Forums » General Electronics Discussion

I'm looking to build out a simple but clean 12v DC source for some audio components I have. I'm a novice when it comes to circuit design (breadboarding, arcade machine repair, etc.) so I'm looking for advice on how to best design this circuit.

In the end i would love to replace the wall wart SMPS with this linear circuit and build it out to support multiple outputs (5v, 9v, 12v, 16v). So I know I'll need to crank up the transformer to support these additional draws.

My questions to the community of experts: How does my first stab look to you?

I tried to setup the RC timings so the caps should discharge without impacting power while the unit is operational.

What are some improvements I can make to the overall design to ensure higher reliability, safety and quality of the power signal?

Thanks in advance! :)

@enusbaum,

You basic design is fine but you do not need R1 or R2. The regulator draws quiescent current which will drian C1 and C2 will discharge into the load.

With the basic 78xx series regulators, the voltage across C1 will never be high enough that you need any kind of bleed resistor for safety reasons.

If this is a general purpose supply then you might want to look at Fig 28 in:

http://www.st.com/st-web-ui/static/active/en/resource/technical/document/datasheet/CD00000444.pdf

There are better regulators available (lower quiescent current, higher ripple rejection, lower output noise etc., etc.) but the generic 78xx family and the slightly later LM340 series are fine.

There is nothing wrong with using Resistors on an Audio PSU design but for R1 & R2 I would go for a 10K Resister to suit Audio

https://www.circuitlab.com/circuit/n6n3a8/op-amp-v-reg/ This Circuit would be a even more stable design to use as a zenner is what is inside a V-Reg Chip & you have an Op-Amp Comparater to keep the output same as V-Ref ZD

What purpose would R1 and R2 serve?

R1 introduces an unnecessary additional current drain from C1 which will increase the ripple voltage across C1 and hence the input to the regulator. No regulator is perfect hence some of the input ripple will appear - albeit heavily attenuated - at the regulated output. Adding more input ripple adds more output ripple. Surely undesirable for audio applications.

Adding R2 just draws an additional DC load current from - and hence power dissipation in - the regulator.

The DC current drawn by R2 will also, in a linear regulator, be drawn as a constant current from C1. Adding yet more ripple across C1.

There are some linear regulators such as the LM317 family of adjustable regulators that require something like R2 to ensure a minimum output current is drawn from the regulator. This is because this current is actually the quiescent current required by the internals of the regulator to function within spec limits.

However, this current is normally either designed to be drawn by the output voltage adjustment resistor ladder or by ensuring that the load always draws sufficient current that the adjustment ladder current plus the minimum load current are always greater than the minimum current required by the regulator.

In general for such regulators it is undesirable to make the adjustment ladder resistors too high because there's also a temperature dependent bias current of a few tens of uA that goes through the ladder and so can cause a small DC offset in the output voltage:

http://www.st.com/st-web-ui/static/active/en/resource/technical/document/datasheet/CD00000455.pdf

The 78xx and LM340 families of regulators do not have such a minimum load current requirement:

http://www.st.com/st-web-ui/static/active/en/resource/technical/document/datasheet/CD00000444.pdf

http://www.ti.com/lit/ds/symlink/lm340-n.pdf