FACTS, reactive power and voltage control - please correct me if I am wrong. 1

[Sergejs]

I am trying to wrap my head around FACTS, reactive power and voltage control. But when I read explanations like this:

Synchronous generators, SVC and various types of other DER (Distributed energy resource) equipment are used to maintain voltages throughout the transmission system. Injecting reactive power into the system raises voltages, and absorbing reactive power lowers voltages.

I am almost certain that this is a wrong way to describe this.

Here's how I see it, correct me if i am wrong:

You do not inject or absorb reactive power, and certainly capacitors do not generate reactive power. Reactive power just oscillates. If for example you have low p.f. (leading or lagging) the energy (that is reactive power) oscillates between generator and inductors or capacitors. If you introduce FACTS you just provide capacitors for inductive network (lagging p.f.) or inductors for capacitive network (leading p.f.) and your reactive power oscillates between L and C, ideally when p.f. is 1 all the energy from L goes to C and vice versa.

Now when I read things like:

...we must raise the voltage higher to push the power through the inductance of the lines (in regards to voltage decrease when p.f. goes low)

It also seems incorrect. I would say - by bringing power factor closer to 1 we just bring Z closer to R, hence the voltage drop across X becomes less and thus we improve voltage regulation by the use of FACTS.

 

[Skogsgurra]

Yes, you are wrong.

Capacitors do generate reactive power. That is a fact.

But this has nothing with free energy to do - simply because reactive power does not produce or consume energy over time.

And, if you need to "push" more current through a line you need to rise the voltage. Easy as that. And that is also waht you are saying. But you seem to have more of a classical circuit view on the subject. That is also OK, but it doesn't mean that you can tell the rest of us that we are wrong.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[Sergejs]

Dear Skogsgurra

I just feel like a lot of people misunderstand electricity just because of the fact that we explain it in a way that does not seem logical.

Capacitors store potential energy (electric field), inductors kinetic (magnetic field) and when we supply that first portion of energy from the generator to this oscillating circuit it just continues to oscillate - this just seem more straightforward to me.

 

[Skogsgurra]

If you think so, OK. But we (most of us in the profession) prefer to use a terminology that has been established over more than hundred years and never really shown any flaws that could result in an alternative view or terminology.

For how long have you been an EE?

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[waross]

I have seen an old diesel power plant run at very light load, (saves fuel and wear and tear on the old engines) but over excited to control the voltage at a city at the end of a long transmission line. The reactive current caused by over excitation circulated back through the source but in the opposite direction to the load current. Reactive voltage drop on a transmission line may be greater than the resistive voltage drop. The reactive current in the reverse direction causes a voltage rise that is controlled to cancel out the resistive voltage drop and hold the voltage at the receiving end at the desired value. It is possible to drive the voltage at the receiving end higher than the voltage at the sending end.
The people who work with these systems may be hard pressed to do the calculations needed with a simple, easy to use explanation that does not include all the effects of reactive current.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[LiteYear]

Of course capacitors do not generate reactive power, regardless of how many decades of sensitive investment have been placed in that notion.

Capacitors can supply and absorb reactive power (from and into energy in electric fields respectively). So can inductors (from and into energy stored in magnetic fields respectively).

By virtue of the fact that most of our useful electrical world runs on magnetic fields rather than electric fields, we've come to think of inductors as "absorbing" reactive power and capacitors "supplying" it. The convention then is that our inductive loads "absorb" reactive power and our capacitive banks "supply" it. In fact, the reactive power is just oscillating between the leading and lagging loads and both are alternatively absorbing and supplying. The convention is useful, very well established, and works well in systems where inductive loads are dominant. It's a bit less helpful, for example, in SWER networks where the capacitive lines dominate and the Ferranti effect becomes significant. The convention can be a bit confusing then, so it's sometimes useful to drop it and go back to the power factor triangle.

 

[crshears]

I'm a bit late to the party; I've been on time off from work, and at work is almost the only time I visit E/T since I'm on dial-up at home and ET is way, way too 'busy' / graphics-intensive for my low-performance dial-up connection...but I digress.

The conventional terminology that we as operators, I mean, controllers, use within my company 'polarizes' the supply/absorb conundrum for us simple folk by stating that inductive devices absorb lagging reactive power and capacitive devices produce lagging reactive power.

We find this convention works very well in practice, both on the predominantly inductive portions of our power grid as found in the industrial cores of a number of our urban centres and on our highly capacitive Ferranti-effect-influenced long-distance high-voltage transmission lines.

Interestingly enough, using this convention does not seem to be an impediment to us grunts grasping the concept of how series capacitors are used in some of our high-voltage tarnsmission lines to reduce the effective impedance of the circuits, allowing for transmission limit expansion, something that both makes us money and pleases our IESO immensely.

And fwiw, LiteYear, I'm not at all sensitive about the topic; I just prefer to apply the KIS principle [ "Keep it simple!" ] whenever possible...and I've found this approach comports very well with the venerated power triangle.

CR

"As iron sharpens iron, so one person sharpens another." [Proverbs 27:17, NIV]

 

[dpc]

LiteYear - so you are hung up on the semantic difference between "generate" and "supply"? Seems to be a distinction without a difference to me.

I think a lot of the confusion comes from the term "reactive power", since power is usually related to real power (kW). In the power industry, capacitors produce lagging vars, inductors absorb lagging vars. This is the convention used in every power flow analysis that I have ever run across. There is no conflict with classic circuit analysis that I am aware of, as long as the terminology is understood.

 

[electricpete]

I'm with skogsgurra and dpc. I don't know what point Buzz Liteyear is trying to make.

It's certainly true that reactive elements are associated with energy storage and the instantaneous power flow might be either in or out of either element at a given point in time….

BUT when we look at sinusoidal steady state, we role all of this into vectors (time dependence is supressed).


If you attach inductor or capacitor to the same source you will certainly have opposite "polarity" of current flow in the two cases (since the inductor lags voltage by 90 and capacitor leads by 90, they are 180 from each other). We can capture this "polarity" by the notion of generating or consuming vars. The terminology is very useful and extends to describe very well the concept of voltage drop across inductive transmission elements. All of this flows naturally from the definitions

apparent power S = V conjugate (I)
real power P = Real{S}
imaginary Q = Imag{S}


=====================================
(2B)+(2B)' ?

 

[LiteYear]

Yep, all agreed. Was just answering the OP's question. Apologies for the offence. My handle seems to cause some angst around here...

 

[electricpete]

Thanks Liteyear. No apologies needed - I apologize.

I just re-read your last paragraph and I now see it was a correct and nuanced response. To paraphrase, you're saying that vars associated with shunt-connected inductor and capacitor have opposite polarity, but the longstanding convention as to which one we choose to call positive and which one we call negative (or which one generated and which one consumed) is arbitrary. That's a good point. The reason for my initial reaction was that Op was imo struggling at a more basic level and your first paragraph (which was correct in context explained by your last paragraph) didn't help the issue he was struggling with imo.

Sorry for butchering your handle.

=====================================
(2B)+(2B)' ?

 

[electricpete]

but the longstanding convention as to which one we choose to call positive and which one we call negative (or which one generated and which one consumed) is arbitrary.

For benefit of op, I should clarify it's not arbitrary if we accept the math definitions of S, P, Q I listed above.
But those definitions themselves contain an arbitrary choice (S = V conj{I} rather than S = conj{V} I)


=====================================
(2B)+(2B)' ?

 

[LiteYear]

Thanks eletricpete. Nice paraphrasing.

At the risk of causing more consternation, I think the following perspective is interesting in understanding the confusion: as already described we need a convention and capacitor = supply works well. The problem is when people make the natural connection from the supply convention <-> supply <-> generate <-> energy source. This is a natural connection because it is true for real power! Energy due to real power really does flow in one direction, from source to sink. So the supply is also the generator is also the energy source. In reactive power, there's actually no resulting energy transfer, despite what our convention would suggest.

I found the following explanation particularly enlightening, despite its somewhat fickle source. Stick with it, the picture developed of alternating current and energy flow is elegant:

http://wiki.answers.com/Q/What_is_reactive_power

 

[LiteYear]

Oh my, and now I've butchered your handle. Sorry electricpete, slipped on the submit button!

 

[electricpete]

Funny, now I don't feel so bad about butchering your handle. It seems there was a buzzlightyear on the forum awhile back.

I'll add to your discussion because it circles back to the original post.

Your link tells us that the average energy transferred by reactive power is zero. That is of course inherent in the definition of reactive power (vars). Only the real power has a non-zero average power.

So when we talk about generating/injecting vars, we're not talking about generating/injecting real power, we're talking about generating/injecting reactive power (of course).

Why do we care about these vars?

In a power system where the branches are primarily inductive, the real and reactive power flow through the system are approximately "decoupled". Specifically:

#1 - Real power flow thorugh a given branch depends primarily on voltage PHASE difference across the branch. Equivalently, if we know the real power flow through a branch (and the branch inductance), we can estimate voltage phase difference across the line.

#2 - Reactive power flow through a given branch depends primarily on the voltage MAGNITUDE difference across the lbranch. Equivalently, if we know the reactive power flow through a branch (and branch inductance), we can estimate voltage magnitude difference across the branch.

Google "fast decoupled load flow" for more rigorous assumptions/derivations to support #1 and #2 above.

Based on #2, we know it will be instructive to trace the flow of vars through the system in order to get a picture of the voltage magnitude profile throughout the system.

Look at the example of Ferranti effect you mentioned. The long line is modeled by a series inductance with capacitance at each end. We have one end open and one end connected to the power system. The vars injected by the capacitor at the open end must flow through the line series inductance to get to the power system (that is the only place they can go). This creates a voltage drop as those vars flow from the open end to conencted end through the series inductance. Accordingly, the open end of the line has a higher voltage. Tracking the flow of the vars helps us mentally model the voltage profile.


=====================================
(2B)+(2B)' ?