Changing ground reference of general electronic devices.

WBahn

Joined Mar 31, 2012
29,979
Read again what I said about the assumption that the neutral and the safety ground are always at or very near the same potential. The contents of that box was designed with that assumption.

If you break the neutral connection and then apply -220 V to it (assuming relative to earth ground), then how is the power that is coming in on the hot wire going to make it back to the transformer?

Sketching out how you are talking about hooking up these boxes to power would sure help the discussion a lot!
 

Thread Starter

coinmaster

Joined Dec 24, 2015
502
If you break the neutral connection and then apply -220 V to it (assuming relative to earth ground), then how is the power that is coming in on the hot wire going to make it back to the transformer?
Hmmmm, wouldn't it just go back to the wall through the -220v power supply?

Untitled.png
 

MrSoftware

Joined Oct 29, 2013
2,188
Think of it like this. Voltages are relative. If neutral is about 0v, and hot varies between about -156v and +156v relative neutral in a sine wave (for 110v RMS), then the AC/DC converter inside is designed to deal with peak voltages of around 156v. Assume you move neutral to -220v DC. Now the hot wire will give you a sine wave between +64v and +376v relative neutral. While it's likely the AC/DC converter can handle voltages above a perfect 110v RMS (156v peak), I doubt the AC/DC converter inside would do well with a potential difference of 376v. That might not be the only problem, but it's the first one that jumped out at me.
 

Thread Starter

coinmaster

Joined Dec 24, 2015
502
I thought as long as the ground references is shifted it doesn't know the difference?
+64v and +376v would only be relative to normal ground.
The input from the hot pin is AC and I thought AC latches on to whatever the current ground reference is?

Oh wait, I get it, I think.
I would have to adjust the hot reference as well right?
 
Last edited:

MrSoftware

Joined Oct 29, 2013
2,188
In the USA, ground and neutral are at the same potential (0v). It is actually connected to an earth anchor, so call it earth for now. The power company is delivering you a hot wire that is 110v RMS relative earth. You can change the ground and neutral pins in your power plug to be -220v, but the hot wire will still be 110v RMS relative earth (0v), not relative your new -220v.

To make it work the way you are thinking, you would need to first disconnect the ground/neutral from the earth, then float the entire power generation and delivery system to be relative -220v instead of relative earth.
 

Thread Starter

coinmaster

Joined Dec 24, 2015
502
Hmmm, so I would need like a 1:1 transformer and and then throw -220vDC onto the output of the transformer so the AC level is shifted. Does that sound right? I question what will happen if I shoot -220vDC into a transformer, I would probably need a buffer.

Assuming I did this and it worked. Other than touching the output terminals or the circuitry with my hands, would there be any concerns?
 
Last edited:

WBahn

Joined Mar 31, 2012
29,979
Hmmm, so I would need like a 1:1 transformer and and then throw -220vDC onto the output of the transformer so the AC level is shifted. Does that sound right? I question what will happen if I shoot -220vDC into a transformer, I would probably need a buffer.

Assuming I did this and it worked. Other than touching the output terminals or the circuitry with my hands, would there be any concerns?
DC voltage can't make it through a transformer, so that's a non-starter right there.
 

WBahn

Joined Mar 31, 2012
29,979
I thought as long as the ground references is shifted it doesn't know the difference?
+64v and +376v would only be relative to normal ground.
The input from the hot pin is AC and I thought AC latches on to whatever the current ground reference is?
But what is the -220 VDC relative to? (seems like I've asked that question a couple of times already)

It is almost certainly relative to the neutral, which is earth grounded. The hot wire is also relative to earth ground, hence the circuitry in that box is going to see voltage differences between that boxes hot and neutral of those extremes.

Oh wait, I get it, I think.
I would have to adjust the hot reference as well right?
Good luck with that one! The hot reference is established by the power utility and it is physically the earth. The only way you could change that, for your equipment, would be to use an isolation transformer. But don't go rushing out buying isolation transformers. There is no reason to think that the circuitry in that DAC box is going to like any of the games you are considering and improper use of an isolation transformer can get you dead.
 

MrSoftware

Joined Oct 29, 2013
2,188
Firstly, I think you've got the wrong approach. What is the real problem that you're trying to solve? Surely there is a better way.

That said; I'm not the expert at this so maybe someone with more knowledge will chime in, but if you wanted to float your entire box to -220v potential, then drive it with 110v AC from the wall, I think coupling capacitors might be a possible way to do it. I think it's a very very bad idea.
 

Thread Starter

coinmaster

Joined Dec 24, 2015
502
DC voltage can't make it through a transformer, so that's a non-starter right there.
Well I didn't mean I would take -220v from the transformer, I was saying that I would use the transformer to get the AC from the wall and then apply -220v from a separate power supply to the AC from the transformer.

But what is the -220 VDC relative to? (seems like I've asked that question a couple of times already)
I told you a few times :p It's relative to earth ground because the amp I'm trying to couple the DAC to is earth ground referenced.

Good luck with that one! The hot reference is established by the power utility and it is physically the earth. The only way you could change that, for your equipment, would be to use an isolation transformer. But don't go rushing out buying isolation transformers. There is no reason to think that the circuitry in that DAC box is going to like any of the games you are considering and improper use of an isolation transformer can get you dead.
Um, won't my above plan work where I use a 1:1 transformer and apply the bias voltage to the other side?

There is no reason to think that the circuitry in that DAC box is going to like any of the games you are considering
You keep saying that but I don't understand it. If I properly shift the ground potential of the DAC wouldn't it not know any different whether it was 0v referenced for 1million volt referenced?
 

Thread Starter

coinmaster

Joined Dec 24, 2015
502
Firstly, I think you've got the wrong approach. What is the real problem that you're trying to solve? Surely there is a better way.

That said; I'm not the expert at this so maybe someone with more knowledge will chime in, but if you wanted to float your entire box to -220v potential, then drive it with 110v AC from the wall, I think coupling capacitors might be a possible way to do it. I think it's a very very bad idea.
Of course it's a bad idea. But it's the best bad idea I got :p
The problem is the simple desire to avoid capacitors and transformers. I've had very positive experiences removing these things from the signal path.
The only way I see direct coupling a dac to a -220v bias point is by floating the dac.
 

Papabravo

Joined Feb 24, 2006
21,159
:confused: What DAC have you got that plugs directly into a wall socket?
The DAC he is referring to is a complete audio component. It takes digital input from one or more sources and produces analog output directly to a power amplifier or to a speaker.
 

Alec_t

Joined Sep 17, 2013
14,280
Right. Your DAC box presumably includes a transformer, operating at either mains frequency or an SMPS switching frequency, for obtaining its power from the mains. This transformer should ensure your DAC has its output galvanically isolated from mains ground (neutral). However, if you now reference that output to -220VDC you will be placing an extra 220V stress on the transformer insulation. It might not like that :(. If the insulation were to break down your DAC and amp could become live and potentially (pun intended) lethal.
 

Thread Starter

coinmaster

Joined Dec 24, 2015
502
It uses a toroidal transformer. The dac itself would be referenced to a ground of some kind, obviously it has to be. probably earth ground referenced. If I changed the ground to -220v it should not know the difference no matter what the voltage is right?
You guys are confusing me with all this "220v of stress", it's only 220v of stress relative to earth ground, if I change the ground of the DAC to -220v it would be 0v of stress.
 

WBahn

Joined Mar 31, 2012
29,979
Again, you are asking things that can only be answered by looking at the circuitry inside the DAC box (not a picture of it, the schematics for it). There's no way we can give a definitive answer without that information.
 

Thread Starter

coinmaster

Joined Dec 24, 2015
502
What I'm asking is, what possible ways could this be an issue for the internal circuitry no matter what the design is?
If I change ground reference, the internal circuitry should be none the wiser no matter what the reference voltage is.
Lets say it runs on +24v.
By default it is +24v above earth ground.
If I change the ground reference to 1 million volts above earth ground it would still be running at +24v as far as the dac is concerned.
 

Thread Starter

coinmaster

Joined Dec 24, 2015
502
What is the -220V referenced to?
I keep answering this question :p
It's referenced to earth ground.

Maybe I need to explain what I'm trying to do again.

I have a dac.
It uses opamps to null the DC at its output, therefore allowing direct coupling to amplifiers without the use of a capacitor.
The amp I want to couple it to has -220v (below earth ground) bias voltage on the input.

The only way to null DC offset with opamps that I know if is by using an "integrator"
Therefore the opamp will try to make the output voltage equal to ground.
However if I change "ground" to -220v then the opamp will now try to make the output -220v but since I changed the entire ground reference for the entire circuit it would not know any difference from -220v and 0v.

Right?
 
Last edited:
Top