Pulse Tube Cryocooler (Part V)
Hyperspace Pirate 14:44
89,494 views · 6,825 likes Watch on YouTube ↗
🔥 Download War Thunder for FREE and get awesome bonuses! PC & Consoles — https://wtplay.link/hyperspacepiratewt and WT Mobile — https://wtm.game/hyperspacepiratewtm
This is Part 5 of my series on attempting to build and test a pulse-tube cryocooler, with the ultimate goal of reaching cold enough temperatures to liquefy air/nitrogen. In this part, I'm going to be testing two designs of pulse tube cooler(s) that are connected to a hermetic compressor on a closed loop using the Gifford-McMahon (valve-operated) cycle.
Previous videos in this series:
Part 1:
https://studio.youtube.com/video/GjRoThMyNGA/edit
Part 2:
https://www.youtube.com/watch?v=WOmjJFk8rl0
Part 3:
https://www.youtube.com/watch?v=cy8aGMH8Tz4
Part 4:
https://www.youtube.com/watch?v=vC2it8LHKSQ
The first design used 1/2" (~13mm) diameter copper pipe that was about 1' or 30 cm long (both the regenerator and pulse tube). The regenerator consisted of packed stainless steel wool at about 5% density. After experimenting with various inertance tube diameters and lengths and orifice valve settings, the minimum temperature achieved was only -11C with air and -15C with Helium.
Helium is typically considered the "magic" "near-ideal" gas for the purposes of gas-cycle cryocoolers, but I really didn't notice much difference in the cooling capacity. The only difference I noticed was that minimum temperature with Helium as the refrigerant happened at around 3x the frequency of air, presumably since the speed of sound in helium is about three times that of air.
The pressures I ran were about 100 psig on the low side and 250 psig on the high side, (7.8 / 18.2 barA). The solenoids I used were only rated for 150 psi, but since the pressure differential was kept within 150 psi, they still worked even at the 250 psig on the high side of the cycle.
The second design I tried was a 1" diameter x 4' long pipe (25.4mm x 1.22m). The same type of packed-steel wool regenerator was used down about half the length, with the remainder being for the pulse tube. In this case, the minimum temperature achieved was approximately -25C with the hot end reaching close to 60C. Cooling the hot end from 50-60C down to ambient gained 2-3 degrees of extra temperature drop on the cold side.
The performance of both designs was lower than that achieved with open-cycle compressed air in part IV of this series (-83C). However, part of that is because only about 200W was being put into the compressor, whereas the compressor from part IV was using well over 1 kW of power. The takeaway from this is that the acoustic network in the pulse tube is extremely sensitive to tuning, any being off by a little bit on any parameters will result in totally ineffective performance.
In my opinion, being able to avoid moving parts in the cold end of the system isn't worth the extra hassle of trying to tune the acoustics and pressure/mass flow phase shift of the pulse tube, so a standard Gifford-McMahon cycle with a mechanical displacer piston is probably a better option for amateurs and hobbyists attempting to get the coldest temperatures possible. Mechanical G-M cycles also typically outperform Pulse Tube G-M cycles by a significant amount.
In upcoming video(s) I'll be examining cooling with a DIY G-M cycle using a mechanical displacer piston.
This is Part 5 of my series on attempting to build and test a pulse-tube cryocooler, with the ultimate goal of reaching cold enough temperatures to liquefy air/nitrogen. In this part, I'm going to be testing two designs of pulse tube cooler(s) that are connected to a hermetic compressor on a closed loop using the Gifford-McMahon (valve-operated) cycle.
Previous videos in this series:
Part 1:
https://studio.youtube.com/video/GjRoThMyNGA/edit
Part 2:
https://www.youtube.com/watch?v=WOmjJFk8rl0
Part 3:
https://www.youtube.com/watch?v=cy8aGMH8Tz4
Part 4:
https://www.youtube.com/watch?v=vC2it8LHKSQ
The first design used 1/2" (~13mm) diameter copper pipe that was about 1' or 30 cm long (both the regenerator and pulse tube). The regenerator consisted of packed stainless steel wool at about 5% density. After experimenting with various inertance tube diameters and lengths and orifice valve settings, the minimum temperature achieved was only -11C with air and -15C with Helium.
Helium is typically considered the "magic" "near-ideal" gas for the purposes of gas-cycle cryocoolers, but I really didn't notice much difference in the cooling capacity. The only difference I noticed was that minimum temperature with Helium as the refrigerant happened at around 3x the frequency of air, presumably since the speed of sound in helium is about three times that of air.
The pressures I ran were about 100 psig on the low side and 250 psig on the high side, (7.8 / 18.2 barA). The solenoids I used were only rated for 150 psi, but since the pressure differential was kept within 150 psi, they still worked even at the 250 psig on the high side of the cycle.
The second design I tried was a 1" diameter x 4' long pipe (25.4mm x 1.22m). The same type of packed-steel wool regenerator was used down about half the length, with the remainder being for the pulse tube. In this case, the minimum temperature achieved was approximately -25C with the hot end reaching close to 60C. Cooling the hot end from 50-60C down to ambient gained 2-3 degrees of extra temperature drop on the cold side.
The performance of both designs was lower than that achieved with open-cycle compressed air in part IV of this series (-83C). However, part of that is because only about 200W was being put into the compressor, whereas the compressor from part IV was using well over 1 kW of power. The takeaway from this is that the acoustic network in the pulse tube is extremely sensitive to tuning, any being off by a little bit on any parameters will result in totally ineffective performance.
In my opinion, being able to avoid moving parts in the cold end of the system isn't worth the extra hassle of trying to tune the acoustics and pressure/mass flow phase shift of the pulse tube, so a standard Gifford-McMahon cycle with a mechanical displacer piston is probably a better option for amateurs and hobbyists attempting to get the coldest temperatures possible. Mechanical G-M cycles also typically outperform Pulse Tube G-M cycles by a significant amount.
In upcoming video(s) I'll be examining cooling with a DIY G-M cycle using a mechanical displacer piston.
Category (YouTube): People & Blogs
Playback is via YouTube's official embedded player. Data from YouTube; Exumo is not affiliated with YouTube.