Generating -5VDC from +5VDC 
(From Richard Friesen)
If you happen to have the March 1984 issue of Radio-Electronics, 
turn to page 78. This issue has the very first instalment of 
Robert Grossblatt's "Designer's Notebook" column. In it, he shows 
a simple circuit which will supply a negative voltage, given a 
positive voltage. It's basically a 555-based oscillator, and a 
voltage-doubling rectifier. He claims the negative-voltage output 
should be good for about 60ma. No-load voltage should be pretty 
close to the input voltage (but negative), although the voltage 
will drop a bit, depending on the load. If you put +5V into the 
circuit, it'll give you around -5V out. load. If you put +5V into 
the circuit, it'll give you around -5V out. If the load makes the 
voltage drop too low (-3V or -4V), you could always just feed the 
circuit with a higher voltage (like maybe 9V or 12V) and then just 
regulate the output down to -5V using a 7905 regulator. I've used 
this circuit a couple of times for powering op-amp's, and it works 
great!

I'm not that great at ASCII-art, but I'll give it a shot. If the 
following schematic doesn't make sense, let me know, and I'll try 
it again...


          +V
           ^
           |
   +-------+---+
   |       |   |                  -V Output
  R1       |8  |4        +----+--->    
   |    7 -------        |    |            Parts List:
   +-----|       |      D2    |            IC1 = 555
   |  +--|       |  +    |    |             R1 = 1.5K
  R2  | 6|  IC1  |---C1--+    |             R2 = 10K
   |  |  |       |3      |    |             C1 = 10uF,16V
   +--+--|       |      D1   C2             C2 = 22uF,16V
   |    2|       |       |    |+            C3 = 1500pF
   |      -------        |    |          D1,D2 = 1N4001 diodes
  C3         |1          |    |
   |         |           |    |
   +---------+-----------+----+
             |
            ===
           (GND)


Note: In the above "diagram", both diodes point down (the anodes 
are at the top). Also, watch the polarity of C1 & C3.
The circuit is set up to oscillate at about 45kHz, with a duty 
cycle pretty close to 50%. None of the values of any of the parts 
are terribly critical, so if the capacitors or resistors are "in 
the ballpark", it should still work okay.