[TR] I give up

[jimmuller at rcn.com]

David  Lylis wrote:
> With nothing left to lose, I took the gauge apart and
> sprayed the inside with carb cleaner and blew it dry.
> It is working again.

An ammeter is a strange sort of beast.  In fact, there is only one kind of meter, a galvanometer, and it responds to current.  It is usually built to have as low a resistance as possible, and to expect very low current.  A voltmeter is made by putting a known high resistance in series with the galvanometer, and calibrating the markings such that a given voltage produces some particular small current that produces a given deflection on the meter.  End of story, more or less.  The resistance is chosen to be high so that very little current is carried, thus not burning out the meter and also so that it draws no significant extra current on the circuit it is measuring.

An ammeter is a step more complicated.  It must carry all the current of the circuit being measured, but it must do so without adding significant resistance so as not to cause a voltage drop.  But it can't be just a pure conductor because that would mean there is nothing for a galvanometer to detect.  So it needs a voltage drop, but a very tiny one.  To that end, it places a known, very low but non-zero resistance wire in series with the circuit.  All current going into one side of the meter must pass through that wire to get out the other side.  The resistance of that wire produces a small voltage drop from one end of the wire to the other.  A voltmeter, i.e. a galvanometer in series with a large resistor, is placed in parallel with that low-resistance wire so as to measuree that voltage drop.

The point of all this description is that an ammeter could fail in two ways.  If one of the terminals was broken or making a poor connection, or if the low-resistance wire was broken, it would look like an open circuit and pass no current at all.  With this failure mode nothing in that part of the wiring would work at all because no current would get through.

The other failure mode is that it carries current just fine but doesn't seem to measure anything.  This could be because something, perhaps water, is providing a short between the main terminals.  It wouldn't do much for the overall conductivity of the meter (which must be as high as possible, but still have some known small resistance), but it might be enough to make the internal galvanometer have no voltage drop to detect.  However the same failure mode could be caused by a broken wire within the voltmeter section, i.e. within the galvanometer or its associated series reistor.  Or instead of a broken wire it could just be a poor connection in the voltmeter section.  Just the act of taking it apart and re-assembling it could fix a problem like this.
--
Jim Muller