4
\$\begingroup\$

Does transmission line theory and transmission line termination take any significance when dealing with long high-voltage power lines? Do power companies terminate their power lines to avoid reflections in their power lines over great distances? (say 1000 km for example)

If so, how do they do it (in general terms)?

Most of the sources I have found talk about electronic and PCB design, I have not been able to find anything about this phenomenon in the electrical power industry so far.

Any article or reading material suggestions are very welcome in the comments.

\$\endgroup\$
4
  • 1
    \$\begingroup\$ I am quite sure that no resistive termination would be used. It wouldn't make any sense at all. It would just be a waste of power. \$\endgroup\$ – mkeith Feb 27 at 7:53
  • \$\begingroup\$ The line impedance could be artificially adjusted using inductors and/or capacitors placed at 0.125 wavelength intervals. At 60 Hz, 0.125 wavelength is around 375km. Slightly longer at 50 Hz. I doubt there are many such lines that run for 375 km with no producers or consumers along the line. NOTE: I am not a power transmission engineer. I have never heard that this is done. I am just talking about theoretical possibilities for matching on very long lines. \$\endgroup\$ – mkeith Feb 27 at 21:09
  • \$\begingroup\$ @mkeith 375km looks like 0.125 wavelength for 50 Hz cable line (3000/8). For 50 Hz overhead line wave length is 6000 km. IMHO. \$\endgroup\$ – AlexVB Feb 28 at 12:04
  • \$\begingroup\$ @AlexVB ok looks like I made a mistake. So 60 Hz would be a bit less than 375km. But still a long way, and the basic comment still applies. \$\endgroup\$ – mkeith Feb 28 at 18:19
3
\$\begingroup\$

Power transmission lines are not terminated with a resistance in the way that data transmission lines are. As a result, there is reflection on power transmission lines. The voltage near the end of a power transmission line is often higher than at its source. This is known as the Ferranti effect. Power utilities compensate for the Ferranti effect by having step-down distribution or service transformers with a variety of taps on them. Distribution or service transformers provide the final "mains" voltage for end-users. Which tap is selected on the service transformer depends upon the voltage on the transmission line where the transformer is located. Thus, although the transmission line voltage varies from point to point, the "mains" voltage supplied to end users can generally be kept within 10% of nominal (in the US).

\$\endgroup\$
4
  • \$\begingroup\$ Thanks for the sources, indeed ferrantini effect is the search tearm I was looking for. Aparently there are also shunt reactances that are placed at various midpoints in the transmission line to mitigate it's capacitance. electricalbaba.com/… \$\endgroup\$ – Joaquin Brandan Feb 28 at 17:37
  • 1
    \$\begingroup\$ I believe that bank capacitors are added to transmission lines not to alleviate the Ferranti effect, but for power factor correction. Motors and other inductive loads are often a large component of the total load, and adversely affect power factor. Capacitors added to the grid reduce the apparent power needed to be supplied by the power plant. \$\endgroup\$ – Math Keeps Me Busy Feb 28 at 17:41
  • \$\begingroup\$ The shunt reactors I mentioned are inductors connected to ground apparenty, not capacitors. (check the link, I know it's not a decent source at all but I'm having a lot of trouble getting reputable sources, it seems this industry is not as accesible as others regarding sources) \$\endgroup\$ – Joaquin Brandan Feb 28 at 21:03
  • \$\begingroup\$ Interesting. I had never heard of this before. Thank-you. \$\endgroup\$ – Math Keeps Me Busy Feb 28 at 23:02
2
\$\begingroup\$

The difference between low-voltage data lines and high-voltage power lines is that the latter are built to transmit power and sinking large part of it into a resistor at the end would spoil their purpose.

What power companies care about is voltage and reactive power. When power flow over the line is below natural loading, the line produces excess reactive power and voltage rises. The opposite happens when the line is loaded higher than natural loading.

To cope with these issues power companies use inductances and capacitors (or equivalent power electronics devices) to, in effect, alter transmission line parameters.

For example you may add inductance between the line and the ground to lower voltage during low load conditions including the case when line is temporary connected from only one side. This is shunt compensation. The inductance is called shunt reactor and in the power engineering area you would say it is a device to consume reactive power.

On the other hand you may want to reduce transmission line inductance to allow more power flow by connecting a capacitor in series with the line. This is called series compensation. It also helps to reduce voltage drop over the line.

Shunt compensation is necessary for extra high voltage (400 kV and above) power transmission lines. Series compensation may be used on 110-220 kV lines to reduce voltage drop over the line and on EHV lines to allow more power flow.

\$\endgroup\$
2
\$\begingroup\$

In a power line you have reflections like phenomena at any load that has a reactive component. This reactive energy bounces back and forth, same way as in RF transmission line. The countermeasure for this is a compensation units (capacitor banks).

Due to very long wavelength there are no nodes, but if this is an issue the AC is converted to DC, then you have a very long HV DC power line and then back converted to AC. This method is also used for load sharing within two separate AC systems (non-synchronized). .

\$\endgroup\$
2
\$\begingroup\$

do power companies terminate their power lines to avoid reflections in their power lines over great distances

No. That's nonsense because the size and weight of the transmission line components isn't what is the target of optimization. Power transmission lines are optimized for energy conservation. You don't want to dump GWh of electrical energy in favour of having tiny components.

And even if you put that aside, the impedance-matched power U²/|Z| of the overhead wires is roughly 1/4 of the power that is actually sent over those lines. And that means there's no way termination could help. Because the equivalent resistance of the consumer at the end of the line is already lower than the calculated termination resistor. So you had to oversize in any case if you wanted to terminate.

For power cables, you could in theory use termination because they are run below their impedance-matched power. But in practice that would mean the cables needed to be cooled all the way.

In reality, the inductivity of overhead wires and the capacitance of cables is balanced by active components near the consumer that are switched on and off depending on the amount of power delivered.

\$\endgroup\$
13
  • 4
    \$\begingroup\$ What do you mean when you say "natural power"? The term seems ill-defined. \$\endgroup\$ – DKNguyen Feb 27 at 7:23
  • \$\begingroup\$ It's that power defined by the impedance of the transmission line. U²/|Z|. \$\endgroup\$ – Janka Feb 27 at 8:21
  • \$\begingroup\$ Please add that into your answer as well as a brief description of why it is relevant. It is not a term that returns any relevant results in Google.. \$\endgroup\$ – DKNguyen Feb 27 at 18:20
  • 1
    \$\begingroup\$ I think the term might be "Surge Impedance Loading" which is also called the "Natural Load"? The power at which the capacitive and inductive reactive powers are equal and opposite and the line is purely resistive". Is that what you are referring to? circuitglobe.com/surge-impedance-loading.html \$\endgroup\$ – DKNguyen Feb 27 at 19:59
  • 1
    \$\begingroup\$ I rephrased and used impedance-matched power instead of “natural power”. \$\endgroup\$ – Janka Feb 28 at 10:13

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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