The EMPF Monolithic Microwave Integrated Circuit (MMIC) environmental coatings project included testing the performance of wafer-level near-hermetic coatings used on several RF devices. RF testing was performed on gallium arsenide RF amplifier test die at regular intervals after temperature humidity biased (THB) exposure. These die were packaged in pin grid arrays (PGAs) and this testing was performed using a manual probe testing set-up. It was performed in an efficient and accurate manner to prevent scheduling delays with the environmental equipment. Below are some tech tips that may be useful for developing an RF test for similar wafer open lid devices.

Design for Test

• Device testability should be designed into the test samples whenever possible. In the case of the MMIC Environmental Seal Project, a matrix of various coating options tested included uncoated, organic, silicon carbide and organic/silicon carbide coated devices. In order to test these devices the chip surface was exposed to the RF probes. A separate test pad chip area was implemented into the design to provide a stable platform for manual RF probing. Traditionally, hermetically sealed devices could not be included in this test group because the test pads for the devices were sealed under the lid.
  
  
• The devices were designed into a PGA package so that a pin and socket could be used during testing. The pin and socket set-up creates a solid platform for securing the devices during testing. At the same time, the sockets allow the devices to be easily removed after probe testing. Similar sockets were used for environmental testing to help facilitate quick and safe transfer of the devices from the test chambers to the storage boxes.
  
  
• Adequate spacing between devices is also a requirement for efficient testing. There should be no overlap between adjacent test pads and no hindrance between devices. In our case, a small chip resistor between FET test pads required lowering of the entire sample stage to prevent crashing the delicate RF probes during testing site changes.

Select Robust RF Test Equipment

• Having the proper test equipment is also an important part of efficient RF testing. This allowed fast transitions between samples, reproducibility, and excellent reliability. Equipment used for RF testing of FETs evaluated in the environmental coating project included a SUSS Microtec PM5 probe station fitted with RF probes. The probes were constructed using a MEMs process that makes them more robust than other types of probes. The test device was powered by a HP model 3610A power supply (VS1) and a Keithly model 2420 (VS2) source meter. Gain measurements were recorded by an Anritsu 37347D network analyzer which allowed measurements at specific frequencies. This set-up proved to be extremely consistent.
  
  
• A main concern when performing manual RF probing is maintaining the planarity of the probes during testing. It is extremely easy to apply excessive pressure on the pad surface and knock one of the signal-ground-signal probes out of alignment. Also, if the height of the test pad is inconsistent, partial contact may result. It is suggested that robust RF probes be utilized to prevent planarity problems. Each probe tip acted as a contact spring providing solid contact with pads of different heights.

Use the Correct Testing Parameters

• Some common RF testing parameters include S-Parameters, Voltage Standing Wave Ratio (VSWR), RF-intermediate frequency (IF) - local oscillator (LO) isolation, phase noise, carrier suppression, gain compression, I/O balance, conversion gain, IP3 (TOI), noise figure, and RF power. It is important to match the correct testing parameter with the proper device function. The S-parameters (scattering parameters) were used to measure device performance for the FETs tested in the environmental coatings project. These parameters are important for verifying proper signal transfer.