Nowadays deep vacuum pumps are fairly expensive, so I'm wondering: how could people reach the deep vacuum required for crookes tube or Geissler tubes in the 19th century ? (I'd be interested in building one myself)

  • $\begingroup$ Torricelli managed to create a vacuum and so did Guericke with the Magdeburg spheres... Given the better tools we have now you should be able to do much better... $\endgroup$
    – Solar Mike
    Oct 1, 2018 at 10:21
  • 1
    $\begingroup$ You don't need to create a deep vacuum to replicate something like the Magdeburg spheres experiment. If makes no practical difference if the internal pressure in the spheres is as high as say 1psi - the resultant external pressure has only changed from 14.7 to 13.7 psi. Just boil some water in the sphere, wait till the steam has driven most of the air out, then seal up the sphere and let it cool down. $\endgroup$
    – alephzero
    Oct 1, 2018 at 11:17

1 Answer 1


Geissler invented the so-called "liquid piston" pump using mercury for the piston. The basic idea of the design is to entrap some of the air between two volumes of mercury in a pipe, and then force the mercury along the pipe to release the air into the atmosphere.

Geissler achieved 0.1 Torr with that design of pump in the 1850s. Sprengel invented an improved version of the same idea. The first models were capable of $10^{-2}$ Torr which was subsequently improved $10^{-5}$ Torr.

https://cds.cern.ch/record/455984/files/p281.pdf has more history, and diagrams (but unfortunately, not of very good quality).


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.