Engineer Kerry D. Wong has published a teardown of the Amrel PPS 35-2 single-channel programmable power supply — and in doing so solved a mystery discovered during a teardown of the dual-channel PPS 2322 back in 2015.
Although the PPS 35-2 is only rated at half of the total power dissipation (35V, 2A single channel) compared to the 2322 (35V, 2A dual channel)," Wong writes, "they weigh just about the same. So it would be curious to see what the difference are inside."
The answer is, perhaps surprisingly given that the two are quite different externally, is that internally the two power supplies are incredibly similar. "The layout is almost identical," Wong writes. "The only obvious difference is the auxiliary power supply riser board and the slightly larger heatsink in the PPS 2322. Both power supplies use the same size toroidal core which explains why the PPS 35-2 is only marginally lighter than its dual-channel cousin.
"Clearly, [the digital controller] board’s design is shared between both the single channel and the dual channel versions of the power supply. Like the controller board found in the PPS 2322, the processor board contains a CPLD (GAL16V8B) chip, two SRAM chips (MPM81C55P-2 is a 2K SRAM and M58725P is a 16K SRAM), and a 80C31 microprocessor (DMC60C31E, which is essentially a ROM-less 8051). The NI chip (CP02613) is presumably the GPIB controller (although I could not find any information on this chip). The firmware is stored in the UV-erasable EPROM (the firmware version of my unit is 7.00B)."
"The reuse of the board designed helped to fill in a gap in the teardown of the dual-channel model, too. "In my teardown of the PPS 2322, three chips next to the process board socket had their markings sanded off. In this power supply though we can clearly see what these three chips are in the picture to the right below," Wong notes. "The two 16 pin DIP ICs are TC4015, dual 4-stage static shift registers. And the 18 pin IC to the right is an Analog Devices’ AD7541 12-Bit monolithic multiplying digital to analogue converter chip (DAC)."
More details are available on Wong's blog post, along with a dump of the EEPROM.