DIY glass capillary tubing?
Posted: Sun Aug 04, 2013 11:36 pm
It was a surprise to me, when a forum search today found no mention of homebrew glass capillary tubing as a gas flow restrictor.
So I went and tried it, even though I know how to get metal capillaries.
Had some bits of relatively narrow soft glass tubing -- about 2.9 mm OD, 1.3 mm ID. Heated the middle of a piece with a regular pencil-flame propane torch, then drew it out into a narrow thread. In all of today's experiments, I cut the thin section and slipped an air hose on a big end.
Later realized that the thin section didn't need to be cut. For glueless external connections, both glass ends could be kept at the original OD. A rigid sleeve of larger tubing could protect the transition and thin sections. Another option is to use only a section of thin stuff, passing it through a small hole in what would otherwise be a dead-end fitting.
Flow was tested by blowing bubbles in water.
Later, it was measured by displacement of water from a bit of soda straw, closed with hot-melt glue.
The first "nozzle" was about 4 inches long, and flowed too much.
For #2 I carefully re-heated and re-stretched the other end of draw #1.
Still too much flow.
Third time was charmed. I made and used a thin section 22 inches long, with OD of around 0.007 inches. With inlet at 10 psig and outlet at atmospheric pressure, the bubble stream amounted to 1/2 cubic centimeter per minute. With inlet at 15 psig (for more pictures), after about 5 minutes the glass tube popped out from the clear plastic hose, without breaking the hollow fiber!
To avoid cloggage, how 'bout cutting the hollow fiber at its thinnest place, and using that as the gas inlet end? It would not surprise me if flow rate is sensitive to direction of flow. To measure sccm flows with outlet under vacuum, I would use a clear inlet hose and dip the other end in some oil.
Other reports, suggestions, or critiques are welcome!
-Rich
So I went and tried it, even though I know how to get metal capillaries.
Had some bits of relatively narrow soft glass tubing -- about 2.9 mm OD, 1.3 mm ID. Heated the middle of a piece with a regular pencil-flame propane torch, then drew it out into a narrow thread. In all of today's experiments, I cut the thin section and slipped an air hose on a big end.
Later realized that the thin section didn't need to be cut. For glueless external connections, both glass ends could be kept at the original OD. A rigid sleeve of larger tubing could protect the transition and thin sections. Another option is to use only a section of thin stuff, passing it through a small hole in what would otherwise be a dead-end fitting.
Flow was tested by blowing bubbles in water.
For #2 I carefully re-heated and re-stretched the other end of draw #1.
Third time was charmed. I made and used a thin section 22 inches long, with OD of around 0.007 inches. With inlet at 10 psig and outlet at atmospheric pressure, the bubble stream amounted to 1/2 cubic centimeter per minute. With inlet at 15 psig (for more pictures), after about 5 minutes the glass tube popped out from the clear plastic hose, without breaking the hollow fiber!
To avoid cloggage, how 'bout cutting the hollow fiber at its thinnest place, and using that as the gas inlet end? It would not surprise me if flow rate is sensitive to direction of flow. To measure sccm flows with outlet under vacuum, I would use a clear inlet hose and dip the other end in some oil.
Other reports, suggestions, or critiques are welcome!
-Rich