# Floating Voltage Source Referance to Ground

#### alexfrey

Joined Feb 23, 2019
22
I saw this comment about floating source “So if the source is completely floating, the voltage between the ground of the instrument and the input signals may be randomly high”
What I don’t understand here how a voltage source which has no referance to ground can be randomly high. There shouldn’t be any value ?
For example thermocouple wired to differential analog input. İt is said if it nongrounded thermocouple we should be careful about common mode voltage which means voltage difference thermocouple and differential instrument ground. But I think the same with above there shouldn’t be any current from thermocouple to differential instrument since no referance exist between them. What is the point I miss here ?

#### ian field

Joined Oct 27, 2012
6,539
I saw this comment about floating source “So if the source is completely floating, the voltage between the ground of the instrument and the input signals may be randomly high”
What I don’t understand here how a voltage source which has no referance to ground can be randomly high. There shouldn’t be any value ?
For example thermocouple wired to differential analog input. İt is said if it nongrounded thermocouple we should be careful about common mode voltage which means voltage difference thermocouple and differential instrument ground. But I think the same with above there shouldn’t be any current from thermocouple to differential instrument since no referance exist between them. What is the point I miss here ?
mostly; the obvious point! The floating "whatever it is" completes a circuit round which current can flow if there's voltage to shift it.

"floating2 means no part of that circuit is grounded.

#### crutschow

Joined Mar 14, 2008
26,991
If a source is completely floating, then It can acquire a static charge that can place it at an arbitrary voltage above ground.
That's why the neutral of power lines are grounded, to avoid a static charge built-up.

#### alexfrey

Joined Feb 23, 2019
22
If a source is completely floating, then It can acquire a static charge that can place it at an arbitrary voltage above ground.
That's why the neutral of power lines are grounded, to avoid a static charge built-up.
Hi crutschow,

What is your comment about common mode voltage issue ? for example wiring a thermocouple , do we care common mode voltage because of static charge built-up even thermocouple is completely floating ?

#### KeepItSimpleStupid

Joined Mar 4, 2014
4,633
Nothing truly floats except in theory. Usually, there will be specs with an instrument as to how far above or below ground the instrument can be floated.

Common mode. is the voltage common to both terminals.

#### crutschow

Joined Mar 14, 2008
26,991
In practice, any differential input amp needs some path to ground to provide a path for the input bias current and keep the common-mode voltage within the amp' s limit.

#### MisterBill2

Joined Jan 23, 2018
8,440
KSS is right, in post #5. "floating" is relative. I did experience that exact problem with a sensor connected to a differential input of high impedance and an input range of +/- 10 volts. The input was going into over-range on a routine basis, but connecting a voltmeter showed no problem. Of course, connecting a model 260 always solved the problem. The fix was a 1 meg resistor from the negative input to the input common terminal. Bias currents would otherwise make both inputs drift up to +12 volts.

#### MisterBill2

Joined Jan 23, 2018
8,440
Interesting comment on an object in space experiencing differential charge,
like our floating circuit, due to material differences, and causing failure.

https://sciences.ucf.edu/news/solving-static-electricity-space/

Regards, Dana.
It seems that it has been found that "the empty void of space" is not really totally empty either. The particles are just much farther apart, and so they can carry charges from far away. Who would have guessed that?

#### KeepItSimpleStupid

Joined Mar 4, 2014
4,633
I did experience that exact problem with a sensor connected to a differential input of high impedance and an input range of +/- 10 volts.
Truely differential is hard to come by, but a DVM does it.

With data acquisition, you run into the term pseudo-differential. Your reading is A-B, but two readings are taken A-REF and B-REF where REF is usually Earth.

Op-Amp differential inputs require a bias return path to ground.

I might have a line powered voltmeter that says negative can be up to 30V from ground. Something inside may not survive higher voltages.

In data-aquisition, it's nice to be able to measure an analog output using pseudo-differential techniques, but having an isolated or non-isolated 0-20 mA current source for an analog out is the companion for the setpoint signal.

Joined Jul 18, 2013
22,839
Floating Voltage source Reference to Ground
As usual, there is confusion introduced by the using Ground in a universal manner without distinction of whether this is power common or earth ground, or Both!
If 'Common' or power common were indicated separately to earth it would make it the much clearer to understand.
Max.

#### ian field

Joined Oct 27, 2012
6,539
As usual, there is confusion introduced by the using Ground in a universal manner without distinction of whether this is power common or earth ground, or Both!
If 'Common' or power common were indicated separately to earth it would make it the much clearer to understand.
Max.
It seems fairly unambiguous on a double insulated USB charger wall wart - they frequently have a screened cable connected to the negative supply line and connector shell. UK 13A plug types frequently have a plastic earth pin that serves only to open the safety shutters - making the "floating" bit also pretty unambiguous.

#### MisterBill2

Joined Jan 23, 2018
8,440
To get a reasonably assured voltage source you can use a medical grade isolation power supply. They must meet some fairly tight standards both for leakage and also reliability. Of course the effort required depends a lot on exactly how much isolation and floating you really need. A lot of sound equipment is floated because of the very high noise fields from their stage lighting equipment.. But it is also balanced, which is why the floating is able to keep that noise out..

#### ian field

Joined Oct 27, 2012
6,539
To get a reasonably assured voltage source you can use a medical grade isolation power supply. They must meet some fairly tight standards both for leakage and also reliability. Of course the effort required depends a lot on exactly how much isolation and floating you really need. A lot of sound equipment is floated because of the very high noise fields from their stage lighting equipment.. But it is also balanced, which is why the floating is able to keep that noise out..
Some musicians put an inverse parallel pair of power rectifier diodes in the earth lead to kill hum loops - but I'm told this is very naughty...……..

#### Wolframore

Joined Jan 21, 2019
2,257
does the diodes work??? I remember getting zapped by one amp on my mic vs another on my guitar. I believe the offending amp had a ground lift not sure if it was activated or not.

balanced cables phase cancel noise/hum... this works very well.

#### ian field

Joined Oct 27, 2012
6,539
does the diodes work??? I remember getting zapped by one amp on my mic vs another on my guitar. I believe the offending amp had a ground lift not sure if it was activated or not.

balanced cables phase cancel noise/hum... this works very well.
0.7V won't zap you - but apparently its enough to lift out of a hum loop.

use a pair of 1N5401 rectifiers and they'll probably fail short circuit if there's an earth fault - but I've seen dodgy knock off 3A rectifiers break in 2 in a certain cheap Chinese VGA monitor.

#### crutschow

Joined Mar 14, 2008
26,991
use a pair of 1N5401 rectifiers and they'll probably fail short circuit if there's an earth fault - but I've seen dodgy knock off 3A rectifiers break in 2 in a certain cheap Chinese VGA monitor.
Yes, I wouldn't want to bet my life on diode failing short instead of open for a high current ground fault.
High ground fault currents will likely cause a diode to blow open like a fuse, unless you maybe use 100A diodes so that the mains fuse blows before the diode.
Technically you would likely want the I²t rating of the diode to be greater than that of the fuse.

Last edited:

#### Wolframore

Joined Jan 21, 2019
2,257
Ahh gotcha... Those diode clamps are semi dangerous.

#### ian field

Joined Oct 27, 2012
6,539
Ahh gotcha... Those diode clamps are semi dangerous.
A lot of TO220 style rectifiers have an instantaneous surge rating greater than the maximum recommended fuse rating. They might not survive as rectifiers - but they'd be unlikely to blow open. The 3A ones such as the 5401 etc are the least I'd take a chance on - and make damn sure not to use cheap dodgy knock offs.

#### MisterBill2

Joined Jan 23, 2018
8,440
Some musicians put an inverse parallel pair of power rectifier diodes in the earth lead to kill hum loops - but I'm told this is very naughty...……..
I am aware of at least one sound system installation with the "green wire" ground totally disconnected because the hum is a bit loud when it is connected. That is because power to a bank of triac light dimmers shares the same conduit and the same wired ground circuit.