Impact of AC Distortion & Noise on Audio Equipment

[solderdude]

Tittle here being: "Impact of AC Distortion & Noise |on Audio Equipment]". Noise from mains not being only sine distortion.

That's the whole point what PMA was talking about... It is different in all situations so will differ from any circumstances in papers as well.

To solve issues in audio:
First try to eliminate at the source.
When that isn't possible try to find out where the nasties get into the audio chain.
That could be airborn, magnetic, via wiring, via mains, via audio cables to name but a few.
Then try to solve the area(s) where the problems seem to find its way into the audio path.

Taking a blunderbuss approach like conditioners, filters, cables, wire routing, using other gear, moving components to another location (sometimes moving a device a few cm/inch is already enough) can help.\

Dimmers, when they are an issue, usually create a buzzing sound, not pops. Pops can be motors or actuators, valves or whatever other devices being switched on or off can do that. May even be from your neighbors.
Hard to diagnose when random and only happening sometimes. Logging may be needed for this.

 

[David_M]

From the speaker's viewpoint it does not matter if a SE or balanced output is connected. The speaker does not care if one connector is tied to (signal)ground or not. Across the speaker terminals there is just a signal. The output resistance of the amp is low and voltage levels are high.

If the noise incoming into the amp or exiting the amp is common mode, then it matters if the amp is SE or balanced. If the former, the noise will be audible (if large enough). If the amp is balanced, then the common mode is common to both + and - speaker terminals and will not be heard as the potential difference of CM noise is zero.

 

[b4nt]

A general question:

If 1khz measurements show no difference, would it automatically mean no difference for all frequencies?

Anyway, I beleive that test was wrong, with sine distortion only. If you feed this to a SMPS 220V/DC converter, it will rectify and filter out all that stuff, your DC output will be clean (with the very usual SMPS output ripple and SMPS related noises). And the DAC will show exact the same performances;

 

[solderdude]

If the noise incoming into the amp or exiting the amp is common mode, then it matters if the amp is SE or balanced.

For amp inputs... of course. That's really the only reason for having balanced inputs.
For balanced outputs for speakers it does not as there is no path to ground from the speaker.

 

[b4nt]

First try to eliminate at the source.

How do you elimitate the noise caused by pluging to mains a 30W resistive solder iron?

 

[David_M]

The nature of possible audible mains interference into audio chain is transient, impulse (switching of breakers, ripple control, motors switching etc.). Not the steady state. So it cannot be measured with steady state distorted wave as an input. Such measurement is close to pointless. It is necessary to catch and analyse interference impulse voltages at the output of audio components.

View attachment 145582
Multiple measurements of amplifier output (persistence mode). Please see random occurrence of wideband transient spikes.
This kind of interferences can be fixed by good line filtering.

Transient noise is just one type of noise. Power lines are basically antennas for AM radio, which in the US, is in the 550kHz to 1720kHz, even for FM radio (88MHz - 108MHz) to some degree. This is continuous noise 24/7. Even our cable TV systems have an 'FM Trap' circuit, to prevent FM radio signals from being passed on to the user by the high bandwidth coax cable.

 

[solderdude]

Anyway, I beleive that test was wrong, with sine distortion only. If you feed this to a SMPS 220V/DC converter, it will rectify and filter out all that stuff, your DC output will be clean (with the very usual SMPS output ripple and SMPS related noises). And the DAC will show exact the same performances;

That's the whole point. In order to see if gremlins make it into an FFT plot these would also have to be repetitive otherwise it would not show up in an FFT. So small 'ticks' that made it through (common mode) and entering in the signal would not be detected.

Amir's measurements show that the 'dimmer voltage' which is fed to the power supply (so not a dimmers spectral garbage from a dimmer in the house) with its reasonably fast but not MHz fast and fairly low transients that are not on top of the AC signal is kind of a valid test signal.
If that crap (the fast transient with switching on after 35 degrees) would make it into the circuit is would show up.

It can't because the rectifier diodes are not conducting yet so it does not matter.
However, the fast transient can cause a filter to ring and create higher frequency spikes.
These too don't make it in the signal with the 3 tested devices but could in some cases.

 

[solderdude]

How do you elimitate the noise caused by pluging to mains a 30W resistive solder iron?

Like I said:

First try to eliminate at the source.
When that isn't possible try to find out where the nasties get into the audio chain.

But let's get into this.
A soldering iron as in a 110V or 230V direct heated one or a soldering station with a transformer or SMPS inside that is switched on using the mains switch or by plugging in the soldering station in the wall with the mains switch on ?

 

[b4nt]

Amir's measurements show that the 'dimmer voltage' which is fed to the power supply (so not a dimmers spectral garbage from a dimmer in the house) with its reasonably fast but not MHz fast and fairly low transients that are not on top of the AC signal is kind of a valid test signal.

Amir distorded a sine wave. Feed a SMPS AC/DC with square, the SMPS wil rectify, filter, convert to DC, as usual. You won't get any additional gremlins in your audio.

The dimmer not exactly "feeding" so the mains, unless you have very poor mains cabling at home.

 

[b4nt]

But let's get into this.
A soldering iron as in a 110V or 230V direct heated one or a soldering station with a transformer or SMPS inside that is switched on using the mains switch or by plugging in the soldering station in the wall with the mains switch on ?

I wrote a 30W resistive iron. This comes with a resistance inside, a wire and a mains plug.

 

[solderdude]

Then the solution is simple.
Use a switch in series in series with the iron, and parallel a capacitor or TVS/MOV in parallel leave the mains side connected.

If that doesn't solve it you really need to look at your audio gear and isolate where it enters the chain.
I hear no ticks in any of my stereo systems when something in the house is switched on.
In the past I have been plagued by this though.
Sometimes it helps swapping L and N on equipment (in my country you can plug in mains 'upside down' even when it has safety earth.
This sometimes helps as the ticks are often common mode. All one needs to do is get it below levels where AM detection in a circuit takes place.

 

[b4nt]

Use a switch in series in series and parallel a capacitor or TVS/MOV in parallel.

Why exactly shoud I add all that? Cause pluging my iron to mains causes transients there?

 

[David_M]

Shunyata has a medical division which is used for cardiac electrophysiology. This is where cardiologists figure out where to ablate (burn) portions of heart tissue to correct irregular rhythms.

https://clearimagescientific.com/case-studies/

Cardiologists are making life changing decisions based upon improvements made by power cables.

It seems most of the signals they deal with are in the
0.050 - 0.250mV
Range

Nice!!!

So there ARE power cables that do actually reduce noise

 

[kopczas]

My air conditioner affects my equipment. There`s a buzz in AMT on my Adam A77x speakers the AC is running. NO buzz on headphones though.
Both are driven by A90.
So I guess that in high gain amp distortion from AC power would be hearable, and should be seen on if someone would measure tweeter output of the amp. (A77x uses 3 amps, 1 per driver).

 

[solderdude]

Amir distorded a sine wave. Feed a SMPS AC/DC with square, the SMPS wil rectify, filter, convert to DC, as usual. You won't get any additional gremlins in your audio.

That's what I said. I also said that when you feed mains with step-alike distortions with SOME equipment this can cause a filter in equipment to ring and create much higher frequency signals that could make it into the audio path. That did not happen with the 3 devices Amir tested.
That doesn't mean that it cannot happen and it also doesn't mean an extra line filter will help.

 

[Johannes AU]

I wrote a 30W resistive iron. This comes with a resistance inside, a wire and a mains plug.

So it should be the same with a 30W incandescent light bulb switched on?

 

[solderdude]

Nice!!!

So there ARE power cables that do actually reduce noise

Yes, they are screened cables and go in hospital grade power sockets and are usually not fed from your regular mains.
Those cables can also have EMC filters inside but have to have low leakage currents.

 

[b4nt]

@amirm @solderdude @pma I have no dimmer here, found this on Internet. I'd suspect a dimmer to alter mains such a way:

https://e2e.ti.com/resized-image/__...372.E27_5F00_230VAC_5F00_trafo_5F00_noise.jpg

 

[Labjr]

My air conditioner affects my equipment. There`s a buzz in AMT on my Adam A77x speakers the AC is running. NO buzz on headphones though.
Both are driven by A90.
So I guess that in high gain amp distortion from AC power would be hearable, and should be seen on if someone would measure tweeter output of the amp. (A77x uses 3 amps, 1 per driver).

You using the XLR input on the speakers?

 

[solderdude]

Do NOT connect audio devices to an old fashioned phase cutting cheap dimmer.
IF you still plan to do that as a test (no idea why you would want to) parallel at least a 100W lightbulb or heater to the equipment you want to 'test' this way.
That's what the B&K simulates.