Gifford-McMahon Cryocooler Part I - Building The Helium Compressor
Hyperspace Pirate 16:32
124,286 views · 5,036 likes Watch on YouTube ↗
Radiacode – scientific gadgets for curious minds. Get 5% off with the code PIRATE2026 at https://radiacode.com/100-series.
In this video I'll be continuing my attempts at making liquid nitrogen with DIY parts by building the Helium Compressor that can be used for a Gifford-McMahon (or pulse tube) type cryocooler.
Previous G/M cryocooler video:
https://www.youtube.com/watch?v=Jj7Q7OqaW4A
The compressor will need to provide around ~300-400 psi of helium to the cold head of the cryocooler, but it also needs to provide a totally clean oilless stream of gas, so in addition to a standard oil separator, there's also a charcoal filter which adsorbs trace amounts of oil that can escape past the separator.
Helium has a very high thermal conductivity compared to air, so compressing it causes dramatically more heating than compressing air. Because of this, the compressor has to be submerged in a bath of water that's actively cooled by using a pump to circulate the water through a radiator with a fan. Running about 300 psi on the high pressure side, 100 psi on the low pressure side, and 1kW of input power, the compressor casing reached a steady state temperature just under 65C with an ambient temperature of 30C. So, it gets pretty hot but the temperature is still low enough to be manageable. Larger helium compressors require their coolant water to pass through a chiller.
In order to control the oil return solenoid valve and the inlet/outlet valves for the cold head, I built a driver board that takes the digital output of a microcontroller (arduino in this case, but it could be an ESP32, rPi, etc.) and drives the 24V solenoid coils. Each channel uses two FETs to ensure that even an extremely weak signal from the microcontroller like 2-3V can still reliably operate the valve.
In the next part of the video, I'll build and run the cold head and hopefully achieve a low enough temperature to make liquid air.
In this video I'll be continuing my attempts at making liquid nitrogen with DIY parts by building the Helium Compressor that can be used for a Gifford-McMahon (or pulse tube) type cryocooler.
Previous G/M cryocooler video:
https://www.youtube.com/watch?v=Jj7Q7OqaW4A
The compressor will need to provide around ~300-400 psi of helium to the cold head of the cryocooler, but it also needs to provide a totally clean oilless stream of gas, so in addition to a standard oil separator, there's also a charcoal filter which adsorbs trace amounts of oil that can escape past the separator.
Helium has a very high thermal conductivity compared to air, so compressing it causes dramatically more heating than compressing air. Because of this, the compressor has to be submerged in a bath of water that's actively cooled by using a pump to circulate the water through a radiator with a fan. Running about 300 psi on the high pressure side, 100 psi on the low pressure side, and 1kW of input power, the compressor casing reached a steady state temperature just under 65C with an ambient temperature of 30C. So, it gets pretty hot but the temperature is still low enough to be manageable. Larger helium compressors require their coolant water to pass through a chiller.
In order to control the oil return solenoid valve and the inlet/outlet valves for the cold head, I built a driver board that takes the digital output of a microcontroller (arduino in this case, but it could be an ESP32, rPi, etc.) and drives the 24V solenoid coils. Each channel uses two FETs to ensure that even an extremely weak signal from the microcontroller like 2-3V can still reliably operate the valve.
In the next part of the video, I'll build and run the cold head and hopefully achieve a low enough temperature to make liquid air.
Category (YouTube): People & Blogs
Playback is via YouTube's official embedded player. Data from YouTube; Exumo is not affiliated with YouTube.