[sdiy] 5V regulators

[sasami at blaze.net.au]

I've been down this track before myself. I run my VCDO 5V 
sections from a separate 5V linear suppply capable of around 
1 amp as limited by the transformer. The regulators 
themselves are well heatsinked - it's an old (1970) computer 
power supply. Experiments with 7805s running off any of 
the synth's higher voltage rails were not that satisfactory. 
While it did work, it loaded down the 15V rail, and 
generated a lot of heat (no news here!). It proved cheaper, 
and more efficient to use a separate supply, with the added 
advantage that no digital hash was making its way back into 
the 15V rail. Why do I say cheaper when I obviously had to 
add another supply? Because if I didn't add a separate 5V 
one, I would have soon found myself needing to upgrage or 
add another 15V one to supply my other modules.  

As for needing 500ma, that is the specified current 
consumption of older 2716 chips. If you are using a modern 
or CMOS variant, the current may be less. 

Ken

>In-Reply-To: <memo.770269 at cix.compulink.co.uk>
>Hi all, thanks for the advise, it is much appreciated.
>I got the power rails wrong to start with I meant +/-15V
>when I stated 12V.
>
>In the cold light of dawn I am thinking about the 500mA
>needed, surely this cannot be correct? I took this straight 
out
>of the VCDO construction notes:
>
>"Power supply requirements to the VCDO are +/-15V at approx
>40mA per rail and a separate +5V rail at 500mA"
>
>I have to confess that the first VCDO I built was tested in 
my +/-12V
>cabinet and the +5V rail was derived using a 7805 for the 
whole cabinet
>and I had no ill effects or hear dissipation problems.
>
>I think a separate +5V supply is best though. I have a 
spare one somewhere.
>
>thanks again
>
>regards.....
>
>Keith
>
>http://home.freeuk.net/lowpass