Guys, do you think this answer about this lm317 is okay?

Someone asked me a question about this LM317 circuit. .

The picture is the schematic diagram of the LM317 adjustable regulated DC power supply circuit. The design uses a single-phase bridge rectifier circuit. In response to this design, someone asked me,

"If you use a rectifier bridge circuit, will the capacitor not short-circuit the live line and the neutral line?"


I want to answer this way: The basic characteristic of a capacitor is to separate the direct current. In the specific analysis, it depends on the frequency of the communication and the capacity of the capacitor. The capacitor I used in this circuit is 104, which is 0.1UF, and the capacity is very small. And the mains will not be short-circuited through the capacitor.

However, if there is a high-frequency interference signal, it will be filtered out by a capacitor short circuit, so this capacitor is also called a high-frequency filter capacitor.

Guys, do you think this answer is okay?

RachelAnne289 said:

and the capacity is very small. And the mains will not be short-circuited through the capacitor.

Guys, do you think this answer is okay?

Click to expand...

The mains will be shorted through the capacitor but only for a short while. Because the capacitance is small, it will charge up quickly and follow the mains waveform.

The reactance at the line frequency is approx 32k Ohms. So the capacitor will carry a current of about 8mA in the steady state. But during the initial couple of mS during the power on phase, the capacitor will appear to be a short but the short circuit current will depend on the actual phase of the line signal during start on.

So your answer is ok. For a noise signal having a frequency of 50kHz, the reactance will be 1000 times lower and the capacitor will act a good short circuit.

Ideally though you should use a smaller capacitor (say 105) in parallel with each of the four diodes. Diodes have their own capacitance but adding a little more is good for the health.

I think it can be okay, it doesn't short mains. But be cautious, the capacitor and transformer inductance formed LC-tank resonance should be eliminated. 100nF is not so small, resonance can cause high frequency noise current peaking which is radiated toward the circuit by wiring and conducted via power lines to the source. So I am not sure it will filter well the high frequency noises, but I don't have too much practical experience with it.

Hi,

Generally I agree with the replies above...

a pedantic view:

* the headline talks about "LM317"..but the problem is in now way related to LM317
* the question is not addressed to "switch ON condition". But for switch ON condition, I agree, that the C1 will cause a short peak... but very soon the current will be dominated by the 22,000 times bigger C2.
* for steady state it is far from being a short circuit. Thus my anser is: NO, it will not short circuit Neutral and Line.

* to your answer: The capacitor is not related to "direct current". At this node there simply is no direct current. An if there was DC, then the capacitor does "nothing", no current.
* what does " frequency of the communication " mean? Is there any communication via mains? or do you mean "frequency of commutation"?
* " if there is a high-frequency " .. that´s the key point. Not interference frome else where. It´s the diodes that generate high frequency swtiching noise. So the capacitor is meant to attenuate this "high frequency diode switching noise".

Other reasons for a capacitor at this node:
* to slightly increase the transformer output voltage
* to compensate for transformer phase shift (improve cos-phi)

Klaus