• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Should a power amplifier be grounded through RCA cables or power supply?

  • Thread starter Deleted member 58865
  • Start date
OP
D

Deleted member 58865

Guest
I'm interested in reducing absolute amount of noise. Does higher voltage power supply reduce absolute amount of noise?
 

sonder

Member
Joined
Mar 6, 2023
Messages
92
Likes
75
If I turn the volume knob to 3'o clock, the amplifier noise floor doesn't get reduced, but the noise floor of the source is increased. But, the amplifier noise floor is higher than the DAC noise floor anyway.

The pot is adding noise and distortion at 9 o'clock, and not wide open, less at 3 oclock, way more than anything else in your system, then it's being amplified at full power. Consider the amplification constant, the less you have that "volume" on the amp set, the more noise and distortion you're introducing right at the edge of the gain.

Bearing in mind the schiit I linked to is a passive dac that's half the price of your entire integrated amp + power supply, you can scale out what the pot you're filtering all your sound through 1/4 open in the amp is doing.

Your other thread has it all in there, either trust the dac and set volume there, or don't and leave your amp wide open and stick a good attenuator in between the two for safety.

Jump back in to the ASR review of your amp, it was done with less than 0.2vrms in and 3 o'clock volume to get that sinad. You're like 1.6vrns in and 9 o'clock.
 
Last edited:
OP
D

Deleted member 58865

Guest
The pot is adding noise and distortion at 9 o'clock, and not wide open, less at 3 oclock, way more than anything else in your system, then it's being amplified at full power.
How do you know this is the case? How is the volume knob on the amp any different from a passive (stepped) attenuator between DAC and amp?

In theory, the volume knob on the amp is no different from a passive potentiometer between DAC and amp.

Does the amp volume knob do anything different than attenuating the input signal?
 

sonder

Member
Joined
Mar 6, 2023
Messages
92
Likes
75
How do you know this is the case? How is the volume knob on the amp any different from a passive (stepped) attenuator between DAC and amp?
Just look at the ASR review of your amp again.

It was measured at 0.17vrms and 3 o'clock for a reason. You're at 1.6 and 9.

(Also the reason higher rating amps are pure power or integrated with better attenuation/ passives in them)
 

solderdude

Grand Contributor
Joined
Jul 21, 2018
Messages
15,891
Likes
35,912
Location
The Neitherlands
I'm thinking about upgrading AIYIMA A07's power supply to either Mean Well RS-100-24 or Mean Well GST60A24-P1J.
RS-100-24 doesn't have to be grounded. GST60A24-P1J has to be grounded.

The grounding is ONLY for the mains filtering. The safety ground is not connected to the DC output. It is only there to allow mains filtering to safety ground.
The grounded converter could, in theory, have lower leakage currents.
Some of the leakage currents from SMPS are caused by an almost obligatory capacitor being present between the DC out and mains in to ensure compliance to emission regulations.
In case of the grounded one that capacitor could well be connected to safety ground leading to lower leakage currents in the DC path.
Lowest leakage currents are from 'medical grade' power supplies.

When AIYIMA A07 is floating, if I rub the metal case, I feel electric zaps on my fingers.
Yes, this is the correct way to look for leakage currents.

I don't get this. Doesn't reducing amplifier volume reduce noise floor of the source as well?

Yes, but most amplifiers have a worse noise floor than digital sources so the actual noise floor one encounters remains the same.

I'm interested in reducing absolute amount of noise. Does higher voltage power supply reduce absolute amount of noise?
Why ? Noise floor is only a problem when it becomes audible and is determined by transducer sensitivity.
When you hear noise from transducers with no signal and the volume turned up to the max. level you would be using it the only ways to lower noise is by using different amplifiers, sources or to attenuate the output of the device that drives the transducers.
 
Last edited:
OP
D

Deleted member 58865

Guest
You're at 1.6
How is output voltage of Topping D10s 1.6V at -20dB or -16dB?
The pot is adding noise and distortion at 9 o'clock, and not wide open, less at 3 oclock
Does the potentiometer on the amp add more absolute amount of noise at 9'o clock than at 3'o clock? Or, is it the noise relative to signal?

Amir didn't measure absolute amount of potentiometer noise at various positions of the pot.

My theory is that reducing DAC volume is not different from reducing the potentiometer volume. Both reduce signal levels. Absolute amount of pot noise stays the same in theory.
 
OP
D

Deleted member 58865

Guest
https://www.audiosciencereview.com/...-of-schiit-sys-passive-pre-amp-switcher.2335/ says
As we lower the level on Sys, naturally the measurement noise as a percentage rises leading to the graphs that you see. In other words you lose signal to noise ratio with passive volume controls.
This means the noise relative to signal increased as the volume on the potentiometer decreased. The absolute amount of noise stays the same in theory.

DAC outputs voltage levels. Potentiometer attenuates voltage levels. Reducing DAC volume reduces voltage signal levels. Reducing pot volume reduces voltage signal levels.
The absolute amount of noise in the system stays the same supposedly.

I think avoiding the pot noise would require a stepped attenuator knob that controls amplifier gain.
 
OP
D

Deleted member 58865

Guest
Why ? Noise floor is only a problem when it becomes audible and is determined by transducer sensitivity.
I still want to decrease noise floor of the amp. But, TPA3255 measurements do not include noise floor at various voltage levels.

I guess voltage levels don't affect noise floor?
 

solderdude

Grand Contributor
Joined
Jul 21, 2018
Messages
15,891
Likes
35,912
Location
The Neitherlands
The noise floor of the amp is determined by the noise of the amplifier circuit, PSRR (Power Supply Rejection Ratio) of the amp circuit and its internal power supply decoupling/regulation.
Not by an external power supply voltage or noise levels unless the the amp design itself is pretty crappy (poor PSSR and noisy components).
The sensitivity to outside noises (common mode, induction of HF) etc. is determined by circuit design, filtering and above all component choice, PCB and wiring layout.
 
Last edited:
OP
D

Deleted member 58865

Guest
The noise floor of the amp is determined by the noise of the amplifier circuit, PSRR (Power Supply Rejection Ratio) of the amp circuit and its internal power supply decoupling/regulation.
Between reducing DAC volume and turning down the amplifier volume pot, which one introduces more noise to the entire system? I guess both introduce roughly the same amount of noise unless the amp volume pot actually injects more noise at lower volume levels.
 
OP
D

Deleted member 58865

Guest
Let's move the discussion about volume setting to
 

solderdude

Grand Contributor
Joined
Jul 21, 2018
Messages
15,891
Likes
35,912
Location
The Neitherlands
I don't know what this means.

It means that the safety ground is only connected to a filtercircuit in the power supply on the 115/230V side and NOT connected to the - (or +) on the DC side.
So the safety ground is not present on the DC side.
 
OP
D

Deleted member 58865

Guest
the safety ground is only connected to a filtercircuit in the power supply on the 115/230V side
That means the power supply is leaking any current to the electrical ground? Is it not good enough? What can go wrong? Should I connect the amp case to the electrical ground even if the power supply is grounded?
 

solderdude

Grand Contributor
Joined
Jul 21, 2018
Messages
15,891
Likes
35,912
Location
The Neitherlands
No it means HF crap on the mains is routed to safety ground. Unfortunately for HF the ground wire in the cable and walls is very high impedance and will not remove that noise.
The 3rp prong is safety ground and not 'clean' at all. It also is not an audio ground. Only for safety reasons.
Depending on how the PS is built there might be somewhat less leakage between mains and DC.

Best to use medical grade power supplies (not meaning medical grade mains sockets/cables) IF leakage is an issue and you get weird sounds or hum. Then this might help (no guarantees)
 
OP
D

Deleted member 58865

Guest
Does it help to use a grounded power supply, then? Someone else pointed out that it protects RCA cables and PCB print-outs from being burnt.

I think GST60A24-P1J would inject less noise into the speakers than the 32V 5A power supply sold with AIYIMA A07.
 
Last edited by a moderator:

solderdude

Grand Contributor
Joined
Jul 21, 2018
Messages
15,891
Likes
35,912
Location
The Neitherlands
Does it help to use a grounded power supply, then? Someone else pointed out that it protects RCA cables and PCB print-outs from being burnt.
Probably AI that made that claim or someone that quoted a response from AI (or Paul or similar 'guru') that made a comment not actually related.

A 2-prong and 3-prong power supply that has a DC output both are isolated from mains. Both can have some leakage (mA range) which may become audible in some devices.
Burning out RCA cables, XLR cables and PCB's will only happen when a power brick fails catastrophically. This is unlikely with any decent power supply. Maybe some no-brand cheap power supply not meeting any requirements might, someday, in an isolated case be able to do something like that.

Safety ground is there for safety reasons and those reasons alone. When safety ground is present, in case of an internal catastrophic failure, the metal parts that can be touched are still safe to touch.
Class-II (double isolated, 2 prong) is double isolated and can not have metal parts you can touch that might come in contact when a catastrophic failure inside occurs.

SAFETY ground... not audio ground.
 
OP
D

Deleted member 58865

Guest
If I live in a rural area, I may have access to electrically clean ground around my house. I can connnect my amp to a rod inserted into the ground.
 
OP
D

Deleted member 58865

Guest
A 2-prong and 3-prong power supply that has a DC output both are isolated from mains.
If a power supply grounds itself through RCA cables, would RCA cables not pick up some noise from the power supply? I know AIYIMA A07 is grounding through RCA cables unless it is paired with a grounded power supply. The power supply noise is routed through the amp, the DAC, and my computer to the electrical ground.
 

solderdude

Grand Contributor
Joined
Jul 21, 2018
Messages
15,891
Likes
35,912
Location
The Neitherlands
If a power supply grounds itself through RCA cables, would RCA cables not pick up some noise from the power supply? I know AIYIMA A07 is grounding through RCA cables unless it is paired with a grounded power supply. The power supply noise is routed through the amp, the DAC, and my computer to the electrical ground.

When the RCA is connected to the chassis (metal parts) of an enclosure then the connected power supply (DC input) is always isolated.
The reason for that could be to prevent hum or to ensure compliance to EMC rules/guidelines/standards.
Only when one has an audio device that has a 3-prong mains inlet the chassis should be connected to safety ground.
There are audiophool brands that make amps that do not comply to these safety standards.

The power supply with safety ground (that outputs DC) does NOT have the DC plug connected to safety ground. These power supplies STILL can have a leakage current that can still become audible in poorly designed equipment.

re: Should a power amplifier be grounded through RCA cables or power supply?​


The answer is simple. When the amplifier is a class-I product (internal 3 prong power supply) the chassis should be connected to safety ground. The internal audio ground doe not have to be. One should NEVER have to rely on an audio cable to provide safety ground.
When the amplifier is class II (double isolated) or is fed via an isolated external power supply it does not have to be 'grounded' (connected to safety ground).
Such a device might still be connected to safety ground via (audio) interlink cables though but does not matter from a safety standpoint.

Ground on mains is ONLY about safety... nothing else. That may 'bite' certain audio systems in the rear though... but it is about safety.

Audio ground (common) ≠ safety ground..... but could be tied together.
 
Last edited:
Top Bottom