Built my first LM317 breadboard voltage regulator unit today.

[Evil Lurker]

Yeah I know, they are pretty simple little circuits. But this time I used a hunk of scrap breadboard, did the component layout myself, and added in a few twists to better suit my needs.

The first twist is that I used dual right angle connectors to plug the PCB directly into each side of my breadboards power buses. So I now have access to the original input voltage and the regulated voltage output. Plus they also make the PCB very steady. I also went ahead and added in a jumper block should I ever wish to disable the regulator and while I was at it added reverse protection diodes. Due to lack of space all I could fit on it were two 50v 100uf elctrolytics on the input and output, I also went ahead and put a .1uf mono cap right next to the input pin. I'm using a 510ohm resistor and 10k pot for voltage regulation. Its pretty sensitive and will allow me to go "all the way down" if I so desire.

I figure I will try and "reform" some new old stock caps and see if I can detect any changes in their ESR levels. I'm kinda sorta thinking that slowly raising the output voltage may not be the best way about waking them up from a deep slumber... I'm kinda leaning towards hooking them up at full rated power and limiting the current that they will be fed via resistors.

[Agent24]

Re: Built my first LM317 breadboard voltage regulator unit today.

Reforming only fixes (current) leakage problems, not ESR, I think?

[999999999]

Re: Built my first LM317 breadboard voltage regulator unit today.

In addition to what Agent24 wrote, slowly raising the voltage is the best way. Reforming to rebuild the oxide layer increases insulation to allow higher voltage operation without excessive leakage. The point of reforming is to not subject them to too high a voltage right away.

However, whether you really need to do it at all depends on how old they are. I've grabbed 10+ year old dormant caps and put them in circuits and they did fine w/o reforming but I usually used caps with a notch or two higher voltage rating than I really needed which is essentially what you are probably doing since most DC powered electronics run off significantly less than 50V.

100uF 'lytics on the regulator board is fine, assuming you have sufficient capacitance upstream at the power source to reduce ripple if/when supplying enough current to need it. In that case even a 10uF tantalum will work.