Why are reed switches made in glass tubes?

Reed switches sense magnetic fields, and glass is not the only material not affected by a magnetic field. They can use plastic for example. Why are most of them in glass?

This question came into my mind when one of them easily cracked and then broke. It could have been in a plastic tube; then it wouldn't have broken.

Also, plastic would be cheaper than glass, and easier to melt and form.

  • \$\begingroup\$ Reed switches and relays come in a variety of packages, glass is traditional, but there are plastic reed switches. Glass is easier to manufacture in some regards \$\endgroup\$ – crasic Dec 16 '15 at 19:27
  • 5
    \$\begingroup\$ Reed relays are extremely sensitive to deformation. Plastic is easier to bend, basically destroying the functionality. Maybe gas tightness is an issue too, but I'm unsure of that. \$\endgroup\$ – jippie Dec 16 '15 at 19:28
  • \$\begingroup\$ For some application gas tightness is critical. Like in high voltage (100v) but low current when you want to maintain a vacuum. \$\endgroup\$ – Spoon Dec 16 '15 at 21:00
  • 4
    \$\begingroup\$ I would say that almost certainly, the one-word answer here is that plastic bends a lot and glass bends very little. In manufacturing, one of the few reasons to choose glass over plastic is "does not bend". This would appear to be a canonical case of that? \$\endgroup\$ – Fattie Dec 17 '15 at 2:54

Glass is clean, dimensionally stable and very strong, doesn't outgas at the operating pressures in the interior of the capsule, won't react with the fill gas in the capsule, and doesn't soften under soldering temperatures.

Here's a beautiful link.

  • 1
    \$\begingroup\$ Glass will outgass, not as much as plastic, but it readily adsorbs gasses during manufacturin, There is a reason all vacuum tubes and crt's have getters in them \$\endgroup\$ – crasic Dec 16 '15 at 19:50
  • 3
    \$\begingroup\$ @crasic: Nice point, but since garden variety reed switches don't operate under high vacuum, the gases which were adsorbed during the manufacture of the tubing won't be sucked out again and they'll stay mostly sequestered within the capsule's glass. Still, your point is valid and I'll have to fix my post. Thanks for that. :) \$\endgroup\$ – EM Fields Dec 16 '15 at 20:36
  • \$\begingroup\$ And what a beautiful link it is! There goes my evening! \$\endgroup\$ – bigjosh Dec 17 '15 at 1:20
  • \$\begingroup\$ @crasic if glass outgassed then the getter would be pointless. Virtually all materials adsorb water vapor, but this desorbing is not considered to be outgassing. In vacuum tubes, the getter is used to get from high vacuum to ultra-high vacuum, and to maintain the latter over a long period of time. Outgassing materials are utterly incompatible with this goal. \$\endgroup\$ – Oleksandr R. Dec 17 '15 at 1:48
  • \$\begingroup\$ @OleksandrR. the only disagreement we have is over the use of desorbtion v outgassing. \$\endgroup\$ – crasic Dec 17 '15 at 2:11

It's likely due to the strength vs thinkness needed for the reed switch. A cylinder of plastic of sufficient strength would be too thick for common magnets to switch. Plastic of the same thickness of the glass would bend without much force at all. While the glass may snap, the plastic would be worse. Keep in mind, the outside is only meant to protect the easily bent metal arms. Once they are bent out of shape there is no easy fix.


The reeds have to be held very strongly in relation to each other, because they are usually spaced very closely together and any bending of the package, usually due to assembly and soldering stresses in the normal manufacturing process and environmental changes, would damage the ability of the reed to work within it specifications.

Glass is a nice brittle, strong enclosure, and further you can find metals that are magnetic and have a similar temperature expansion rate so specifications can be maintained across the specified temperature range.

Plastics and many other materials could be used, but they wouldn't be as reliable without enlarging them (more material to make up for the lack of strength), and dealing with temperature expansion mismatches.


With glass you can see inside. That's useful since reed switches have orientations. With a opaque package, the orientation would have to be marked, but there isn't much room on something the size of a reed switch. That would leave cryptic markings. With glass you just don't need that.

Glass is also more hermetic and inert than other materials of the same thickness and strength that are candidates for reed switch casings.

  • 8
    \$\begingroup\$ Sorry but I don't buy that marking argument, at least not for modern components, we write quite some things on our smd components these days, also a flat side on an otherwise round component won't be that bad either. \$\endgroup\$ – PlasmaHH Dec 16 '15 at 20:44
  • \$\begingroup\$ The only reed switches I know of which require orientation are mercury-wetted, and I believe almost all - if not all - of them are used to make mercury-wetted reed relays, with the proper orientation printed on the relay's cover. \$\endgroup\$ – EM Fields Dec 17 '15 at 14:28
  • \$\begingroup\$ The other advantage of the clear package is that you can see if the switch is activating when it's close to your magnet. With an opaque package, you have to use a multimeter to check it. \$\endgroup\$ – Johnny Dec 17 '15 at 16:22
  • \$\begingroup\$ Orientation matters? For other than some mercury wetted switches, that's utter nonsense. The only thing that matters to a reed switch is if the field around it is strong enough to attract the contacts toward each other in order to reduce the reluctance of the magnetic circuit. \$\endgroup\$ – EM Fields May 12 '16 at 0:50
  • \$\begingroup\$ @EMFi: You're not going to get the same results with the magnetic field perpendicular to the motion as when at right angles to it. \$\endgroup\$ – Olin Lathrop May 12 '16 at 10:44

I imagine you're asking the question "why are reed switches still made of glass?"

Glass is traditionally used. Reed switches find rarer use every day since there are Hall-effect switches and other devices that are smaller, cheaper, and more reliable. Reed switches were invented in 1936, according to Wikipedia.

Nobody retools a buggy whip factory, just as nobody will retool a reed switch factory. If all reed switch manufacturing disappeared tomorrow, it's unlikely a new site would appear to replace the lost capacity.

All of the other mentioned reasons (imperviousness to gas, rigidity, and so forth) can be addressed using plastics. Microchips are currently manufactured in plastics when at one time they came in steel packinging (TO-packaged LM711 comes to mind) as well as ceramic (old intel 4004's, even up to the 8086), but the usefulness of these items and the expense of that technology drove retooling to plastics.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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