Guys I know this has been discussed before but the more I read the more confusing it become.😕
Objective: make a clean power supply from a 12V RMS transformer with a bridge rectifier.Current draw is less than 1 Ampere at any time.
Does anyone has a simple CLC circuit to share with me. Please be gentle as I am a hobbyist and know how to solder. So please help me with a fully functional circuit with all tits and bits than concepts.
Hope someone will help.
Objective: make a clean power supply from a 12V RMS transformer with a bridge rectifier.Current draw is less than 1 Ampere at any time.
Does anyone has a simple CLC circuit to share with me. Please be gentle as I am a hobbyist and know how to solder. So please help me with a fully functional circuit with all tits and bits than concepts.
Hope someone will help.
Single polarity supply needs 1 transformer, 1 bridge rectifier, 2 smoothing capacitors.
You can use 1 inductor (for CLC), or 1 resistor (for CRC).
A useful recommendation is 2200uF for each amp of output.
Since your maximum output is 1A then 2200uF would be sufficient if you only had 1 smoothing capacitor.
You could reduce that to 1mF + 1mF for the CRC, or keep the same value and use 2m2F+2m2F for a bit less ripple.
The capacitors should be 25V (or more), since 16V will see a bit too much voltage.
You can use 1 inductor (for CLC), or 1 resistor (for CRC).
A useful recommendation is 2200uF for each amp of output.
Since your maximum output is 1A then 2200uF would be sufficient if you only had 1 smoothing capacitor.
You could reduce that to 1mF + 1mF for the CRC, or keep the same value and use 2m2F+2m2F for a bit less ripple.
The capacitors should be 25V (or more), since 16V will see a bit too much voltage.
Thanks Andrew. Can you please explain this a bit more "You could reduce that to 1mF + 1mF for the CRC, or keep the same value and use 2m2F+2m2F for a bit less ripple.
The capacitors should be 25V (or more), since 16V will see a bit too much voltage. "
The capacitors should be 25V (or more), since 16V will see a bit too much voltage. "
Last edited:
A 230:12+12Vdc transformer (or 115:12+12 in the USA) cna and will see a range of supply voltage.
If you run the 230Vac on 254Vac (the maximum in the UK) then the output will be:
Vac secondary = 254?230*12*transformer regulation.
Assuming a transformer regulation of 10% (very small transformers can go as high as 30%) then the maximum secondary voltage is 254/230*12*(1+10%) = 14.58Vac
The maximum peak voltage of the sinewave is 14.58*aqrt(2) = 20.6Vpk
The bridge rectifier can drop much less than 700mVf whne at very low current demand.
I generally use 500mVf for checking worst case smoothing capacitor voltage.
Max cap voltage ~20.1Vdc
This is well above what a 16V capacitor can hold.
Use 25V or more.
The simplistic way to predict smoothing cap voltage is:
Vac * sqrt(2) - 1.4V which for a 12V transformer gives 15.6Vdc, giving the "WRONG" impression that a 16V capacitor is OK !
Many on this Forum do the check in the simplistic form. And worse they tell other Members to copy their WRONG method !!!!
If you run the 230Vac on 254Vac (the maximum in the UK) then the output will be:
Vac secondary = 254?230*12*transformer regulation.
Assuming a transformer regulation of 10% (very small transformers can go as high as 30%) then the maximum secondary voltage is 254/230*12*(1+10%) = 14.58Vac
The maximum peak voltage of the sinewave is 14.58*aqrt(2) = 20.6Vpk
The bridge rectifier can drop much less than 700mVf whne at very low current demand.
I generally use 500mVf for checking worst case smoothing capacitor voltage.
Max cap voltage ~20.1Vdc
This is well above what a 16V capacitor can hold.
Use 25V or more.
The simplistic way to predict smoothing cap voltage is:
Vac * sqrt(2) - 1.4V which for a 12V transformer gives 15.6Vdc, giving the "WRONG" impression that a 16V capacitor is OK !
Many on this Forum do the check in the simplistic form. And worse they tell other Members to copy their WRONG method !!!!
A CRC or CLC filter gives better attenuation of the rectified mains sinewave.
Instead of using a single 2m2F smoothing cap you can model a 1mF+1mF pair of smoothing caps in the CRC topology and you will probably find that ripple is lower than for the single higher value cap.
Try using PSUD2 to help with the modeling.
Instead of using a single 2m2F smoothing cap you can model a 1mF+1mF pair of smoothing caps in the CRC topology and you will probably find that ripple is lower than for the single higher value cap.
Try using PSUD2 to help with the modeling.
I tried playing with PSUD2, a bit complex for me. I am sorry but did you mean 20,000uf by using 2m2F? and I guess 10,000uf by 1mF? What would be the appropriate inductor value to choose?
I think Andrew meant you could stick with two 2200uF caps or, given the rule of thumb he states of 2m2 uF per amp of output which would suggest a single 2200uF cap would be sufficient capacitance, use two 1,000uF (1m uF) caps (with a resistor or inductor in between the two).
1mF = 1 milli-Farad
2m2F = 2.2 milli-Farad
A single 2m2F is often recommended as sufficient for 1A load current.
But two 1mF in CRC will probably give less ripple than a single 2m2F.
Check by modeling in PSUD2
2m2F = 2.2 milli-Farad
A single 2m2F is often recommended as sufficient for 1A load current.
But two 1mF in CRC will probably give less ripple than a single 2m2F.
Check by modeling in PSUD2
Hi Andrew,
Can you tell us how do you get the value of 2,200uF per amp?
Any calculation involved or based on past experience?
Thank you.
Can you tell us how do you get the value of 2,200uF per amp?
Any calculation involved or based on past experience?
Thank you.
that is a rule of thumb for ~1V RMS of ripple at 120Hz
Calculate the capacitance you need from the following formula ...
C = ( I L / Δ V ) * k * 1000uF
where I L is load current, ΔV is ripple voltage
k = 7 for 100Hz or 6 for 120Hz ripple frequency
where I L is load current, ΔV is ripple voltage
k = 7 for 100Hz or 6 for 120Hz ripple frequency
So let's say Kinku goes with two 2200uF filter caps. His next question is likely, again, "how do I size the inductor in the CLC filter?"
set the LC corner freq to get the amount of attenuation from the 1st caps ripple voltage. a 2 pole filter here provides -12 dB /octave.
note > under damped L is very dangerous in practical use! both DC losses as well AC and transient analysis required!
note > under damped L is very dangerous in practical use! both DC losses as well AC and transient analysis required!
Last edited:
SGK you are right that is my question,I would go with 1mF on both side of inductor but what is the value of inductor ,at least range,I read somewhere between 1.8mH to 3 mH ?
those values put Fc close to the ripple frequency so depending on the filters Q could actually see gain in ripple voltage. need to model in a circuit simulator Q= Xl/Rdc
not recommended for novice , use a resistor eg CRC 1st order filter / atten at 120Hz ~20LOG Xc/(R*Xc)
Xc=1/(2*PI*f*C), f= 120
if C= .001, R = 1 ohm 20*LOG(0.57)
it's pretty silly using passive filters to get "clean power" you need to use a solid state V regulator following the CRC.
most any voltage regulator needs at least 3 volts headroom minimum in the input so shoot for using a 13 -15 VAC transformer.
not recommended for novice , use a resistor eg CRC 1st order filter / atten at 120Hz ~20LOG Xc/(R*Xc)
Xc=1/(2*PI*f*C), f= 120
if C= .001, R = 1 ohm 20*LOG(0.57)
it's pretty silly using passive filters to get "clean power" you need to use a solid state V regulator following the CRC.
most any voltage regulator needs at least 3 volts headroom minimum in the input so shoot for using a 13 -15 VAC transformer.
Last edited:
what are you planning to power ?
eq how clean does clean need to be.
look here > 1/2 of this circuit will be closer to clean http://sound.westhost.com/project44.htm
can use a common 12.6V at 1-2 amp XFMR with a single winding
eq how clean does clean need to be.
look here > 1/2 of this circuit will be closer to clean http://sound.westhost.com/project44.htm
can use a common 12.6V at 1-2 amp XFMR with a single winding
Last edited:
It is for a audio project with Raspberry pi,so the current consumption will be close to 800mA.I am going with Salas regulator(Reflektor) to power it but thought will make the supply to Salas a bit cleaner. So if anyone can help me with just a CLC or CRC circuit to lower ripple before Reflektor it will be great.
Hagerman has an easy to use online calculator that computes required power supply LC filter values. It's the last one, titled, "Choke Filtered Supply".
Hagerman Technology LLC: Design Theory
Hagerman Technology LLC: Design Theory
- Status
- Not open for further replies.