Thanks for your insight! I had the same thought regarding Class A. In theory, transients shouldn't affect the power supply, but I always test my theories. I plan to build several designs to see if there is any difference in sound for any reason. I also plan to build +5V supplies of the same designs for a USB DAC. Thanks again!
One point is the slew rate required from the opamp in a regulator. That slew rate requirement is minuscule. The output of the opamp which drives the pass device will never change more than a few 10's of mV, 100mV tops.
100mV change at 20kHz is a slew rate of 0.026V/us ...
Jan
100mV change at 20kHz is a slew rate of 0.026V/us ...
Jan
3 years ago i used to test very expensive, precise(three decimals precision), high power(4kw) smps, around 100khz switching frequency and all the op-amps(about 20 pcs per board ) used there were opa2227 ( around 2V/us slew rate).All the power sources we built (up to 300kV/65 kVA) used the same op-amp and the test for the 300kV/65kVA supplies was to short circuit the output for 300 times, at 100kV, 200kV and 300kV...All power supplies made by that company used a 2.3 V/us op-amp for sensing the currents and voltages change...The capacitors would be the ones to handle all the trouble there, not the op-amps.
Last edited:
@Safesphere - you can try this one too
VRDN: bipolar regulator PCB for line level ckts: ±11V to ±20V @ 1.5A with "De-Noiser"
VRDN: bipolar regulator PCB for line level ckts: ±11V to ±20V @ 1.5A with "De-Noiser"
Hello safesphere,
I think if I were in your shoes, I might buy a power transformer with 18VAC or even 20VAC secondaries. Then I think I would probably build a cascade of two voltage regulators in series. There's plenty of voltage headroom to do this: 18VAC + Schottky rectifiers gives 24VDC raw. Let the first regulator have 2V in-to-out and let the second regulator have 4V in-to-out. Then: 24V --> reg1 --> 22V --> reg2 --> 18V
The only objective of the first ("upstream") regulator is to reduce ripple and create a much cleaner supply for the second ("downstream") regulator. Now, for the second regulator, you can apply whatever nifty circuits you like, secure in the knowledge that they will operate in a clean(er) environment.
And you get an extra 40-70dB of PSRR at the mains and 2Xmains frequencies.
You can also install RFI filtering and even ferrite bead suppression, before each of the two regulators.
_
I think if I were in your shoes, I might buy a power transformer with 18VAC or even 20VAC secondaries. Then I think I would probably build a cascade of two voltage regulators in series. There's plenty of voltage headroom to do this: 18VAC + Schottky rectifiers gives 24VDC raw. Let the first regulator have 2V in-to-out and let the second regulator have 4V in-to-out. Then: 24V --> reg1 --> 22V --> reg2 --> 18V
The only objective of the first ("upstream") regulator is to reduce ripple and create a much cleaner supply for the second ("downstream") regulator. Now, for the second regulator, you can apply whatever nifty circuits you like, secure in the knowledge that they will operate in a clean(er) environment.
And you get an extra 40-70dB of PSRR at the mains and 2Xmains frequencies.
You can also install RFI filtering and even ferrite bead suppression, before each of the two regulators.
_
Attachments
A friend of mine is restoring a Sony TA-ax6 amp and its phono preamp supplied with +-16.4 v derived from +-24v derived again from the main +-32 regulator aimed at supplying the first stages of the power amp, and i saw it's power supply idea like an interesting one, especially because you can supply any type of phono preamp with it and very similar idea with this one: https://www.diyaudio.com/forums/power-supplies/359652-fine-ic-voltage-regulators-25.html#post6348373
Attachments
Last edited:
With so much great advice here, I plan to start with building 2 or 3 different designs to see if any of them sounds better. I can use the same transformer/rectifier to compare different regulators. Then I can put them in series like you've suggested and see if this makes any difference in my system. Than you for the idea!
Well, to make the scratches etc as "silent" as possible, it needs to handle them swiftly and without distorsion...
//
My phono prreamp is the original first Rega Aria MM/MC
I should be getting soon one of the good models of Lyra
Hello,
I've just finished design and now order pcba for a balanced TPA headphone amp. Do you know if for the Salas SSLV1.3 UltraBiB shunt regulator, the gerber files are available? Or, for other superreg designs? I need positive and negative supply both.
Thanks,
JG
I've just finished design and now order pcba for a balanced TPA headphone amp. Do you know if for the Salas SSLV1.3 UltraBiB shunt regulator, the gerber files are available? Or, for other superreg designs? I need positive and negative supply both.
Thanks,
JG
O.K., so, I understand the gerber is not public. I have a question. If I just make a board design based on the superreg design here, am I breaking any written on non written rule?
Thanks,
JG
Thanks,
JG
Hello Jan,
I have bought the boards for your regulator and I have a quick question. There are ten 120uF caps on the BOM. Can I replace them with either 100uF or 220uF without negatively affecting performance? If I can, which of them (C1-C10) should be 100uF and which 220uF? And which must stay 120uF? Thank you so much!
Alex