An Engineer in Wonderland – the glory of reed switches

I was pondering the failure of my car roof light to operate, wondering if newer cars had gone beyond the corrosion-prone two-pieces-of-metal-and-a-bit-of-plastic door switch.

When reed switches came to mind.

For the uninitiated, these consist of two flat flexible magnetic wires – reeds – held parallel and overlapping in a sealed glass tube.

A correctly applied magnetic field induces a N pole on one reed and an S on the other, at which point they bend towards each other and touch.

Operation has in-built hysteresis as, once attracted, the reads don’t unstick until the field has been reduces well below the engagement level.

With very little protection, they will operate for years in almost any climate, or immersed in almost any liquid, because the contacts are snuggled away in an inert atmosphere behind a hermetic seal.

Like Hall-effect switches, they are operated by an applied magnetic field.
And although they are not available with the sort of on-board signal-conditioning that makes Hall switches so versatile, they do operate entirely quiescent current.

Types are available to switch several hundred volts, and others will switch over an amp.

And with nothing to oxidise the contacts, any tiny current can be used to detect whether the switch is open or closed – although, oddly, I can’t find any spec sheets to tell me what the minimum operating current is.

Sadly, life is too short to manufacture a couple of reed switch assemblies to make my interior light work, but the failure did reminded me just how good reed switches are.

There us some more detailed actuation stuff here.


Respond below, or to



  1. Thanks for the story.
    I sort of hope that it isn’t true, because if it is, someone must have under-designed something somewhere.

  2. I worked for BT in the early 1980s and reed switches were used extensively in TXE4 exchange equipment.
    I heard the following story – I can’t verify whether or not it is an industry legend.
    Originally, the glass envelopes were filled with nitrogen, but experiments were done and it was concluded that dry air could be used instead, which would be cheaper.
    However, over time, and at the relatively high operating voltages of exchanges, some contacts would arc-weld closed.
    The card would then be removed from the frame, but the mechanical shaking it received would free the contact, so that when it was put on the tester, the tester would show no fault.
    Eventually, the problem was diagnosed, and the instruction added to operating procedures that in the event of such failures, the engineer should apply a sharp blow to the offending card to see if it cleared the fault.

Leave a Reply

Your email address will not be published. Required fields are marked *