DIY Modular High Vacuum System
DIY Modular High Vacuum System
DIY Modular High Vacuum System
One of my most ambitious personal projects.
One of my most ambitious personal projects.
Background
Background
Background
From the first moment that I had interacting with a high vacuum system, I was completely captivated. On their own they are marvels of engineering; systems where the most minute levels of gas in the system can not only be detected, but can ruin an experiment. But arguably more incredible is the science that can be carried out in such a unique experiment. For me, I was enamored by electrostatic fusors, electron beams, physical vapor deposition, and more.
Knowing that I would like to run these experiments and use these tools in the future and being unable to afford a commercial high vacuum system, I decided that I wanted to make one.
From the first moment that I had interacting with a high vacuum system, I was completely captivated. On their own they are marvels of engineering; systems where the most minute levels of gas in the system can not only be detected, but can ruin an experiment. But arguably more incredible is the science that can be carried out in such a unique experiment. For me, I was enamored by electrostatic fusors, electron beams, physical vapor deposition, and more.
Knowing that I would like to run these experiments and use these tools in the future and being unable to afford a commercial high vacuum system, I decided that I wanted to make one.
Initial Thoughts and Design
Initial Thoughts and Design
Initial Thoughts and Design
Very early on, I had determined that a project of this magnitude would need to be broken down into smaller parts and some bounds had to be added. Importantly, I am a graduate student working with limited time and an even more limited budget. This meant that, as often as possible, I would need to purchase components second hand or build them from scratch myself.
With these thoughts in mind, I first wanted to try tackling the vacuum system itself. While I knew that I would eventually need a full two stage vacuum system to reach the true "high vacuum" regime, I started with a single stage pump. Or rather, the suction end of a refrigerator compressor.
Hooked up via a machined top plate to a mason jar with epoxy potted electrical feed through or copper plug, I pulled my first vacuum and demonstrated low pressure arc discharge.
Very early on, I had determined that a project of this magnitude would need to be broken down into smaller parts and some bounds had to be added. Importantly, I am a graduate student working with limited time and an even more limited budget. This meant that, as often as possible, I would need to purchase components second hand or build them from scratch myself.
With these thoughts in mind, I first wanted to try tackling the vacuum system itself. While I knew that I would eventually need a full two stage vacuum system to reach the true "high vacuum" regime, I started with a single stage pump. Or rather, the suction end of a refrigerator compressor.
Hooked up via a machined top plate to a mason jar with epoxy potted electrical feed through or copper plug, I pulled my first vacuum and demonstrated low pressure arc discharge.
Acquisition of Vacuum Pumps and Gauge
Acquisition of Vacuum Pumps and Gauge
Acquisition of Vacuum Pumps and Gauge
These initial experiments made it immediately obvious that a true vacuum system was going to be needed. At this time, I was able to acquire two pumps: a used Varian DS102 rotary vane pump and a used Edwards nEXT400IID turbo molecular pump.
After some general upkeep and some new seals, the Varian roughing pump was ready for use and pulling quickly to ~10^-2 torr. However, the turbo pump required a lot more work. Data is difficult to find but this pump was supposedly a working pull from a tightly integrated piece of high vacuum equipment so came with a non-standard vacuum flange. To remedy this a custom adapter was machined on a manual mill from a piece of aluminum stock to adapt the unknown flange to a standard ISO160 type. Additionally, an Arduino control unit was created to allow speed readouts and control over spin up/down parameters.
These initial experiments made it immediately obvious that a true vacuum system was going to be needed. At this time, I was able to acquire two pumps: a used Varian DS102 rotary vane pump and a used Edwards nEXT400IID turbo molecular pump.
After some general upkeep and some new seals, the Varian roughing pump was ready for use and pulling quickly to ~10^-2 torr. However, the turbo pump required a lot more work. Data is difficult to find but this pump was supposedly a working pull from a tightly integrated piece of high vacuum equipment so came with a non-standard vacuum flange. To remedy this a custom adapter was machined on a manual mill from a piece of aluminum stock to adapt the unknown flange to a standard ISO160 type. Additionally, an Arduino control unit was created to allow speed readouts and control over spin up/down parameters.
At the same time, a used MKS 901P vacuum transducer was purchased as the first purchased vacuum gauge could not measure below 10^-3 torr. This transducer similarly received a control unit for easy readout and powering. With these upgrades I finally crossed the high vacuum barrier with pressures below 10^-5 torr, maxing out my gauge's ability to measure.
At the same time, a used MKS 901P vacuum transducer was purchased as the first purchased vacuum gauge could not measure below 10^-3 torr. This transducer similarly received a control unit for easy readout and powering. With these upgrades I finally crossed the high vacuum barrier with pressures below 10^-5 torr, maxing out my gauge's ability to measure.
Construction of Primary Chamber
Construction of Primary Chamber
Construction of Primary Chamber
With a functional vacuum system tested and working, the construction of a proper high vacuum chamber could begin. With the turbo pump adapter having a ISO160 flange, it made the most sense to build something to directly attach to this. This resulted in the most cost effective solution being the purchasing of a stainless steel ISO160 Tee. This choice was not only cost effective, but also came with two full ISO160 flanges for future expansion. However, this tee's ability to add feedthroughs was quickly outgrown. This was remedied by cutting into the tee and welding in KF25 and KF50 stubs. The end result is an ISO160 tee with two free ISO160 flanges, two KF25 flanges, and two KF50 flanges.
With a functional vacuum system tested and working, the construction of a proper high vacuum chamber could begin. With the turbo pump adapter having a ISO160 flange, it made the most sense to build something to directly attach to this. This resulted in the most cost effective solution being the purchasing of a stainless steel ISO160 Tee. This choice was not only cost effective, but also came with two full ISO160 flanges for future expansion. However, this tee's ability to add feedthroughs was quickly outgrown. This was remedied by cutting into the tee and welding in KF25 and KF50 stubs. The end result is an ISO160 tee with two free ISO160 flanges, two KF25 flanges, and two KF50 flanges.
With the bulk construction complete, the chamber was ready to be used. However, it is nearly impossible to use a chamber without numerous feedthroughs. With most commercial feedthroughs costing >$100, my only option was to learn to make them myself. This topic is discussed in the below section
With the bulk construction complete, the chamber was ready to be used. However, it is nearly impossible to use a chamber without numerous feedthroughs. With most commercial feedthroughs costing >$100, my only option was to learn to make them myself. This topic is discussed in the below section