4
$\begingroup$

I am planning to make a small-scale ionic thruster, however, I am not planning to invest in a vacuum chamber so I am thinking of only achieving vacuum in the plasma chamber itself, by sealing the outlet with an elongated tube so that the plasma still have somewhere to go.

$\endgroup$
3
  • 3
    $\begingroup$ Welcome to the world of vacuum -- it can be an expensive place -- or an endlessly jury-rigged, silver-solder-blobbed, and wax-slathered one. Note that vacuum chambers, especially smallish ones, may be less expensive than one might expect. $\endgroup$
    – ikrase
    Jan 18, 2023 at 8:19
  • 1
    $\begingroup$ An additional note: This is going to inevitably involve (at least moderate) high voltages. While the relevant voltages (3 to 10 kV, potentially from a Neon Sign Transformer) are not overwhelmingly dangerous, they definitely can kill (especially if there is current or capacitance behind them) and they don't act like low voltages or line voltage -- can jump gaps or burn through insufficient insulation (most wire insulation is only rated for 300 or 600 volts). Much higher voltages, above 20 kV, can unleash a variety of weird and hazardous effects and there is no substitute for proper training. $\endgroup$
    – ikrase
    Jan 18, 2023 at 8:27
  • 1
    $\begingroup$ .... Many people on Youtube, especially using microwave oven transformers, are reckless and do not have a good handle on the hazards! $\endgroup$
    – ikrase
    Jan 18, 2023 at 8:28

2 Answers 2

1
$\begingroup$

Partial answer to get the ball rolling:

When we make plasmas in gas, we either make a crazy powerful and destructive discharge at atmospheric pressure (e.g. arc welding, lightning, forks in microwave ovens) or we do it at low pressure of several Torr where Paschen's law tells us breakdown and discharge is a heck of a lot easier, lower power, less destructive and much more beautiful (e.g. fluorescent lights, neon signs + bulbs (remember those?), plasma globes) and of course ion thrusters.

You will need to get to this low pressure (somewhere in the roughly 5 to 20 Torr range) in order to make a continuous plasma without dumping in enough power to melt your project in seconds. If you can find anyone in your area who still works with neon signs or lights, has some equipment laying around, at least for art projects, they may find a collaboration with you sounds quite fun! They're not likely to be very busy making neon signs these days.


Paschen curves obtained for Helium, Neon, Argon, Hydrogen and Nitrogen, using the expression for the breakdown voltage as a function of the parameters A, B that interpolate the first Townsend coefficient.

source (click for full size) Paschen curves obtained for Helium, Neon, Argon, Hydrogen and Nitrogen, using the expression for the breakdown voltage as a function of the parameters A, B that interpolate the first Townsend coefficient.

A Plasma ball photographed from above. The 1/60s exposure is needed to capture the plasma filaments rather than blurring them to ribbons

source A Plasma ball photographed from above. The 1/60s exposure is needed to capture the plasma filaments rather than blurring them to ribbons.

$\endgroup$
13
  • 1
    $\begingroup$ The graph mean that I can achieve plasma at 10^0 torr with lowest power right? $\endgroup$ Jan 18, 2023 at 2:35
  • $\begingroup$ @yuzhechong The units for the x axis are pressure $p$ (Torr) times gap distance $d$ (cm) meaning that for a gap of 1 cm, the pressure that has the lowest breakdown voltage is around 1 Torr. If the gap is 5 cm, then the pressure for lowest breakdown voltage would be 0.2 Torr or 200 mTorr. It's not an exact thing, you can see it's a fairly soft minimum in the breakdown voltage, at least on the high $pd$ side, and for realistic geometry that deviates from two smooth parallel planes; things with corners or roughness, you can often get breakdown at a lower voltage, or even unwanted arcing. $\endgroup$
    – uhoh
    Jan 18, 2023 at 5:49
  • 1
    $\begingroup$ A note regarding this: Typically, a real ion thruster would be generating the plasma in a region that is at higher pressure, and then accelerating it into an area that is much closer to zero pressure. $\endgroup$
    – ikrase
    Jan 18, 2023 at 8:23
  • $\begingroup$ @ikrase that's a very good point! For real, optimized thrusters that need to generate enough thrust to justify themselves. The don't run at the bottom of the Paschen curve, but they may require special materials to deal with the high power densities necessary to maintain such a plasma. For this school project it might be easier to just have the demo immersed in a uniform low pressure (air or otherwise) at whatever the lowest pressure they can get with a simple roughing pump, rather than trying to inject a controlled flow-rate into a specific location. $\endgroup$
    – uhoh
    Jan 18, 2023 at 10:34
  • 1
    $\begingroup$ @yuzhechong I think it's not about lab rental necessarily , but building a relationship with someone who can provide you access to a lab. Probably you would want, at minimum, a two-stage rotary vane vacuum pump. $\endgroup$
    – ikrase
    Jan 19, 2023 at 14:56
1
$\begingroup$

Another partial answer:

Most high vacuum pumps are very expensive. Except one: venturi aspirators. https://en.wikipedia.org/wiki/Vacuum_ejector.

The vacuum is limited by the vapor pressure of the fluid generating the venturi. In the case of cold tap water, this is in single digits Torr.

Vacuum aspirators are simple (no moving parts) and can be made by in a DYI machine shop.

It even looks something like a rocket engine !

enter image description here

$\endgroup$
6
  • 2
    $\begingroup$ First: By vacuum industry standards, venturi aspirators do not reach "high" or even "medium" vacuum. Second: The confluence of the HVAC industry and cheap Chinese-made tools have lead to the widespread availability of sub-$300 (pre-pandemic) rotary vane pumps (these meet medium vacuum). Unfortunately the supply of surplus small oil diffusion pumps (high vacuum, must be backed by a medium vacuum "roughing" pump) has mostly dried up. $\endgroup$
    – ikrase
    Jan 18, 2023 at 8:20
  • 2
    $\begingroup$ @ikrase I spent (too) many nights playing with a duoplasmatron-like ion source I was building in a chamber with an oil diffusion pump (with louvres for throttling) and a needle valve bleeding argon. I often got carried away, drove the filament and discharge current way high and lit up the whole chamber just for fun! $\endgroup$
    – uhoh
    Jan 18, 2023 at 10:39
  • 1
    $\begingroup$ @uhoh what filament you used as cathode? $\endgroup$ Jan 18, 2023 at 11:05
  • 1
    $\begingroup$ @yuzhechong filaments for electron emission have traditionally been made from thoriated tungsten (tungsten wire with a few percent thorium). The thorium lowers the work function so that at a given filament temperature the electron emission rate can be much higher. That means for a given emission rate, the temperature can be lower, so the lifetime of the filament can be much longer. $\endgroup$
    – uhoh
    Jan 18, 2023 at 12:06
  • 1
    $\begingroup$ @uhoh - at least it wasn't a mercury diffusion pump. Hopefully none of those left in use anymore... $\endgroup$
    – Jon Custer
    Jan 18, 2023 at 15:57

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.