Is there any reason why the circuit below is a bad idea?
I want to power both an Arduino and a split supply OpAmp cct from a single switch mode 15V o/p wall wart psu.
I was planning to use this arrangement (capacitors omitted from drawing), where the Arduino would be the top load and I would take the +5V regulated output from the Arduino as one half of the split supply for the op amp and the negative -5V supply from the negative input of the wall wart, the other half. The GND would be taken from the output of the 5V regulator for both opamp and Arduino.
Would the voltage regulator work with a significant load here or am I being really dumb? Bye the way the +10V is probably 100mA max and the -5V at most 1mA.
Any thoughts on a better (but simple) way to do this would be much appreciated.
Wouldn't this mean the 7805 becomes a current sink? The data sheet specifically says sinking current through the GND connection will damage the device.
I want to power both an Arduino and a split supply OpAmp cct from a single switch mode 15V o/p wall wart psu.
Can't see how you are going to do this with a single regulator.
You could use a 5V and 10V regulator then your +5V becomes your ground.
Alternately create a common signal for the op amps with a potential divider, and leave the arduino at ground.
I am beginning to think that this arrangement should work.
I should have said, the requirement is for a split supply for the op amp where the GND of the op amp and the GND of the Arduino are at the same level. But, all powered from a single wall-wart switch mode PSU. I don't want to use a transformer.
The output from the Op Amp will eventually go to an analog input, but only after rectification, so there is no possibility of the analog input going negative.
Mike - care to expand on
That wouldn't work
. Notionally I always though of linear regs as little more than this kind of zener arrangement and it would reduce my component count.
. Notionally I always though of linear regs as little more than this kind of zener arrangement and it would reduce my component count.
But they are not, they are essentially a pass transistor and amplifier. In that previous circuit you had the output connected to the input through a resistor.
This one is better and stands a chance of working, although I can't say I like it. Have you worked out the power dissipation in the components?
If it were me I would use a voltage mirror IC to get you the -ve rail.
Thanks Mike. The collector current of the transistor will be the current through the Arduino, which fortunately is quite small in this application, so power should be alright.
It is a bit dirty - I haven't used a DC to DC mirror chip - any recommendations?
Presumably this gives a much higher resolution than home-photo etching? I use transparency film and a home made UV LED light box, but haven't tried anything as fine pitched as those pesky little SMD chips.
Also, no nasty chemicals with milling.
So do you stick to single-sided, or flip the board over mill both sides?
I have only used single sided so far but I could do double. I think you could actually get finer with home photo etch than my miller. You are restricted by the tool size and how flat you can get the Z axis. I have some compensation software I am working on at the moment.