Are there reactance effects for high frequency DC?

[hurricanehenry]

Hello. I am wondering if there are any reactance effects for a varying DC voltage? i.e., an AC wave with offset so that there is no negative DC component. I am wondering if transmission line impedance mismatch will have any involvement if the signal never goes negative? (i.e., line reflection phenomenon that can reduce usable power).

Or is the entire concept of impedance matching (and reactance) completely irrelevant for anything other than AC.

Thanks.

[Zero999]

The signal doesn't have to go below negative to have an AC component.

For example, a voltage that changes between 1V and 2V at a frequency of 1GHz is really 1.5VDC + 1GHz AC and transmission lines will apply, unless the signal path is very short of course.

[cvanc]

^^^What Hero999 said^^^

"High frequency DC" is called "AC"

[Brumby]

^^^What Hero999 said^^^

Such a signal will have an AC component and a DC component.  Each component has its own properties which must be considered and they can both exist together by appearing as you have described.

[T3sl4co1l]

AC is the part of the signal which is not constant.

Tim

[mrpackethead]

and you can have multiple ac signals on the same wire, at the same time.

[Brumby]

... but only 1 DC.

[Ian.M]

<pedant>
There is no DC, as for that to actually exist would require the circuit to be in existence carrying an unchanging current from the big bang to the end of the universe.  *EVERYTHING* is AC, but some of it is a very very very low frequency, and the concept of DC is a useful simplification of sufficient accuracy.
</pedant>

[PA0PBZ]

*EVERYTHING* is AC

Everything is DC until it changes amplitude 

[max_torque]

I've got some ultra low frequency batteries you can buy if you like?   

[Ian.M]

For primary cells, anything higher frequency than 1 nHz, unloaded is junk!

[jitter]

Hello. I am wondering if there are any reactance effects for a varying DC voltage? i.e., an AC wave with offset so that there is no negative DC component. I am wondering if transmission line impedance mismatch will have any involvement if the signal never goes negative? (i.e., line reflection phenomenon that can reduce usable power).

Or is the entire concept of impedance matching (and reactance) completely irrelevant for anything other than AC.

Thanks.

You might consider DC a 0 Hz AC signal. Your AC signal can include many different frequencies, including 0 Hz. Yes, your transmission line probably works differently at 0 Hz, but so does it do at 1 kHz, 1 MHz, 1 GHz, etc.
And remember that voltages are taken with respect to a reference point. Change the potential of a reference point, and suddenly you find your AC signal might not have an offset without changing anything to the actual signal.

[rdl]

How can DC be AC?

Direct Current which changes voltage, no matter how fast, is still DC. The current does not change direction (alternate).

[jitter]

What I clumsily tried to say is that DC belongs in the frequency spectrum just as much as 0.000001 Hz or 1000000 THz do and that a transmission line has different characteristics at different frequencies.

[Brumby]

How can DC be AC?

Direct Current which changes voltage, no matter how fast, is still DC. The current does not change direction (alternate).

Perhaps in the strictest physical sense ... but try and do any circuit analysis or design with that model.

It's only when you look at it as an AC signal with a DC bias - ie, the two components addressed individually on their respective attributes - that you get the ability to work with it.


This idea of working with a model that (some might argue) does not directly represent a physical phenomena, is not uncommon.  The Universe is a weird and wonderful place - with complex things going on.  Often, it is only by breaking something down to it's component parts or some abstract mathematical construct that we are able to get a handle on it.

Consider the use of complex arithmetic for phase and amplitude in signal processing.  Heck, part of the mathematics has an element that is even called imaginary.  But throwing in this 'imaginary' bit makes the maths work.


Perhaps the key to the 'confusion' here is simply a matter of semantics and/or perspective...  The term "DC" can be viewed slightly differently and the inferences may change - demonstrated by the different perception of someone with an EE/hobbyist background and someone who does not.

<pedant>
There is no DC, as for that to actually exist would require the circuit to be in existence carrying an unchanging current from the big bang to the end of the universe.  *EVERYTHING* is AC, but some of it is a very very very low frequency, and the concept of DC is a useful simplification of sufficient accuracy.
</pedant>

I did like this ... and the posts that followed, but there is one fundamental quality about it - it is correct.

For example - a computer is turned on at 9:00am and then shut down at midnight every day.  If we measure the 12V rail ... is it DC - or is it an AC square wave with a frequency of 11.57 uHz, a duty cycle of 62.5% and really bad rise times?

[Brumby]

Then there's this little gem...

And remember that voltages are taken with respect to a reference point. Change the potential of a reference point, and suddenly you find your AC signal might not have an offset without changing anything to the actual signal.

 

[techie1234]

How can DC be AC?  Direct Current which changes voltage, no matter how fast, is still DC. The current does not change direction (alternate).

Perhaps the key to the 'confusion' here is simply a matter of semantics and/or perspective...  The term "DC" can be viewed slightly differently and the inferences may change - demonstrated by the different perception of someone with an EE/hobbyist background and someone who does not.

I think that is indeed the key point, as the terms AC and DC are sometimes used too loosely and highly depend upon the context.  Often it is said that any flow whose voltage doesn't drop below 0 is "DC" when in fact it is alternating at a non-trivial frequency.  And a 100% constant (i.e., 0.00000000000...Hz) voltage is as impossible to achieve as a perfect insulator or absolute zero.  It just depends upon your frame of reference and tolerance for fluctuation.

[Zero999]

How can DC be AC?

Direct Current which changes voltage, no matter how fast, is still DC. The current does not change direction (alternate).

That's not true. The current does change direction. All conductors posses a certain parasitic capacitance, so whenever the voltage changes, a current will flow. Using conventional current flow: when the voltage increases, electrons flow into the cable, charging up the  parasitic capacitance and when the voltage decreases, electrons flow back out of the cable, discharging the capacitance. The current flow will depend on the frequency, change in voltage and impedance of the cable.

Direct current which changes voltage is in fact two signals: a DC component which remains constant and the AC part which continuously fluctuates.

[hurricanehenry]

Thanks everyone for a very elucidating discussion.

[Brumby]

Getting back to the thread title...

The answer is a very simple 'Yes'.

Reactance is a function of the time rate of change of voltage and/or current.  It has nothing to do with polarity as such.  This time rate of change is represented by the AC component - and that is what you use in reactance calculations.

This is where the 'philosophical' discussions on the merits of a 'high frequency DC' signal are irrelevant.  It doesn't matter HOW you look at it.  The simple fact of the matter is that the MATHEMATICAL description of an AC signal with a DC bias is complete and accurate.


You could have a 'philosophical' discussion on a circuit model with Tinkerbell sitting or standing on a see-saw waving her wand up and down - but the mathematical model is all you need to correctly analyse what's going on and what effects you will observe.

[techie1234]

I guess I'd never really thought about it (and of course we're 100 years too late...) but if instead of the terms "AC" and "DC" becoming established, the terms "Alternating Voltage (AV)" and "Constant Voltage (CV)" had come into use some of the confusion would be avoided.  Perhaps we could have also fixed conventional flow nomenclature at the same time...

[R005T3r]

Nope: Voltage is not a fundamental physical dimension, current is, as a result we talk about current, not voltage.

[Mechatrommer]

nope: no voltage no current. you cant make up current source if you dont have potential (charge) difference available between 2 space connected by a transfer medium. btw, ac is whenever the electrons' net momentum change. and they always follow the rules because they dont have a choice, not like human.

[Chalcogenide]

Well, if we really want to go deep, every current is a superposition of pulses corresponding to a single electron crossing a boundary (pulses which are shaped according to the Schockley-Ramo theorem). In fact, in the real world, every "DC" current is also affected by noise, which are just its AC components.

[hurricanehenry]

nope: no voltage no current. you cant make up current source if you dont have potential (charge) difference available between 2 space connected by a transfer medium. btw, ac is whenever the electrons' net momentum change. and they always follow the rules because they dont have a choice, not like human.

Yeah this is something I've always not been too happy about, whenever the focus is on current. I've even read an author who said that BECAUSE of current flow through a resistor, it CREATES a voltage. Now, in all fairness, the resistor is sitting in the circuit, and the actual potential difference is being caused by a voltage source.

(Since I'm the OP I think we should allow a small side discussion on this, as it's related to my original question and I want to know what is real situation with this current v voltage thing).

[Mechatrommer]

in physical reality, you can make V and R on the macro/larger/external system scale, current will follow afterward. in effect the internal nodes potential however will vary in time based on each component's reactance (ie the current flow through it, with some right combination, they will oscilate. and then dampen if not agitated by another current flow. the hardest part is try to understand the balance of them (the internal nodes), thats what ee school is for. btw dc is whenever there is nonzero net of electrons transfer at a particular time and space reference, so if you pick less than a period of a freq as your time reference, then you'll have a dc component in pure ac signal, my 2cnts.

[Brumby]

nope: no voltage no current. you cant make up current source if you dont have potential (charge) difference available between 2 space connected by a transfer medium. btw, ac is whenever the electrons' net momentum change. and they always follow the rules because they dont have a choice, not like human.

Yeah this is something I've always not been too happy about, whenever the focus is on current. I've even read an author who said that BECAUSE of current flow through a resistor, it CREATES a voltage. Now, in all fairness, the resistor is sitting in the circuit, and the actual potential difference is being caused by a voltage source.

(Since I'm the OP I think we should allow a small side discussion on this, as it's related to my original question and I want to know what is real situation with this current v voltage thing).

Sometimes trying to discuss relationships can be best served by looking at a extreme and/or limiting case...

In the case of voltage and current - what can be said when there is a current through a superconductor?

[R005T3r]

Superconductors...They advertise the material has 0 DC resistance, but in reality it's not like that: even if it's infinitesimal to a certain degree of magnitude does not mean it's actually zero: it approaches to zero, but it's not zero!. So, what's the deal about superconductors? It's that they dissipate so low power that it will make the difference in a mains distribution system, as a result if anyone uses it we will have an efficiency boost of the 30% on the actual global distribution system (and it's a huge amount)

However, the "ideal" wire such as the ones you see on the simulator, it will never-ever exist. As for current, if you will have 1A flowing trough your superconductor, it will be: 1.6x10^19 electrons per second again, you can say current flowing trough an object, but as a regard of the voltage drop? it would be near to zero but there will be some value... This means that the definition is well formulated.

[T3sl4co1l]

Superconductors...They advertise the material has 0 impeadence

No, superconductors have zero DC resistance.

Period, end of story.

No one has ever said they have zero impedance, except erroneously!

Indeed, type 2 superconductors have AC resistance -- it's even macroscopically visible!

Type 1 superconductors, carefully prepared, have quite low AC losses, even at high frequencies: resonators, of polished niobium at 2.2K, have a Q factor higher than most quartz crystals -- about 3 million.

The loss is small but nonetheless present, and it's poorly understood.  One could only guess at the mechanisms; surface quality and purity probably have a lot to do with it.

All superconductors cease at some finite frequency, because that is the frequency characteristic of the binding energy of the Cooper pairs, ~1meV, which is in the 10s of GHz microwave region.

This can be used as a very fast switch, actually: the superconducting electrons are a thermal reservoir somewhat separate from the material itself (giving rise to the electronic heat capacity -- usually a small fraction of the total heat capacity, which is predominantly due to phonons -- acoustic vibrations).  If the electrons are heated by zapping them with microwaves or higher frequencies, superconduction stops instantly, then eventually recovers as the electrons come back to equilibrium.

Tim

[R005T3r]

Yes, I wanted to mean DC resistance, not impedance... My bad.

[Zero999]

All superconductors cease at some finite frequency, because that is the frequency characteristic of the binding energy of the Cooper pairs, ~1meV, which is in the 10s of GHz microwave region.

This can be used as a very fast switch, actually: the superconducting electrons are a thermal reservoir somewhat separate from the material itself (giving rise to the electronic heat capacity -- usually a small fraction of the total heat capacity, which is predominantly due to phonons -- acoustic vibrations).  If the electrons are heated by zapping them with microwaves or higher frequencies, superconduction stops instantly, then eventually recovers as the electrons come back to equilibrium.

I take it that means a superconductor stops superconducting when exposed to photons with a high enough energy to break the Cooper pairs? So if a superconducting magnet were exposed to visible radiation it would lose some of its magnetic field.

[T3sl4co1l]

Yes; of course, this only applies to the penetration depth of light in the material -- perhaps 100s of nm.

It's of practical use (I think?) in superconducting thin films, but in a bulk material, there's plenty of sub-surface left to handle things; and for that matter, in a practical superconducting cable, there's a jacket of copper metal around everything, which is even less transparent.

If the current has a place to go, the magnetic field magnitude won't change.

Tim