I have been playing with the idea of building my own linear power supply but have a couple questions.. The transformer I planned to use has no data beside output voltage. So is there a way to measure how much current the transformer can safely output ?? If built properly should I have the output grounded or not so I don't short out my scope because of a mix up ?? The plan was to use a basic LM317T setup but should I include adjustable current ?


Thanks
Jay

 

I built a PS from a transformer and a LM317 that allows both current control and voltage control with current limiting. It's more elaborate than you'd want, because I had a 4-20mA meter I wanted to use and I built that into it with a switch so I can look at either voltage or current. I did that mostly so I could hook it to a battery without overheating the transformer if the battery was discharged enough to draw a hefty current.

The only clue I had about the transformer was that over ~600mA, the thing got too hot to touch. I figured that couldn't be good for it, so I keep the max current below that. Once the current limiting cuts out, the voltage is set to 13.8V but I can dial in whatever I like.

 

Measure the secondary voltage with no load. Add increasing load until it drops 10%. Use that as the nominal secondary voltage.

The Nat Semi and TI datasheets show a current limited regulator. I don't care for it, but it's one way.

 

You can also look up transformers with a similar physical size and the same output voltage.
Their current rating would likely be similar to the one you have.

Note that you need to derate the AC RMS output current of a transformer by at least 60% to get the maximum DC output current from a full-wave bridge-rectifier capacitor filter on its output.

 

If built properly should I have the output grounded or not so I don't short out my scope because of a mix up ??

Normally the circuit is not connected to the main's safety ground but the case is.
Often there is a terminal connected to this case which can be connected to the circuit ground, if needed.

But that's unrelated to shorting out the scope, which is usually only a problem if you are trying to measure voltages on a line-operated circuit with no isolation transformer.

should I include adjustable current

Adjustable current limit is nice but it adds significant complexity to the circuit.
The LM317 has a built in current limit of about 1.5A and that's usually sufficient to prevent damage if the output is momentarily shorted accidentally.

 

should I include adjustable current ?

If I want to be sure I don't hurt my voltage supply when I am going to have decades to make a stupid move, I use an LM723. They come with a foldback option that is bulletproof. If I want a current limit, I build an adjustable constant current supply. If you want that to have a voltage limit, feed it from a voltage regulator.

Features, features, features!
You can't reasonably put all the features in one circuit and get deadly accuracy unless you spend all week to design it and use about a quad of op-amps. Or, you can build a module for each purpose and connect them to a single transformer-rectifier-filter capacitor rig.

Just a suggestion. If you want to be sure, but you aren't really sure, make modules.