Any class A amp does not have constant current when signal is applied. It can vary from just above zero to just below saturation assuming no clipping.
Other design specs could include p-p ripple voltage on the DC output, RF rejection on the incoming AC line, holdup time capability for AC interrupts and/or dropouts. These all vary according to the needs of the load circuit.
You've got two secondaries and two bridge rectifiers. From there, you could have two identical, isolated filter and regulator circuits, and just connect them at the outputs, the negative of one to the positive of the other. This has several advantages, though they may be subtle. One is you can use the best regulator in both locations. Usually the positive regulator is better (for specs like low noise) than the negative one, so that would mean LM317. Having the same circuit twice helps reduce confusion over where one component goes vs. its complement, and all electrolytics and diodes go the same direction in both sections.
C18/C19 can be 0.1uF and R11/R12 can be 1 ohm. I'd do that AND have a 0.1uF directly across each secondary. Furthermore, I'd replicate those three components across each of the eight rectifier diodes. That may seem like a lot, but they're not expensive and I'm convinced it gets rid of 60hz buzz that could otherwise creep in. Here's a thread (one of many - this may not be the best description) describing more than you want to know about the topic:
https://www.diyaudio.com/community/...rmer-snubber-using-quasimodo-test-jig.243100/
You show Jumper JP1, but with no description of what it does. When installed it appears to lower the +28V output voltage to about 3 volts.
You say this is for a microphone preamp, but this still leaves questions. Why does +28V have a 300mA load and the -28V only a 30mA load? Most low-level circuits pull about the same current from both supplies. What does the +24V output power?
C18/C19 can be 0.1uF and R11/R12 can be 1 ohm. I'd do that AND have a 0.1uF directly across each secondary. Furthermore, I'd replicate those three components across each of the eight rectifier diodes. That may seem like a lot, but they're not expensive and I'm convinced it gets rid of 60hz buzz that could otherwise creep in. Here's a thread (one of many - this may not be the best description) describing more than you want to know about the topic:
https://www.diyaudio.com/community/...rmer-snubber-using-quasimodo-test-jig.243100/
You show Jumper JP1, but with no description of what it does. When installed it appears to lower the +28V output voltage to about 3 volts.
You say this is for a microphone preamp, but this still leaves questions. Why does +28V have a 300mA load and the -28V only a 30mA load? Most low-level circuits pull about the same current from both supplies. What does the +24V output power?
Thanks for your comments. I wasn't very clear. The circuit would be a dual channel strip (i.e two mic preamps with EQs). All discrete.
To answer some question:
- I realise that pre-regulated and post regulated voltage ripple is a requirement but that's pretty much what we're discussing here since I don't know what should I be aiming for.
- The +24V rail is for LEDs and relays (I mentioned it in a previous post).
- Jumper JP1 is in order to be used only once when the trimmer is set to the correct value so, it can be ignored.
- In the past I have used 10nF caps across each diode. Is 1ohm+100nF a better solution?
I will revise the schematic and upload it when I find some free time.
To answer some question:
- I realise that pre-regulated and post regulated voltage ripple is a requirement but that's pretty much what we're discussing here since I don't know what should I be aiming for.
- The +24V rail is for LEDs and relays (I mentioned it in a previous post).
- Jumper JP1 is in order to be used only once when the trimmer is set to the correct value so, it can be ignored.
- In the past I have used 10nF caps across each diode. Is 1ohm+100nF a better solution?
I will revise the schematic and upload it when I find some free time.
I am absolutely no expert . However , as I recall , mic level is about 3 mV and line level is 0.775 Vrms .
So that's a voltage gain of 258 V/V ( 48 dB's )
Then any noise in the line level is going to amplified by further stages .
So even though the mic amp may have a great PSRR , its still critical to have ultra quiet supply rails .
For audio , split bobbin transformers are good because they have low capacitive coupling between the primaries and secondaries .
But E cores can hum , and r cores can be difficult to source ,
So consider the Hammond 229 series , which have rating up to 48VA , but these need a circuit board to mount on .
As I've read on this site , power transformers can be pushed up to about 1/2 their power rating .
IMO for audio equipment , the power supply is 1/2 the sound .
... and there are no short cuts .
Put it this way , for all your time and effort , plus the case and sourcing all the parts , there is no point in going cheap on the power supply .
Instead of chip regulators , consider the Sigma r22 dual rail supply .
https://www.amb.org/audio/sigma22/
Plus on the way , you could learn about a differential pair ... which is a very cool circuit .
Here is a great series of articles , where noise on the supply rails , caused by various battery chemistry's ,
chip regulators and shunt regulation is discussed .
https://www.tnt-audio.com/clinica/regulators_noise1_e.html
.
So that's a voltage gain of 258 V/V ( 48 dB's )
Then any noise in the line level is going to amplified by further stages .
So even though the mic amp may have a great PSRR , its still critical to have ultra quiet supply rails .
For audio , split bobbin transformers are good because they have low capacitive coupling between the primaries and secondaries .
But E cores can hum , and r cores can be difficult to source ,
So consider the Hammond 229 series , which have rating up to 48VA , but these need a circuit board to mount on .
As I've read on this site , power transformers can be pushed up to about 1/2 their power rating .
IMO for audio equipment , the power supply is 1/2 the sound .
... and there are no short cuts .
Put it this way , for all your time and effort , plus the case and sourcing all the parts , there is no point in going cheap on the power supply .
Instead of chip regulators , consider the Sigma r22 dual rail supply .
https://www.amb.org/audio/sigma22/
Plus on the way , you could learn about a differential pair ... which is a very cool circuit .
Here is a great series of articles , where noise on the supply rails , caused by various battery chemistry's ,
chip regulators and shunt regulation is discussed .
https://www.tnt-audio.com/clinica/regulators_noise1_e.html
.
