By William J. Garner
While working for a firm doing remote rural telephony by transmitting and receiving TDMA modulated signals at around 450 MHz in bands separated by about 100 KHz, I was asked to travel to a California manufacturing plant to determine the test procedures for a power amplifier that incorporated feed-forward techniques to reduce the third order inter-modulation distortion so that the re-growth of the modulation sidebands did not extend over to the adjacent channel. The power amplifier was in a rack along with a receiver and its power was supplied by a -48 DC battery then converted using a DC to DC converter to supply the power amplifier with a positive 35 DC volts.
When I got to the lab to do the testing, I noticed the amplifier was on, but the panel switch was in the down position that said “OFF.” I went ahead with the testing to affirm that the power amplifier met the third order inter-modulation specs and decided to see why the panel switch was always on regardless of the up or down position. One of the design engineers in the lab said that they had all failed closed in the field, but since it wasn’t a main concern since they were always on during field operation located in remote areas such as mountain tops where the antennas could give the best distance range.
I removed the screws holding the unit in the rack and slid the unit from the rack. I removed the 48 volts plug from the rear of the unit. Then I removed the two screws attaching the switch to the panel and unfastened the two wires from the two switch screw terminals. I noticed that this switch was one used in everyone’s home that had a rating of 120 VAC at 10 amps. I wanted to see the condition of the contacts so I had to drill out the two rivets holding the plastic body from the switch mechanism and used a stereo microscope to view the two contacts which were fused together and reminded me of an electric weld.
Thinking back to my college days in the electrical section of a physics course, I remembered the problem of two electrodes being charged with opposite DV voltages and if the electrodes were brought close enough, an arc would occur. This was dependant upon the level of the charge, the size and shape of the electrodes and the atmospheric gas between them.
Now a switch handling AC would not have a problem since the AC voltage is cycling up and down so the chance of any arcing is small because the differential voltage is always changing. With this reasoning, I decided to call a local electrical supply house to see if there are switches made for DV switching. Sure enough, they had one for handling 200v DC with an amp rating of 5 amps. I was concerned about the DC switch being too big, but they gave me the measurements and found that it had slightly deeper dimension, but everything was the same as the AC switch. They hand delivered the switch and it just fit in the panel location. I noticed that when toggling the switch off to on, the action was snapped very fast that means the spring action was very strong. Connecting the 48 volts wires and the -48 volt plug and putting the unit into the rack, we turned the unit on then off with no sound of arcing occurring.
All the other units in the field and production line installed the DC switch with no “Closed” failures occurring.