Best way to reduce ripple effect in transformer power supply?

Hey,
I want to design a low noise power supply and I wonder which method is better.
My transformer output is 14V AC at 20VA.
The power supply will be used for ham radio receiver application.
I will most likely feed that power supply to the 7812.

I am considering two solutions.
1. 2nd order RC Low-pass filter
https://www.adpslab.com/en/audio/audio-projects/14-simple-low-ripple-psu-for-dac.html
Schematic:

Low pass filter network

The decoupling network is a second order RC low pass filter. This network decreases noise transmission of the high-frequency content of any current that does pass through the resistor. The series elements are 2 small value resistors and DC voltage drop is acceptable because the current is low.

Click to expand...

2. pi-filter
https://www.sunpower-uk.com/glossary/what-is-ripple/

A more effective method of reducing the ripple voltage is the addition of a π-filter (pi-filter) at the output of the rectifier. This low pass filter consists of a two smoothing capacitors, as well as a choke to provide high impedance to the ac ripple.

Click to expand...

My question is, which is a better solution to reduce the ripple effect and get a stable voltage output?

How much rejection do you want? The common 7812 regulator will already have at least 60 dB ripple-rejection at 120 Hz.

Hi,

you say "reduce ripple" and you say "low noise".
The ripple is fixed frequency (twice mains frequency), while noise is a mixture of random frequencies.

So what do you want to solve?
If you want to solve both: Then what do you want to solve first?

And give a target value. (Please don´t say something like "best")
****

When you use a 20VA transformer you may expect less than 0.8A DC output current - not to overload the transformer.
RCRC: The resitors will drop the output voltage a little, but it is an uncritical circuit with good damping functinality.

PI: it generates less voltage drop and it will reduce ripple much more. But it generates a resonance at about 2.8Hz.

Read datasheet of 7812 about
* output noise
* load regulation
* PSRR

Klaus

Basically, I want a stable power supply that will behave like the gel battery.
I've talked with several hams and I noticed that people even say that "7812 is not enough" and they tend to power their radios with gel battery. They say it's the best method to avoid AC/mains noise and stuff. But I'd like to stick with power supplies, that's why I am looking for various ways to improve the output.

I'm not sure how 2.8Hz resonance would affect my receiversr, but they work at 3.5MHz or so (7MHz 40m band, 14MHz 20m band).

Ahhh, and I'm not sure which one affects the receiver more (ripple or high frequency noise from mains), but I guess it's the latter?

Hi,

I'm not sure which one affects the receiver more

Click to expand...

You need to know your enemy, it makes it easier to fight againt it.
I don´t know the answer.

In my eyes it makes no sense to to reduce ripple, if the reciever is sensible
* against HF noise
* absolute supply voltage drift
* temperature
...

Klaus

Klaus, I looked up the topic and it seems that I need a low RFI power supply - low radio frequency interference. Considering that, which solution is better?

Hi,

the bulk capacitor circuit is that that important for RFI.

How I´d do it:
* Transformer
* rectifier
* bulk capacitors (the capacitor need at least to ensure minimum operating voltage at the 7812 input at full load. But up to 10x larger capacitors won´t hurt.)
Please note that bulk capacitors usually are not made for RF. Adding some ceramics capacitors in parallel will improve this.
* 7812 circuit according datasheet
* common mode and differential RF filtering with appropropriate capacitors and chokes for your RF frequency. (To keep RF away from 7812 and keep voltage stable on application side)
Maybe you can use EMI filtering chokes. You need to read datasheets on this.
For other chokes I recommend to add a resonance damping RC combination.

Klaus