Voltage regulators and capacitors

Yes, they are really needed.

When three terminal regulators were knew, it was easy to dismiss the capacitors, "why do we need them, it's just a voltage regulator?". But then they were really great oscillators without the proper bypassing. Everyone soon learned the necessity of those capacitors.

If you have the regulator right next to the large filter capacitor of a power supply, the input capacitor may not be needed, since the filter capacitor might do the job of keeping the regulator from oscillating. Or it might not, since the filter capacitor is there to filter the 60Hz, and is so large in value that the accumulated inductance completely negates its value at higher frequency where it would be need to stop oscillation.

The old National databook would always list three values for that input capacitor, depending on the type of capacitor used. The actual value was less important than the capacitor's effectiveness at bypassing the higher frequencies.

The 0.1uF on the output would be a ceramic capacitor.

I don't seem to recall having problems with a .1uF ceramic capacitor on the input of a 3 terminal regulator, but I'm too lazy to dig out the National databook to find out exactly what they said.

Michael

It's certainly easy and cheap enough to just put a small bypass cap of 0.1 to 1 uF at the input close to the pins. It's also a good idea to have some sort of output capacitor to handle brief current surges of the load. But if the output capacitor is much larger than the input capacitor, and especially if you disconnect the regulator from the main filter capacitor, it's a good idea to put a diode from output to input to avoid back biasing the regulator, where the output voltage is higher than the input (by more than a diode drop).

Paul

0.33uF capacitors seem to be rare as rocking horse guano, but 0.47uF are available. OK to use 0.47uF?? Also are metallised polyester capacitors acceptable or should I use electrolytic??

TVMIA