Thanks, all for the input.

I decided to go with the LM2687 for the following reasons:

    * Used it before and I know it works well with analog circuitry.
    * Reasonably low part count.
    * Using a port driven charge pump will require output oscillation to
      sustain the negative voltage. This can be a hassle when debugging,
      as the oscillator may stop.
    * I wanted a small negative voltage and most solutions mirror the
      input voltage. This part let's you generate a regulated -1.2 volts.
    * Photovoltaic method too exotic for me :-)


Mark Skeels
Engineer
Competition Electronics, Inc.
TEL: 815-874-8001
FAX: 815-874-8181
www.competitionelectronics.com


Vitaliy wrote:
> Spehro Pefhany wrote:
>   
>>>> Sheesh.
>>>>
>>>> How can that work?
>>>>         
>>> The reverse B-E junction produces light, which illuminates the B-C
>>> junction which acts like a photocell.
>>>
>>> I've tried it on TO-18 case transistors with the metal case cut open,
>>> and you can see the light!
>>>       
>> Forgot to mention, you should probably toss the transistor after trying
>> this, because zener-ing the B-E junction tends to degrade beta,
>> more-or-less permanently.
>>     
>
> Fifteen or so years ago I saw a schematic of a "solar" radio which was 
> powered by a power transistor w/ the lid sawed off.
>
> This is one thing I love about the PICList. They don't teach this stuff at 
> school, and I didn't even think about the oscillator to -ve trick, or the 
> opto trick. :)
>
> Re: original question.. years ago Russell suggested splitting grounds using 
> a Schottky. You use the "Schottky" ground as your "normal" tround, and the 
> "normal" ground (now with a negative offset) for your opamp.
>
> Vitaliy 
>
>   
-- 
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist