[sdiy] LED's on negative rail??

[media at mail1.nai.net]

>> I'm working on a few things that run almost entirely off the postive
>> rail. Since LED's draw a significant amount of current (the light
>> has to come from somewhere), I'm thinking that I should run the LED's
>> off the negative rail to balance things out.  However, I've never seen
>> that done.  Is there any reason why??  Is it a noise issue??
>> As far as I can tell, the LED's would "see" the same difference.
>
>Hmmm, nice idea...
>I guess you mean getting whatever it is to switch to ground?
>so you have
>
>-V ----|<---- siganl

I wasn't thinking of switching it to ground, I was thinking I could pinch
off the current with a PNP transistor.  As it turns out, I was wrong.  To
use a single transistor on the negative rail the base would have to go
negative, but it could be done with two transistors (or a negative going
input signal).  I don't know anything about discrete FET's, except that
there are six types that I don't understand :)

>The problem is you need to make sure that the signal NEVER goes positive,
>>it must be either 0v or open circuit.. easy to do with a CPU, not so easy
>>with TTl/CMOS/Opamp.

Wouldn't ground be "positive" as far as the LED is concerned??

>If you're concerned about noise issues, one solution is to drive the LED
>through one side of a differential transistor pair so that current drawn
>from the PSU remains fairly constant.

I'm not sure what you mean.

>This is the technique I used on all the Wiard modules. All the voltmeters,
>VU meters and the like are LM339s quad comparators powered from +/- 15
>volts. The outputs are open collector and you can run the LEDs between
>>noisy (digital) ground and -15 volts (use two ground systems in any
>analog >design, one for a quiet voltage reference, another for noisy
>current >demands).
>
>Then you run a trace from the -15 volt pin on the LM339 to the incoming -15
>volt terminal BEFORE your power supply input filters or regulators. The
>noise generated on the -15 volt line is now filtered out with the rest of
>the rail noise, and doesn't get inside your circuit. Relatively little
>current noise appears on the LM339 positive supply pin.

That sounds like a great idea!!!  Let me see if I understand this
correctly.  The output of the comparator is an open collector with a
grounded emitter, so you use a resistor to pull down the output to the
negative rail, and it goes up to ground when the comparator swings on??
Except for making repairs, I've never used a comparator chip -- I've always
used op-amps without the feedback.