Induction Coil

 

Induction Coil I do not know what this was used for, it came as an unexpected part of an auction lot. It appears to be part of some larger apparatus since only the front and top (as shown here) of the unit are polished. The ebonite panel is stamped "REG.No.381 No 7988". Adjacent to the C terminal is an ellipse enclosing what looks like 8TH. The windings are encased in a solid block of wax and are continuous. I am now pretty sure that it is part of an early spark transmitter. This site gave me the idea:trench transmitter

Since then I have seen a picture of a transmitting receiving set sold by Gamages of Holborn prior to 1914 which included an almost identical adjustment lever and contact arrangement. This was illustrated and described in the BVWS Bulletin (vol.25 no.1). A similar item is shown in the centre of a photograph of a pre World War 1 radio station on page 237 of "Radio Radio".



Peter Norris who viewed this page has provided some interesting information about the automobile buzz coil. He has a Model T ford coil and provided this link They were housed in a modular box 130 x 80 x 50mm, not including the contact breaker mechanism on the end and the connections were to 3 button shaped terminals, one on the end and two on one long side. This enabled the coils to be slid into a coil box. It was common to have one coil per cylinder as this avoided having to distribute the HT. The engine was instead fitted with a low voltage 'distributor' which cycled the LT current to the appropriate coil at the right time. The armature was very light so the frequency of operation was high - hence the term buzz. The spark on mine can be drawn out to over inch so they are pretty powerful devices and some care is sensible.

He suggests that "P+ is battery positive, S is spark plug and C is the LT contactor (battery negative). You could try 6V between P+ and C and expect up to 4 amps to flow with activity at the contact breaker. Try that momentarily and if no vibration, disconnect, open the points adjuster, reconnect and then bring the points into contact which should start the armature vibrating. Throughout this you should arrange a spark gap between S and C using a short length of stiff wire - gap about 1mm - to unload the secondary winding. Not much to go wrong with them apart from points burnt and eroded away, condenser failure which would exacerbate points burning or an open circuit in the secondary. An avo meter check should give a good indication of health but you could go ahead and try it with maybe an car headlamp bulb in series in case there is a short present."