MEASUREMENT DEVELOPMENT
high voltage measurements. For example, to prepare for the first prototype boards, we needed to acquire high voltage, high power supplies to power the circuits for debugging. We also needed to find (or in some cases, develop our own) DC and AC probes to make the high voltage measurements. Floating circuits are common in SMUs and some can float to the full voltage of 3kV. Significant analysis was required to determine where to place probes and which equipment to use to make specific measurements.
All equipment had to be properly grounded to ensure safe operation. The increased energy from the high voltage meant that any measurement mistake had the potential to damage the test equipment or the SMU circuits under test, leading to costly downtime for repairs. In many cases, a high voltage arc on the board meant several days of debugging and repair due to the number of components damaged. Until we could determine the source of a problem, we had to repeat the process of powering a unit, running a test until the arc occurred and hoping to detect its source, repairing the unit, and rerunning the test. In some cases, we had to work in a dark room to spot the source of the arc.
Simply making measurements was difficult given the high voltages involved. The usual procedure was to start with the power off, determine which test was needed, set the probes, power up the unit, and make a measurement. Moving a probe required powering off the unit, relocating the probe, and then repowering the unit to make the next measurement. Although this was an extremely time-consuming routine, it was the only way to work with floating high voltages safely. High voltages also made it impossible to handle a DUT under power; DUTs were located in an interlocked fixture that had to be opened (turning off the power to the fixture) before changing the DUT and rerunning a test.
When designing low voltage instruments, engineers can typically identify hot or damaged components by touch, but that wasn’t possible with the Model 2657A. A thermal video camera was invaluable for finding these components safely. When an arc occurred, this camera also helped us identify damaged areas of the circuit board quickly (Figure 1). Designing the Model 2657A SMU itself was only part of the process of high power/high voltage system development. Users needed an interlocked high voltage test fixture to connect to their DUTs and the sales team needed it to demonstrate the instrument’s capabilities safely. Special interface boxes were essential to connect multiple high and low voltage instruments’ LO terminals together safely, as well as special protection interface modules that could connect the HI terminals of multiple low and high voltage instruments and protect the low
voltage instruments from damage in case of a device failure. Finally, we had to engineer a way to cable and ground the multiple instruments and interface boxes together in a system with multiple configurations (Figure 2).
High voltage instrument design brings with it a whole new level of prototyping and testing challenges. Things that can usually be taken for granted, like the availability of suitable lab instruments, lab locations, and cabling, grounding, and safety interlock requirements all demand special consideration. Instead of a single PCB, we needed multiple smaller PCBs to meet voltage group spacing requirements for creepage and clearance. System debug was much more complex due to the difficulty of making measurements because of the high voltages involved. Even something seemingly as simple as connecting two instruments to the same DUT was more complicated.
These complications also made it difficult to avoid schedule slippage. Despite all of the challenges that complicated the process, we created a high voltage system design that combines high safety for users and test equipment with the characterization capabilities that the next generation of power semiconductors demand. At the same time, we expanded our own product development capabilities substantially, developing new skills and practices that we can apply to the next high voltage design project we undertake.
© 2013 Angel Business Communications. Permission required.
Issue I 2013
www.siliconsemiconductor.net 33
Figure 2. A test system configuration illustrating cabling and connections for measuring MOSFET characteristics with the Model 2657A and another Keithley SMU
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42