Strange audio interference on shared power supply (video)

Jump to Latest
J

jb91

Member
Joined 2020
#1
Hi, I was pointed to you guys from another forum. I am hoping you might be able to give some advice on how I might fix this audio interference issue I am getting please? see this video.

I have the components set up as follows:

-A Meanwell RPS-400-18 Power supply connected to a IEC inlet filter powers both:

-Zotac EN72070V PC

- Sure Electronics AA-AB32996 Amplifier (powered through the screw down terminals)

-The audio is carried from the PC to the amplifier via a 3.5mm Jack to RCA cable.


Edit: I bought a cheap ground loop noise isolator from Amazon to test and the problem has completely gone away. I'm not sure how this will affect sound quality- apparently they can cause low frequency issues- however that's another problem to tackle as I build the speaker/s into the sheet metal enclosure. Thanks everyone for the input.

-The Amplifier powers 2x Dayton PS95-8 Speakers via speaker cables.

When setup this way I get a lot of interference (coming out of the speakers) which appears to be from the PC (when I wiggle the jack connector, the audio interference changes). I have tried connecting the 3.5mm jack to a USB port using an Apple 3.5mm to USB-C Adapter (which contains a small DAC) but this also didn’t help.

If I connect either the PC or the amplifier to a separate power supply (so one running from the Meanwell PSU above and the other to a wall plug adapter) the issue (almost) disappears.

Any advice or things to try would be greatly appreciated, very stuck currently! Thanks

EDIT: I bought a cheap ground loop noise isolator from Amazon to test and the problem has completely gone away. I'm not sure how this will affect sound quality- apparently they can cause low frequency issues- however that's another problem to tackle as I build the speaker/s into the sheet metal enclosure! Thanks everyone for your input.
 
Last edited:
#3
Quote-- "when I wiggle it the audio interference changes "-- did the other forum tell you to wiggle it ?


The female 3.5mm socket does not stand up to much abuse and the web is full of ---no sound/cutting off etc when jacks are plugged in .


I have replaced many with better quality ones ,so the question is before many remedies are given when it changes do you mean it GOES AWAY ?
 
#4
Congratulations OP, you are about the gazillionth to have encountered a ground loop when using an unbalanced audio connection between two devices sharing a common ground. An eternal classic e.g. among mobile boombox builds.

As you have discovered, resolving this requires galvanic isolation - in this case provided by the two separate power supplies. Other possibilities of providing it include:
* a USB isolator (higher-end DACs also tend to include one) or
* a line-level isolator

Both have to be chosen carefully. The latter may result in rather poor frequency response and/or distortion performance, the former needs to be checked for supported data rates (some may be restricted to USB 1.1 speeds) and output current.

If using multiple power supplies is not an issue, I would suggest going that route, it is generally the cheapest and easiest option. In addition, I would also look for an audio cable with lower shield resistance, or bond amplifier ground to audio output ground using a decently heavy-gauge wire.
 
#5
Looking at the Meanwell data sheet it's not obvious where the AC safety ground is connected but it seems to be called "FG". It's important that any switch mode power supply be well grounded because safety ground is also the primary circuit noise sink.
Having said that, a direct connection between the safety (input) ground and the DC output ground is also asking for noise problems, aka the classic ground loop. The block diagram shows a (hopefully small) capacitor between "FG" (frame ground?) and the output ground. This should not be a problem but it could be. Just be sure there is no direct connection between output ground and "FG".
The biggest problem might be that the "FG" is not grounded, in which case Y-capacitors in the input filter creates half the line voltage on the "FG". Be sure "FG" is well connected to the 3rd pin (ground) on the ac cord.
 
Last edited:
#6
duncan2 said:
Quote-- "when I wiggle it the audio interference changes "-- did the other forum tell you to wiggle it ?


The female 3.5mm socket does not stand up to much abuse and the web is full of ---no sound/cutting off etc when jacks are plugged in .


I have replaced many with better quality ones ,so the question is before many remedies are given when it changes do you mean it GOES AWAY ?
Click to expand...

Ideally I don't want to use the 3.5mm jack at all as it isn't located in a good place on the computer for the way it is mounted. for this reason I am planning to use the USB-3.5mm adapter . By changes- it just changes the pitch etc. of the noise- i.e. you can tell the 'source' of the interference is coming from the computer as opposed to it being generated on the amp for example.
 
#8
steveu said:
Looking at the Meanwell data sheet it's not obvious where the AC safety ground is connected but it seems to be called "FG". It's important that any switch mode power supply be well grounded because safety ground is also the primary circuit noise sink.
Having said that, a direct connection between the safety (input) ground and the DC output ground is also asking for noise problems, aka the classic ground loop. The block diagram shows a (hopefully small) capacitor between "FG" (frame ground?) and the output ground. This should not be a problem but it could be. Just be sure there is no direct connection between output ground and "FG".
The biggest problem might be that the "FG" is not grounded, in which case Y-capacitors in the input filter creates half the line voltage on the "FG". Be sure "FG" is well connected to the 3rd pin (ground) on the ac cord.
Click to expand...

Thanks- I have just checked and there is continuity between FG (frame ground) and the IEC ground connector so I do have a good connection between these. I have no continuity between -ve and FG when I check continuity. Some FAQs say the PSU is built as an isolated supply.
 
#10
sgrossklass said:
Congratulations OP, you are about the gazillionth to have encountered a ground loop when using an unbalanced audio connection between two devices sharing a common ground. An eternal classic e.g. among mobile boombox builds.

As you have discovered, resolving this requires galvanic isolation - in this case provided by the two separate power supplies. Other possibilities of providing it include:
* a USB isolator (higher-end DACs also tend to include one) or
* a line-level isolator

Both have to be chosen carefully. The latter may result in rather poor frequency response and/or distortion performance, the former needs to be checked for supported data rates (some may be restricted to USB 1.1 speeds) and output current.

If using multiple power supplies is not an issue, I would suggest going that route, it is generally the cheapest and easiest option. In addition, I would also look for an audio cable with lower shield resistance, or bond amplifier ground to audio output ground using a decently heavy-gauge wire.
Click to expand...

Thanks a lot, I have just done some research based on what you said and youtube videos etc. of people with the same issue seems like it could well be that! I just didn't know what it was called. Does that mean something as simple as this
 or this would fix the issue?

I did read that the differential input connector that is on the amplifier can have a positive impact on eliminating noise but I can't seem to find such a cable for this? Is it naive to assume there is some kind of USB/3.5mm jack to differential connector type cable?
 
Last edited:
#11
Thanks, that's really helpful but a lot of it goes well over my head! I have just checked continuity between the two negative RCA inputs and there is continuity, in addition there is continuity between the negative RCA inputs and the -ve terminal at the power input for the amplifier.
 
#12
What is the resistance between the negatives and the safety ground on your power supply?


But please don't make any changes to your safety ground. The best thing to do is to live with using a second power supply to minimize the noise, as others have suggested. The case/safety ground should not be messed with. At least you have a reasonable solution.
 
Last edited:
#13
pinholer said:
What is the resistance between the negatives and the safety ground on your power supply?

I don't get a reading- should I?

But please don't make any changes to your safety ground. The best thing to do is to live with using a second power supply to minimize the noise, as others have suggested. The case/safety ground should not be messed with. At least you have a reasonable solution.
Click to expand...

That would be easiest but unfortunately isn't an option in this case because of space restrictions etc. I need to try and find a solution to getting them both running on the same supply. Perhaps a galvanic isolator will do the trick?
 
#14
sgrossklass said:
Congratulations OP, you are about the gazillionth to have encountered a ground loop when using an unbalanced audio connection between two devices sharing a common ground. An eternal classic e.g. among mobile boombox builds.

As you have discovered, resolving this requires galvanic isolation - in this case provided by the two separate power supplies. Other possibilities of providing it include:
* a USB isolator (higher-end DACs also tend to include one) or
* a line-level isolator

Both have to be chosen carefully. The latter may result in rather poor frequency response and/or distortion performance, the former needs to be checked for supported data rates (some may be restricted to USB 1.1 speeds) and output current.

If using multiple power supplies is not an issue, I would suggest going that route, it is generally the cheapest and easiest option. In addition, I would also look for an audio cable with lower shield resistance, or bond amplifier ground to audio output ground using a decently heavy-gauge wire.
Click to expand...

I bought a cheap ground loop noise isolator from Amazon to test and the problem has completely gone away. I'm not sure how this will affect sound quality- apparently they can cause low frequency issues- however that's another problem to tackle as I build the speaker/s into the sheet metal enclosure! Thanks for the help
 
You must log in or register to reply here.
Top Bottom