Working on a 24v unregulated power supply, I dug out a transformer from a broken solder station. The transformer is 115 60hz on the primary and 0-12-26 on the secondaries.

Using this formula, C = 0.7(I)/ΔE(f) I worked out a capacitor value of ~2800uF for the capacitor. Digging in my junk box I find a 3300uF cap. I connected to the output of a rectifier and with a 2 amp load I get ~25 volts and a ripple to low (or to fast) for my DMM to measure. Seems like this will work Ok. Only one problem, the CAP is rated 40volts and the rectifier output is ~39 volts with no load - to close for comfort in my book.

After this I started to "play". Using a 470uF 100v Cap I measure ~25v with a 2 amp load, and ~12volt ripple (DMM set to measure AC, so I assume this to be the ripple voltage). I then added a second 470uF Cap in parallel to the first and Then measure a ~6v ripple, a 3rd 470uF reduced it to ~4 volts, ect. This is all to be expected.

Anyways, after all the playing around I noticed that the output without a Cap measured ~24v DC and ~23 volts AC, no load (again I assume this is almost a 100% ripple) with a Cap I would measure ~39 volts and very little ripple at no load.

Now my questions:
Is the reason the voltage increases when the cap is attached due to the ripple adding to the DC charge?

I plan to use part of the 24volt and connect it to a LM317 to get a variable voltage, how much ripple is too much for the input of the LM317?

btw, this PS will be used to test non-critical equipment so precision and regulation isn't necessary.

 

"I plan to use part of the 24volt and connect it to a LM317 to get a variable voltage, how much ripple is too much for the input of the LM317?"

If the negative peak voltage value is at or below the drop out voltage specification of the regulator when drawing maximum rated current at maximum desired adjustable voltage then it's too much ripple voltage.

Lefty

 

"Is the reason the voltage increases when the cap is attached due to the ripple adding to the DC charge?"

Either that sentence lost something in the translation, or... No.

Without a cap, your meter is reading either the average or the RMS of the pulsating DC voltage from the rectifier. With the cap, the voltage is charging up to the peaks of the pulses, and maintaining that voltage (or nearly so) until the next peak. The average of this is much larger than the average of the unfiltered pulses.

 

"Is the reason the voltage increases when the cap is attached due to the ripple adding to the DC charge?"

Either that sentence lost something in the translation, or... No.

Without a cap, your meter is reading either the average or the RMS of the pulsating DC voltage from the rectifier. With the cap, the voltage is charging up to the peaks of the pulses, and maintaining that voltage (or nearly so) until the next peak. The average of this is much larger than the average of the unfiltered pulses.

Ah yes, forgot about the Peak to Peak, thanks for knocking me in the head.
<I miss my scope>

 

There is also loading on the transformer, which can be significant. Think internal resistance.

 

You didn't say if you were rectifying with a single rectifier (half wave) or a bridge (full wave). If the former is true you can replace the rectifier with a bridge. This will enable you to half your filter cap (uF) value.

As far as the voltage being to close to the max rating of the cap you specified... definitely!

In regard to your DMM's capability to measure ripple at 60Hz is concerned ... yes it can.

Edit: When 60Hz is full wave rectified the output of the bridge is 120Hz. The higher the source freq is the smaller filter caps have to be.

 

There is also loading on the transformer, which can be significant. Think internal resistance.

True, but he said that he tested it with a load of 2A already. So that should not be a factor here.

 

You didn't say if you were rectifying with a single rectifier (half wave) or a bridge (full wave). If the former is true you can replace the rectifier with a bridge. This will enable you to half your filter cap (uF) value.

It is a 4amp bridge.

As far as the voltage being to close to the max rating of the cap you specified... definitely!

I will need to keep digging on my junk box. I am trying to do this without buying anything for it. This is just one part of the whole project and I am useing all scavanged parts. Besides these caps are exspensive. I could almost buy a regulated 24v supply for the price of the cap alone!

So I may go with 4, 470uF 100v caps in || that I have on hand.

 

If you're a young fellow on a minimum budget I would recommend the art of dumpster diving. They can be a gold mine. You just need to know what dumpsters before doing a triple lindy in it.

 

If you're a young fellow on a minimum budget I would recommend the art of dumpster diving. They can be a gold mine. You just need to know what dumpsters before doing a triple lindy in it.

well I'm not really a young fellow, I guess I could throw my son into the dumpster, My business provides a good source of all kinds of parts and strange oddities. I'm not really on a minimum budget - just a personal challenge to build a complete project out of Scavenged parts. I'll post pictures of it when it is complete.

 

I'm not really on a minimum budget - just a personal challenge to build a complete project out of Scavenged parts.

Hey, 'field expedience'. I can relate to that!