I don't have an oscilloscope to see how bad the voltage regulation and noise output is. I presume for the price, pretty bad. BUT if the power supply is used to "convert" 120 VAC to 12VDC (as it does) and the 12VDC is then used to drive a linear step down voltage chip (such as a 7805, NCP1117, or similar). Would the poor regulation and/or noise be passed to the output of the second stage and the final voltage?
P.S. Yes this is the same power supply as listed in the diaphragm pump thread, but I'd like to discuss its use in a different application. The power supplies are pretty inexpensive as long as they are usable.
I woudl just add some capacitors, (less than the amount which will make the psu go unstabl) unless its a really sensitive app like a radio receiver receiving a weak satellite signal.
Ive had so many cases of micro's not working because of "psu noise", which ended up being faulty software, or floating input pins on the micro, etc.
In fact, dont even bother adding caps, as there will be enough output caps in the output of the psu.
Another point is that adding a linear reg to the output wont do anything to help common mode noise.
It was only an example, used to describe the setup to help everyone in the evaluation of the effects of how well a "secondary" voltage regulator may clean up the output of a power supply that isn't very pretty.
And I was only pointing out that if you think you're going draw an amp at 5 volts for example, you will be dissipating 7 watts in the regulator.
I'm posting this for others who may read this post and not be aware of this fact. From your question however, I would not assume you have an extensive knowledge of linear regultors - perhaps my mistake.
(a) perhaps a little rusty; but more so (b) never in this application, of trying to correct a poor power supply output. I have "cascaded" them to when connected to a good power supply, but not to (what may be) a poor supply.