0
\$\begingroup\$

This isnt a naive question - I hope!

In virtually every situation where a signal is sent by wire, it uses (at leasst) two wires - one for some kind of reference voltage, and (at least) one for the data. The signal is amplitude moderated - the difference between the signal wire(s) and reference wire voltages being above, below or between certain voltages.

Naively, if two floating-voltage devices are connected by a single wire, effectively this fixes their relative voltage, and amplitude modulation wont work. They share a common ground or similar. So naively, reading a voltage on one wire can't be done - a voltage has to be between two potentials.

But potential takes time to build, and time to dissipate. It's not ideal and square.

Therefore, is there any effect (impedance? lead/lag/timing behaviours? other?) such that a circuit could receive a single wire from a floating device, and still determine if the voltage at the other end has changed up or down, or by how much, simply by (for example):

  • observing the transient effects as charge equalises at the junction with the wire, and potential settles at a new level; or by
  • comparing the voltage between the incoming wire and some well grounded point internally that can be taken as a reference voltage for the purposes of observing transient behaviours at the wire input? (So it can be treated as "more static in nature" or because it leads/lags the wire potential in behaviour)
  • Observing the potential rise and fall at two points a few MM apart on the conductor, at or near the junction, if the junction was designed to create a measurable lag over some short distance.

Has this been used in any practical form?

CLARIFICATION: To be clear, I'm thinking of pure basic charge-flow-based effects, transient behaviours, and other measurables, within the recipient PCB and its circuitry, or similar. Not radiated effects (electromagnetic radiation), surface-wave effects along the wire, or conversion to and from a field outside the conductors/components. (At least I think that's what I mean!)

\$\endgroup\$
6
  • \$\begingroup\$ If they share a common ground then you have two wires (signal and ground) and can communicate at normal. \$\endgroup\$ Commented Jan 26, 2021 at 17:30
  • 1
    \$\begingroup\$ You can get rid of all the connections and use radio. If you still want a connection then that one connection won't stop radio working. \$\endgroup\$
    – Andy aka
    Commented Jan 26, 2021 at 17:33
  • \$\begingroup\$ I'm thinking of purely charge-based effects, rather than radiated effects (electromagnetic radiation), or conversion to and from a field outside the conductors. \$\endgroup\$
    – Stilez
    Commented Jan 26, 2021 at 17:49
  • \$\begingroup\$ @jsotola - yes and fixed,thank you \$\endgroup\$
    – Stilez
    Commented Jan 26, 2021 at 18:02
  • \$\begingroup\$ If you define charged-based to mean that charge must move from the sender to the receiver, then you might consider something like a pair of particle accelerators firing electrons back and forth (and you would need two since even a small excess of electrons on the receiver will block further transmission), although arguably in this case the air is serving as a second conductor. \$\endgroup\$ Commented Jan 26, 2021 at 18:22

2 Answers 2

1
\$\begingroup\$

In virtually every situation where a signal is sent by wire, it uses (at least) two wires - one for some kind of reference voltage, and (at least) one for the data. The signal is amplitude moderated - the difference between the signal wire(s) and reference wire voltages being above, below or between certain voltages

Well .. sort of. This model breaks down for fast signals over long wires where transmission line behavior takes over. It also doesn't cover the popular 4-20ma current loop signalling system, and things like Ethernet which is isolated by tiny transformers and the ends can float as much as 1500V apart without being out of spec.

There is also radio transmission, which uses electromagnetic field coupling to transfer a signal using zero wires!

But potential takes time to build, and time to dissipate. It's not ideal and square

I think you're heading in the right direction here.

May I present: the Goubau line! This is effectively a radio wave, but the presence of the wire anchors it in space around the wire. Current doesn't actually flow in the wire.

Any form of sharp transient signal that drives a single wire will travel along the wire, behaving like a transmission line with the self-inductance of the wire and its capacitance to ground (even though that may be far away). The signal will tend to bleed off the wire, and may be reflected from the far end. The limit case of a single wire driven by an alternating signal is the classic radio aerial such as you might find on your car.

Some more general ideas are presented here.

\$\endgroup\$
3
  • \$\begingroup\$ See clarification added to OP, if that helps refine the answer? Will come back and see what it says shortly! \$\endgroup\$
    – Stilez
    Commented Jan 26, 2021 at 17:57
  • \$\begingroup\$ @Stilez Hmm. Gold foil electroscope? Single connection, detects net positive or negative charge. \$\endgroup\$
    – pjc50
    Commented Jan 26, 2021 at 22:56
  • \$\begingroup\$ Hmm. A good question. Suppose the other end was truly floating, with no earth loop or stray connection. For.example, you had a floating sine generator in isolation, in a vacuum, similar to what Bruce Abbott suggests, would you cause electrons to move in its outputaor what would hapoen to charges in them? If not, what is the effect of a truly floating sine generator on the charge at its outputs, or the charge movement in a wire attached to them, or the transient effect to the charges in it's outputs or that wire, if it runs isolated in a vacuum or similar? \$\endgroup\$
    – Stilez
    Commented Jan 26, 2021 at 23:17
0
\$\begingroup\$

Naively, if two floating-voltage devices are connected by a single wire, effectively this fixes their relative voltage, and amplitude modulation wont work. They share a common ground or similar. So naively, reading a voltage on one wire can't be done - a voltage has to be between two potentials.

Correct. If the devices are truly floating (not coupled to anything via resistance, capacitance or inductance) they cannot develop a voltage difference.

The following circuits are functionally identical. The 'ground' reference point is completely arbitrary and has no effect. In each case the signal is going to an open circuit - so no current can flow, no voltage can be developed, and no transmission line effects can occur.

schematic

simulate this circuit – Schematic created using CircuitLab

Therefore, is there any effect (impedance? lead/lag/timing behaviours? other?) such that a circuit could receive a single wire from a floating device, and still determine if the voltage at the other end has changed up or down, or by how much, simply by (for example):

  • observing the transient effects as charge equalises at the junction with the wire.

Charge won't 'equalize' because the generator cannot push any current into the wire. Instead, voltage at the other end of the generator will simply go up and down uselessly as it drives an open circuit (refer to circuit A).

  • comparing the voltage between the incoming wire and some well grounded point internally that can be taken as a reference voltage.

You cannot measure voltage on the incoming wire because you don't have a common reference point. Your voltmeter's 'ground' terminal is open circuit (refer circuit B).

  • Observing the potential rise and fall at two points a few MM apart on the conductor, at or near the junction, if the junction was designed to create a measurable lag over some short distance.

In circuit C the junction is 'grounded' so nothing gets past it. Of course this ground is just an arbitrary reference point, but the result is the same no matter where we put it.

However in practice nothing is truly floating. There is capacitance to other objects and real ground, inductive coupling to mains wiring etc., even when the circuit is galvanically isolated. Radio waves are everywhere, the Earth's atmosphere has a strong electrostatic gradient, and even deep space has constant microwave background radiation that can be detected with sensitive instruments.

\$\endgroup\$
3
  • \$\begingroup\$ Question. I have an open ended bare wire, attached to some circuit (grounded to.earth or other fixed potential point) on a pcb. In simple terms,I do.something at the far and of the wire, that would tend to cause charge to accumulate/drift to the loose or PCB end of the wire. For example place the loose end in proximity to a + or - static charge. Can that action cause a measurable effect in any way at the PCB end? For example, can I in any way detect a change in charge distribution, even transiently, at the PCB end? If so, ... (cont) \$\endgroup\$
    – Stilez
    Commented Jan 26, 2021 at 22:28
  • \$\begingroup\$ ... since a potential essentially is that which causes charge movement, can I use the same principle to detect a + or - change in the loose end potential, by detecting some corresponding (transient or orherwise) change to charge distribution at the PCB end? \$\endgroup\$
    – Stilez
    Commented Jan 26, 2021 at 22:29
  • \$\begingroup\$ You could induce a 'static charge' on the wire using a nearby object with varying electric field, or you could do it with a generator at one end of the wire which has a counterpoise, both of which would produce an an AC current in the wire. But now you have an antenna, and you specifically excluded radiated effects and fields outside the wire. physics.stackexchange.com/questions/286805/… \$\endgroup\$ Commented Jan 27, 2021 at 2:31

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.