For the small package the power supply may same on the PCB of amplifier stage.
But about the noise that will happen in the system and out to speaker.
Which one have more noise between same on PCB or Separated PCB ?
But about the noise that will happen in the system and out to speaker.
Which one have more noise between same on PCB or Separated PCB ?
Separate PCB's for low noise, in fact you don't really need a PCB for the PSU if it's just a couple of caps & a choke, keep it simple, your amp though. If you do go down putting it all on one PCB make sure you pay attention to layout regarding ground current. First connect the OP stage, then next lowest current draw valve with the IP valve right at the end.
Andy.
Andy.
Both can be exactly the same, depending on how you implement it.
Think about it: The only difference is how much distance is between the two circuit sections. That distance could be made from air, or it could be an empty section of PCB material, it's doesn't matter either way.
I'd do it separately, as @Diabolical Artificer said. In addition, it gives you more flexibility, both with the location and the design of the power supply.
Ask first: will the power transformer couple noise into the signal circuit?
If the PT is large, or has high leakage inductance it may cause problems for LP RIAA stages or line amplifiers. In this case, the transformer might be better to be moved away from the signal circuit.
One thing you can always do, to minimise the noise-coupling of a power supply:
- keep the length of wiring: transformer -> rectifier, and rectifier -> electrolyt capacitors as short as possible, and preferably <50mm.
These wires carry the recharge pulses to the caps; the peak current in these wires can reach 5* - 7* the DC current!
Long wires and large current peaks means big B-fields (low impedance magnetic fields), and these couple into the "victim" circuit very easily.
Conequently:
If the power transformer cannot be mounted close to the signal circuit, the power supply (Raw DC rectifiers & caps) should be mounted near the transformer, not near the signal circuit.
If the PT is large, or has high leakage inductance it may cause problems for LP RIAA stages or line amplifiers. In this case, the transformer might be better to be moved away from the signal circuit.
One thing you can always do, to minimise the noise-coupling of a power supply:
- keep the length of wiring: transformer -> rectifier, and rectifier -> electrolyt capacitors as short as possible, and preferably <50mm.
These wires carry the recharge pulses to the caps; the peak current in these wires can reach 5* - 7* the DC current!
Long wires and large current peaks means big B-fields (low impedance magnetic fields), and these couple into the "victim" circuit very easily.
Conequently:
If the power transformer cannot be mounted close to the signal circuit, the power supply (Raw DC rectifiers & caps) should be mounted near the transformer, not near the signal circuit.
A ground plane is a good idea. But do keep the rectifier+reservoir capacitor separate, with a single connection to the rest of the ground plane. Also be aware of creepage distances between copper, if you are using high voltages.
Merlin,
from your article, page 6, "Also note that ground currents accrue as we get closer to the reservoir, so in this case V1 is produced by only one noisy current whereas V2 is produced by the sum of two noisy currents."
That looks like a typo to me, "V2" and "V1" ought to be the other way around.
kind regards
Marek
from your article, page 6, "Also note that ground currents accrue as we get closer to the reservoir, so in this case V1 is produced by only one noisy current whereas V2 is produced by the sum of two noisy currents."
That looks like a typo to me, "V2" and "V1" ought to be the other way around.
kind regards
Marek
I integrate it all on one PCB these days. More compact and less wires. Layout is important no matter what.
You're right, well spotted, will fix when I get round to it
I'm building the Salas 6V6 line Amp at this time. I'm going with separate power supply chassis which will only contain transformers, all the rectifiers, CRCRC filters, and a delay timer of the B+. All linear. I need three voltages, heater, B+ and a 12V supply for the headphone module I'm adding, the relay volume control, and the relay input selector. So basically just the "raw DC" voltages needed will be emanating from this "brick". In the preamp chassis will be the regulators for both B+ and heater and pos/neg regulator for the headphone Amp module (Neurochrome HP-22 board). So regulation will be near the circuit and "raw DC" will be via an umbilical cord. I believe this will help take the length of the umbilical out of the audio equation as the regulation will be inches from the circuit, not three feet away.
I'm doing it this way mostly because I have limited time for this hobby, so I want a dedicated PS box that I can possibly reuse for another project if the B+ is close. So for me its about productivity and choosing a second project where I won't have to spend time building a PS at all because I'll look for a project that works with this same B+ and heater voltage. For me about use of my time in this hobby and wanting to do multiple projects.
I'm doing it this way mostly because I have limited time for this hobby, so I want a dedicated PS box that I can possibly reuse for another project if the B+ is close. So for me its about productivity and choosing a second project where I won't have to spend time building a PS at all because I'll look for a project that works with this same B+ and heater voltage. For me about use of my time in this hobby and wanting to do multiple projects.