Dual supply ICs that aren't microscopic?

[paulca]

Hmm.   I'm not sure I need 100 or 200mA.  My headphone amp which has 2 dual op amps and a power LED shows as using less than 20mA even at full tilt.  The rest of the opamps in my project will not be producing power for speakers/headphones, but just buffering and pre-amping.

Maybe I can get away with a smaller output.  It's hard to know.  What is the standard way of assessing this?  Do I add up the "absolute maximum current" for all the ICs or do I take a test such as the example headphone amp and multiple it by the extra amps?

I do like the idea of the contained isolated (or even unisolated) potted PSUs linked to.  Saves me the hassle and saves me the potential of ballsing up the PSU stage and getting horrid artifacts from a clipping switching regulator or similar.  Also the switching regulator, self built will probably not be prototypable on a breadboard.  I learnt that lesson recently, they don't like it, too much capacitance in the BB.

An off board PSU to limit EMI is an option, but would mean more things flitting around my desk.

Now I'm in procrastination mode.  The potted PSU units are not cheap enough to buy to test and then reject.  Where as the switching self build or switching IC build is a fair project in itself.

[Kalvin]

Is it going to be portable? If not, you should consider the plain old linear power supply as your power consumption is very small.

For the portable device you probably want to use a switched mode psu - either charge pump or inductive - with good filtering. Filtering should be quite easy to do as your power requirement is quite modest and you do not require to have large magnetics.

There are some mini 12V re-chargeable batteries available if you want to experiment a little. This would be a compromise between a linear psu and portability, but you would not have to struggle with the noise problems. However, you need to implement some sort of charging capability which will add some complexity. Just my 2c.

[paulca]

Is it going to be portable? If not, you should consider the plain old linear power supply as your power consumption is very small.

I could go linear if I could get the operating voltage down to +-9V, as long as I can find LDO linear and inverting linear regulators.

It's not portable persae but I want the option to run it off 12V nominal battery power.  So that results in the standard automotive power range, 9V-15V.  It just makes it easy to power from anything including a basic 12V wall wart, Lithium 3 series battery or a feed from a solar panel setup.

The headphone amp I have which will form the basis of my amplifier design will run on such voltages and I don't notice any degradation of quality until it gets down to around 7 or 8V, but I haven't characterised that in anyway with a scope, just turning the voltage down until I noticed a sound quality change.

[tszaboo]

The isolated converters are just a PWM signal fed into a transformer. The transformer is driven to saturation (EMI everywhere). There is absolutely no regulation.

I'm certain that cheap converter modules exist that are built as you say.

But I think you're underestimating  the modules I suggested. They are indeed regulated, check the data sheet.
Like I said, I've used them myself to power opamps at +/-12 V and have had absolutely no noise problems or EMI. The high frequency ripple is killed in the end by the decoupling caps at the opamps.

But you go to your church, I'll go to mine.

It said on their website it is fully regulated... Probably only on the + for the dual version, that is why there is a 5%  variation int the output. And it works at 0% load. OK. Looks like these are newer models, with more intelligence built in.

[paulca]

I'm thinking I might try one of those.  It will make things a lot simpler and allow me to focus on the op amps which will be taxing enough.

Still need to decide on how much power I actually need.

The Maximum input current on the NE5532P is -10/10mA.  I figure I need at most 8 of them.  3 input preamps, 1 mix down buffer, 2 aux output buffers and 2 as a current sharing pair for headphone output.  They have current limiting outputs so I should be able to stay within the 10mA and thus my supply requirements would be 80mA.

Adding a 25% grace gives me 100mA.  The 4mA or so for a power LED is negligible.  Does that sound reasonable?  Do I need to accommodate loss in the passives?

[David Hess]

Another way if regulation is not required is to use a 1:1:1 or N:1:1 transformer and drive it with a simple inverter.  Some switching regulators can be configured to drive a transformer as part of an inverter.  An inverter can also be regulated from either the input or output side to provide a constant output voltage.  One nice thing about this is that the outputs may be completely floating.  Coilcraft makes some off the shelf transformers for this.