Any issues not using ground lead on potentiometer?

[tomzarbo]

Re: Any issues not using ground lead on potentiometer?

When you wrap in foil, are you attaching the foil to a good ground? If you're just wrapping the foil around your wires and not attaching it to anything, then you may as well wrap it in a piece of paper. If the foil is picking up any noise, it needs a ground path to shunt the noise to.

Hmm, can you remove the heatsink and see what the amp chip(s) is/are, or is it epoxied in place? You have me very curious now as to what this little amplifier really is. Especially with the power numbers they claim, and the difference between the L/R channels and the sub channel.

Okay, so my little 'shielding' experiment didn't have a chance.

The heatsink is glued on there, but Rory in his statement on the Q&A section said that the markings on the chips were 'obliterated' before gluing on the heatsink, so I guess we're outta luck there.

I just looked at the amp board again, and it appears this is an inductor-less class-D design. That design will generate a lot of stray noise, and it could just be inherent to the design. If designed properly, that noise should be out of the audible band, but you never know.

I'm beginning to think that this little board may have problems.

FWIW I always ground the potentiometer's case. I had a noisy tube amp project and grounding the case cured it. Now I just make it standard practice.

Okay, that makes sense, what should I ground it to, the negative terminal on the power supply incoming? I can drill/tap a screw on the alum. plate to ground things to if it helps, no problem there. I would ground the pot body and what else would you suggest grounding? The amp board is only touching the alum. plate where the main volume potentiometer is bolted to it, but the board itself is basically right next to the alum. plate. The red piece of electrical tape is there to prevent any direct contact between the board and the alum.

If I wanted to 'ground' the amp to the board, how would I even do it? I don't see metal showing on any of the mounting holes on the amp board. The only place I can think of to ground it would be on the negative terminal of the power coming in.
I'm assuming that the main volume pot mounted on the amp board although it is metal, would not actually be conducting anything to anything else.

So again, it works fine with a jumper in place, no real noise. The only noise is when I try to wire in a potentiometer... and again to reiterate, the same thing is happening with both the bare naked board and the one mounted to the alum. plate.

I'm a bit bummed, this little board is quite nice otherwise. But without a way to adjust the sub volume, it's not going to fly in my book. Wonder why they just didn't build it with the sub pot built in?

TomZ

[1100xxben]

Re: Any issues not using ground lead on potentiometer?

Okay, that makes sense, what should I ground it to, the negative terminal on the power supply incoming?

Yes, without knowing any other details of the board, that would be the safest place to tie the plate to.

...Wonder why they just didn't build it with the sub pot built in?

Maybe because it's noisy?? ;)

[craigk]

Re: Any issues not using ground lead on potentiometer?

since nothing is working try star grounding.

[tomzarbo]

Re: Any issues not using ground lead on potentiometer?

since nothing is working try star grounding.

So one good connection point on the aluminum plate which I can tie a grounding terminal to from every component until the noise ceases?

For instance, I'd drill a hole and tap a screw through the aluminum plate, then secure it on the back side with a nut, then proceed to 'ground' things... the sub pot, negative side of the power supply input, negative, on the amp board, etc. until the noise stops.... if I'm understanding the concept correctly. So I pretty much need to do this on a metal plate and not have stuff just sitting on my workbench for this to work.

Thank you for the suggestion Craig. I'll give it a shot after I mow the grass tonight... if I can still see and breathe... allergies are killing me this year.

TomZ

[craigk]

Re: Any issues not using ground lead on potentiometer?

So one good connection point on the aluminum plate which I can tie a grounding terminal to from every component until the noise ceases?

For instance, I'd drill a hole and tap a screw through the aluminum plate, then secure it on the back side with a nut, then proceed to 'ground' things... the sub pot, negative side of the power supply input, negative, on the amp board, etc. until the noise stops.... if I'm understanding the concept correctly. So I pretty much need to do this on a metal plate and not have stuff just sitting on my workbench for this to work.

Thank you for the suggestion Craig. I'll give it a shot after I mow the grass tonight... if I can still see and breathe... allergies are killing me this year.

TomZ

you have the correct idea. I hope it works.

[1100xxben]

Re: Any issues not using ground lead on potentiometer?

I would ohm everything out first. If you already have connection between your input power ground and the ground side of your potentiometer for example, then all you'd be doing with the additional star grounding is creating a ground loop, which is not a good idea.

[craigk]

Re: Any issues not using ground lead on potentiometer?

I would ohm everything out first. If you already have connection between your input power ground and the ground side of your potentiometer for example, then all you'd be doing with the additional star grounding is creating a ground loop, which is not a good idea.

this is not correct. when two or more devices are connected to a common ground through different paths, a ground loop occurs. Currents flow through these multiple paths and develop voltages which can cause damage, noise or 50Hz/60Hz hum in audio or video equipment. To prevent ground loops, all signal grounds need to go to one common point and when two grounding points cannot be avoided, one side must isolate the signal and grounds from the other. all grounds will being going to one common point with star grounding and there is not chance for a loop to occur.

[tomzarbo]

Re: Any issues not using ground lead on potentiometer?

Okay, I think we may be getting somewhere...

I soldered two 5.1k resistors to the center terminal of the sub pot, and then the other ends of the resistors to the outer two terminals on the sub pot. If I understand correctly this simulates a 10k pot.

Well, that did a lot to eliminate some noise. Grounding the metal body of the subwoofer pot to the negative side of the 12v input from the power supply did the rest. It's pretty quiet now, I can live with this for sure. You can hear the whining sound faintly if you put your head right in front of the subwoofer, but when music is playing at any real volume, it all but disappears. I think this will work.

Funny, I was just about to post that adding the resistors didn't work -- I was typing it on the computer... then I saw that I had written down 5k on the little diagram I made. By mistake I soldered two 500k resistors, which did basically nothing at all. Thank God I caught that.

So right now, it looks like a solution will be to order up a few 10k pots and ground the sub pot to the 12v negative on the amp board.

PE sells two 10k pots. This one:

http://www.parts-express.com/10k-ohm...meter--023-530

and this one:

http://www.parts-express.com/10k-aud...shaft--023-610

I'll order a few of both and see which works better/quieter.

Any ideas on the best way to solder to the back of a pot body? Sand, flux, and lots of heat?

Thanks so much for the good advice. As soon as I get the parts I'll try the new pot on the board mounted on the alum panel and hopefully I'll be in business.

TomZ

[1100xxben]

Re: Any issues not using ground lead on potentiometer?

this is not correct. when two or more devices are connected to a common ground through different paths, a ground loop occurs. Currents flow through these multiple paths and develop voltages which can cause damage, noise or 50Hz/60Hz hum in audio or video equipment. To prevent ground loops, all signal grounds need to go to one common point and when two grounding points cannot be avoided, one side must isolate the signal and grounds from the other. all grounds will being going to one common point with star grounding and there is not chance for a loop to occur.

If you're going to start randomly grabbing "ground" points of a circuit and connect them to another common point (star ground), then you first need to check continuity between them and verify they are not already connected. If they are already connected and you pull off your own "ground" wires to another common point, you will quite possibly end up with ground loops.

Ground loops do not always occur between two different "devices". Ground loops can easily occur within a single piece of electronics, and even within a single circuit board if the grounding scheme is not designed properly. Ground loops through the ground wire in your house will generally cause a 50/60 Hz issue, but they can easily occur at any frequency when you're dealing with inter-device ground loops.

Here is a rough picture of how I'm guessing the circuit (the relevant portion of the circuit) is laid out, and why tying the input power ground to the aluminum plate AND the "potentiometer ground" to the aluminum plate is a bad idea. Adding both of the red wires will create a ground loop. Now, tying the case of the pot to a ground potential at some point should be just fine, as long as it's completely isolated from any other ground point.

[1100xxben]

Re: Any issues not using ground lead on potentiometer?

... the sub pot, negative side of the power supply input, negative, on the amp board, etc...

Just to clarify something here... When you say "negative, on the amp board", are you referring to the negative speaker output? If so, you should never tie the speaker output negative to anything unless you know for sure exactly what type of amplifier output stage you have. I would bet money that these boards are a bridged output, which means the negative speaker output is being driven and is anything but ground.

[tomzarbo]

Re: Any issues not using ground lead on potentiometer?

Just to clarify something here... When you say "negative, on the amp board", are you referring to the negative speaker output? If so, you should never tie the speaker output negative to anything unless you know for sure exactly what type of amplifier output stage you have. I would bet money that these boards are a bridged output, which means the negative speaker output is being driven and is anything but ground.

What I meant by "negative on the amp board" was where the (-) side of the 12v in from the power supply comes in. I did not do any grounding to the speaker outputs.
There doesn't seem to be any place on the amp board to 'ground' anything. I've seen plenty of amp boards where the mounting holes on an amp board will have an uncoated portion around the hole that has exposed metal that is meant to be a 'grounding' point, but this amp board doesn't have that... so I picked the 12v (-) point to ground the amp board to.

Attempting to ground the main volume on-board pot to ground did nothing to reduce the noise.

The tweaks I tried last night were on the raw board... I have not yet done this to the version that is mounted on the alum. plate. It was getting late and I already goofed up on the resistor values... I didn't want to mess something else up potentially worse.



I'm having two of each of the above listed 10k pots overnighted and as soon as I get them, I'll take the 50k pot out of the alum. plate project and replace with the 10k, as well as ground that pot to the 12v negative terminal on the amp board. Hopefully there will be a similar noise reduction.

I have another amp board coming and will construct another plate amp on another alum. plate I have with the new 10k pot. If both re-do's are good, then I'll believe we have the problem licked. Fingers crossed...

Thanks you guys for the guidance and help.
I'll be sure to post up what happens when I get the parts in.

Gracias,

TomZ

[craigk]

Re: Any issues not using ground lead on potentiometer?

If you're going to start randomly grabbing "ground" points of a circuit and connect them to another common point (star ground), then you first need to check continuity between them and verify they are not already connected. If they are already connected and you pull off your own "ground" wires to another common point, you will quite possibly end up with ground loops.

Ground loops do not always occur between two different "devices". Ground loops can easily occur within a single piece of electronics, and even within a single circuit board if the grounding scheme is not designed properly. Ground loops through the ground wire in your house will generally cause a 50/60 Hz issue, but they can easily occur at any frequency when you're dealing with inter-device ground loops.

Here is a rough picture of how I'm guessing the circuit (the relevant portion of the circuit) is laid out, and why tying the input power ground to the aluminum plate AND the "potentiometer ground" to the aluminum plate is a bad idea. Adding both of the red wires will create a ground loop. Now, tying the case of the pot to a ground potential at some point should be just fine, as long as it's completely isolated from any other ground point.
[ATTACH=CONFIG]57496[/ATTACH]

your diagram is not a correct representation of what a star ground is.

[[email protected]]

Re: Any issues not using ground lead on potentiometer?

Why not just get more resistors and forget the pot all together.
No star ground is possible as no ground wire goes to the amp. Its a wall wart with just plus and minus.

[1100xxben]

Re: Any issues not using ground lead on potentiometer?

your diagram is not a correct representation of what a star ground is.

This is exactly my point...

You suggested to try a star grounding scheme:

since nothing is working try star grounding.

Tom suggested what I drew in my diagram above:

So one good connection point on the aluminum plate which I can tie a grounding terminal to from every component until the noise ceases?

For instance, I'd drill a hole and tap a screw through the aluminum plate, then secure it on the back side with a nut, then proceed to 'ground' things... the sub pot, negative side of the power supply input, negative, on the amp board, etc. until the noise stops....

You informed him that his idea of creating a start ground was correct:

you have the correct idea. I hope it works.

I was simply trying to show that this method is not truly creating a star grounding scheme and would most likely create a ground loop.

[tomzarbo]

Re: Any issues not using ground lead on potentiometer?

Why not just get more resistors and forget the pot all together.

Because if I don't actually physically use the 10k pot I won't be able to feel 100% sure that it will work as expected. Judging from this thread I suspect there may be others who would like to duplicate this.

http://techtalk.parts-express.com/sh...ighlight=micro

If I can be sure the noise is gone for sure this will make a really nice micro-sized setup for most of the small speakers/subwoofers that are available. Even though 8 ohm drivers are suggested I was using it with a set of Neo Nanos and a Voxel sub... both 4 ohms if I'm not mistaken... and the heat sink never even got warm, even at fairly loud volumes, playing for several minutes. This was with both a 12v and 16v power supply, I think 18v is the max recommended.

At this point, I'm going to try just grounding the sub volume pot to the negative 12v on the amp board. If that does it, then I won't fiddle with any more grounding at that point.

TomZ