for different base currents.

Synchronization of instruments

Since we want both instruments to be programmed (to avoid GPIB delays), we want all instruments in the test setup to be synchronized. Initially, this is not a problem. For example, if several SMUs have the same firmware and are programmed with the same test parameters, each step will take the same amount of time to execute. The difficulty comes with memory location recall and auto-ranging steps, which take an indeterminate amount of time.

In situations like this, an external, dedicated trigger controller is required to ensure that measurements from multiple instruments occur simultaneously. This is particularly useful when the test system uses equipment from different manufacturers, or even when it comes from the same manufacturer but has different triggering methods.

The process is as follows (the example used is based on Keithley instruments, but similar methods can be used for instruments from other manufacturers):

1. The trigger controller outputs a trigger signal to each instrument.

2. Call the source memory location from memory.

3. Enable source outputs for all instruments.

4. Each instrument executes according to the user-defined delay.

5. Once the delay operation is completed, each instrument outputs a trigger signal to the controller.

6. The trigger controller waits for the trigger signal output by each instrument (delayed output).

7. The trigger controller sends a trigger signal (measurement input) to each instrument.

8. Each instrument starts measuring operation.

9. After completing the measurement, each instrument sends a trigger signal to the controller.

10. The trigger controller waits for the trigger signal (measurement output) output by each instrument.

11. Return to step 1 to start the next test.

Specific transistor tests

There are usually two important breakdown voltages to measure for HBTs: The first is the collector-emitter breakdown voltage, which can be measured with the base open or shorted. Figure 3a shows the setup for measuring the collector-emitter breakdown voltage with the base open (BVCEO or V (BR)CEO), and Figure 3b shows the setup for measuring the collector-emitter breakdown voltage with the base shorted (BVCES or V(BR)CES). The other breakdown voltage is the collector-base breakdown voltage (BVCBO or V(BR)CBO), which is usually measured with the emitter open. Figure 3c shows this test setup. In these measurements, the source-measure unit sweeps the voltage across the HBT while measuring the current. The current will remain very constant until the breakdown voltage is reached, after which the current will suddenly increase.

Figure 3: a: Collector-emitter breakdown voltage, base open circuit;

b: Collector-emitter breakdown voltage, base short circuit;

c: Collector off current, ICBO, and collector-base breakdown voltage, emitter open circuit.

Other parameters that are usually measured for RF power transistors are the collector-emitter sustaining voltage, BVCEO(sus) or VCE(sus), the collector-emitter breakdown voltage when reverse biased at the base-emitter junction (BVCEV or BVCEX), and the emitter-base breakdown voltage when the collector is open (BVEBO).

Junction leakage current

It is also important to describe the leakage current when the device is off, because leakage current wastes power when the device is not operating and can reduce the operating time of battery-powered equipment. The most commonly measured leakage current parameter is the collector-off current (ICBO), measured between the collector and the base with the emitter open (Figure 3c). The base reverse bias leakage current, also called the emitter-off current or emitter-base-off current (IEBO), is the other most important leakage current and is the leakage current in the base when the device is off.