Lectures have finished; students now are into the exam period; and my thoughts naturally turn to research for the summer. To be more accurate, they first turn to marking the aforementioned exams and other assignments, but research will quickly take over.
One of the projects we have going involves recording small electrical signals from a system in our lab. Small signals for us mean a few millivolts, with very little current. Probably the major problem we face is electrical noise. That means that the signals we want are contiminated by those from other sources – just like it’s hard to hear a bird singing in the garden if you live next to a busy road – the signal from the bird is contaminated by the road noise.
Some sources of electrical noise are unavoidable. Thermal noise (or Johnson noise) is unavoidable, since it comes from the thermal energy of electrons in the system. There are ways of reducing it, such as cooling your apparatus with liquid nitrogen. Shot noise is a different effect and is arises when there are very low currents. Since each electron carries a discrete charge (small, but not infinitessimal), charge doesn’t flow smoothly – and that gives you a fluctuating current.
But the problem we have is contamination from other electrical sources. There is so much electrical equipment around the place that the room is full of alternating electric and magnetic fields. These fields will induce currents to flow in the equipment – e.g. through Faraday’s law – a changing magnetic flux through a loop will induce a voltage around the loop. Also, any piece of wire can act as an aerial and electrons in it respond to alternating electric fields.
There are simple ways of reducing this effect – e.g. by having your cables nicely shielded but it’s not perfect. Strategically-placed aluminium foil can be a great help sometimes – it cuts out electromagnetic waves because it is so conductive, as you can demonstrate with a mobile phone. Clever routing of the wiring to avoid large-area loops (i.e. reduce the extent to which Faraday’s law can give you voltages you don’t want) is good practice too.
However, our lab sited next door to the switch room for the whole building (where the mains power comes in to supply most of the Faculty of Science and Engineering) which is hardly ideal, and so we might just have to put up with the problem for a while.