Benchtesting MaSCOT

[Apologies that this is the first post about MaSCOT. I’ll try to keep up a development blog _and_ backfill with add’l project notes and resources. Maybe.]

The full MaSCOT system has been assembled and is ready for tank testing.

IMG_0010

IMG_0014

I ran the sealed system on the desktop for the first time last night. While it was on I recorder internal air temperatures using the Bosch BMP280 on the companion board, plus the Linux ACPI temperature values.

jetson_temperatures

The system was idle for the majority of the test (overnight). I started using my recorder app at the end, hence the tail up.

The temperatures look relatively high, but this is with the system in air, not in water. This will sink heat away from the enclosure far better than air.

Also, there’s a clearly a bug with my code to read from sys (I’m using libudev, not a shell call), as it’s not updating its values. Back to the drawing board… Even with that issue, the on-die temperatures are somewhat lower than the air temperature, which I don’t believe for a second.