I have been asking this over various threads and have recieved many good thoughts, I appreciate it greatly. I looked into a CRC power supply, but thanks to many of you, I found that such a design is impractical with a highcurrent device like an amplifier.
Ok I think what I designed is a PI Filter, but please correct me if I'm wrong. The design uses a 10mf capacitor followed by a series inductor rated at 18uh and a DCR of .117ohms. Then followed by around 40,000uf of capacitance. It gave I think around 300 micro farads of ripple, which is pretty low compared with 100,000uf of capacitance.
Ok I think what I designed is a PI Filter, but please correct me if I'm wrong. The design uses a 10mf capacitor followed by a series inductor rated at 18uh and a DCR of .117ohms. Then followed by around 40,000uf of capacitance. It gave I think around 300 micro farads of ripple, which is pretty low compared with 100,000uf of capacitance.
Hi,
that's a very small inductor.
It's not going to store much energy.
An air cored 1mm wire 35Turns in 5layers on a 13mm bobbin will give about 19uH and 0r044.
You should be able to get at least 10 times your inductance and stay under 0r1 with a gapped iron core.
I wonder if 0r1 & 1 to 2mH would be a better target?
Your caps seem OK and will support one 4ohm amplifier or two 8ohm amplifiers.
Keep in mind the first cap ripple rating.
that's a very small inductor.
It's not going to store much energy.
An air cored 1mm wire 35Turns in 5layers on a 13mm bobbin will give about 19uH and 0r044.
You should be able to get at least 10 times your inductance and stay under 0r1 with a gapped iron core.
I wonder if 0r1 & 1 to 2mH would be a better target?
Your caps seem OK and will support one 4ohm amplifier or two 8ohm amplifiers.
Keep in mind the first cap ripple rating.
Hi Pj,
what version of PSUD2 are you using?
My copy only allows LC that are greater than 1mH. How did you specify 18uH?
I substituted 1.2mH for 0r1 as the last series element and the ripple hardly changed.
output voltage went from 30.78V/31.34V(RCRC) to 31.66V/32.2V (RCLC) i.e 560mVpp to 540mVpp (8r0 load)
Where are you getting 300uVpp ripple at what load.
Keep in mind that cascaded single pole filters are brilliant at attenuating mains frequency. Your challenge is making the Pi filter good at supplying all the current demands of the amplifier. That is what Pi is not so good at and why they tend to be used on high voltage low current amplifiers (tubes/valves).
what version of PSUD2 are you using?
My copy only allows LC that are greater than 1mH. How did you specify 18uH?
I substituted 1.2mH for 0r1 as the last series element and the ripple hardly changed.
output voltage went from 30.78V/31.34V(RCRC) to 31.66V/32.2V (RCLC) i.e 560mVpp to 540mVpp (8r0 load)
Where are you getting 300uVpp ripple at what load.
Keep in mind that cascaded single pole filters are brilliant at attenuating mains frequency. Your challenge is making the Pi filter good at supplying all the current demands of the amplifier. That is what Pi is not so good at and why they tend to be used on high voltage low current amplifiers (tubes/valves).
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.