Calculating line impedance for distance relaying

[mbrooke]

When calculating line impedance (for zone 1 and zone 2 settings), what do I need to take into account and are there any special considerations? Flat vs horizontal conductor spacing?

 

[steve66]

Of course the configuration comes into play. Looking at my old textbook, it looks like its really math intensive, even for simple configurations.

Can you let a computer program do the heavy lifting for you? I don't have Powerworld, but it sounds like it might be what you need:


http://www.powerworld.com/files/TransLineCalcHelp1.pdf

 

[mbrooke]

Of course the configuration comes into play. Looking at my old textbook, it looks like its really math intensive, even for simple configurations.

Awesome stuff, thank you!

Can you let a computer program do the heavy lifting for you? I don't have Powerworld, but it sounds like it might be what you need:


http://www.powerworld.com/files/TransLineCalcHelp1.pdf

No, just a calculator and basic programs Though Powerworld would be ideal in this case.

 

[mbrooke]

Of course the configuration comes into play. Looking at my old textbook, it looks like its really math intensive, even for simple configurations.

Can you let a computer program do the heavy lifting for you? I don't have Powerworld, but it sounds like it might be what you need:


http://www.powerworld.com/files/TransLineCalcHelp1.pdf

And oh, for the experts: what temperature to use when basing the calculations? Do I just pick the median average of the highest and lowest temps the conductor will encounter? This has me stumped, but Im guessing as long as I try the highest and lowest and make sure neither force zone 1 to over reach and zone 2 to under reach I should be fine?

 

[steve66]

Awesome stuff, thank you!

Yea, I know saying "It's really complicated." isn't much of an answer. But it's really complicated.

I don't even recognize one of the symbols anymore. What does a capital pi looking thing mean? As in:

"sigma ln A_k = ln CapitalPiSymbol A_k" ??

 

[Ingenieur]

Yea, I know saying "It's really complicated." isn't much of an answer. But it's really complicated.

I don't even recognize one of the symbols anymore. What does a capital pi looking thing mean? As in:

"sigma ln A_k = ln CapitalPiSymbol A_k" ??

it is the product of the defined range of the series of numbers
A1 x A2 x A3 x..... x An
k = 1 thru n
as far as the op a program is needed for all but the simpliest ckts

 

[steve66]

it is the product of the defined range of the series of numbers
A1 x A2 x A3 x..... x An
k = 1 thru n
as far as the op a program is needed for all but the simpliest ckts

That makes sense, since adding logs is like multiplying numbers.

 

[mbrooke]

Yea, I know saying "It's really complicated." isn't much of an answer. But it's really complicated.

I don't even recognize one of the symbols anymore. What does a capital pi looking thing mean? As in:

"sigma ln A_k = ln CapitalPiSymbol A_k" ??

No, the paper was of a lot of help, to which I appreciate The only weak answer is one thats never given IMHO.

 

[Ingenieur]

Most college texts on power engineering will include a software package
usually does basics like:
transmission line parameters
Phase to seq component conversions
basic fault calculators
basic system modelling

there are also std tables for various geometries, conductor sizes, etc http://www.rd.usda.gov/files/UEP_Bulletin_1724E-200.pdf

example http://www.cengage.com/search/productOverview.do;jsessionid=90CD646F03F1834D9EBDA1EA5EC379BF?N=16&Ntk=all%7CP_Isbn13&Ntt=power+system+analysis%7C9781111425777&Ntx=mode%2Bmatchallpartia

l

 

[mbrooke]

Neat.


would anyone have a basic example? There is an easy way out with differential relaying, however a backup like step distance is highly recommended should communication go down.


Speaking of communication, all modern 345kv and 115kv systems have communication based relaying?

 

[mbrooke]

This any good?


http://www.gegridsolutions.com/Multilin/notes/artsci/art14.pdf

 

[Bugman1400]

The line impedance needs to be obtained from a line constants software. Please do not attempt this by hand. The pos seq impedance (Z1MAG, Z1ANG) is not too hard but, I guarantee you will not come close for the zero seq impedance (Z0MAG, Z0ANG). There are just too many factors (cond spacing, pole/structure height, soil resistance, static wire size and spacing, mutual, k-constant, etc). With typical Zone settings of 85-90%, you could easily overshoot the Zone 1 and overtrip the line for out of zone faults.

If you have the software and want to model, I typically start with the Plan & Profile dwgs from the T-line engineers.

 

[mbrooke]

The line impedance needs to be obtained from a line constants software. Please do not attempt this by hand. The pos seq impedance (Z1MAG, Z1ANG) is not too hard but, I guarantee you will not come close for the zero seq impedance (Z0MAG, Z0ANG). There are just too many factors (cond spacing, pole/structure height, soil resistance, static wire size and spacing, mutual, k-constant, etc). With typical Zone settings of 85-90%, you could easily overshoot the Zone 1 and overtrip the line for out of zone faults.

If you have the software and want to model, I typically start with the Plan & Profile dwgs from the T-line engineers.

3 Hypothetical questions:

1. What if I deliberately under reach my ground zones? Zone 1 at 55%, zone 2 at 80% and zone 3 at 120%?

2. Ive heard of directional ground elements taking over for this.

3. If I can't obtain the correct values, could I just have differential relaying do all the work for me provided I have
2 independent channels of communication?

 

[Bugman1400]

3 Hypothetical questions:

1. What if I deliberately under reach my ground zones? Zone 1 at 55%, zone 2 at 80% and zone 3 at 120%?

2. Ive heard of directional ground elements taking over for this.

3. If I can't obtain the correct values, could I just have differential relaying do all the work for me provided I have
2 independent channels of communication?

1.) Or, you could have it done right and get someone to model or create the settings for you.....and the liability that goes with that falls on them.
2.) Same as #1
3.) You could have curr diff perform the protection and hope that the comm doesn't go down. My experience is that when one goes down they both do....because the channels are not diverse. Curr Diff also does not provide any backup. Also see #1.

If you need help...ask. You know where I'm at.

 

[mbrooke]

1.) Or, you could have it done right and get someone to model or create the settings for you.....and the liability that goes with that falls on them.
2.) Same as #1
3.) You could have curr diff perform the protection and hope that the comm doesn't go down. My experience is that when one goes down they both do....because the channels are not diverse. Curr Diff also does not provide any backup. Also see #1.

If you need help...ask. You know where I'm at.

Even if primary and backup relaying has its own OPGW? (just thinking aloud)

Anyway, 67 could cover me for ground relaying, but isnt selective co-ordination lost?

 

[mbrooke]

1.) Or, you could have it done right and get someone to model or create the settings for you.....and the liability that goes with that falls on them.


If you need help...ask. You know where I'm at.

And in all honesty I agree, however I would still like to learn some of this for myself. In my view there is to much compartmentalization, I myself would like to know how its done, even with the computer programs that give me the easy way out (pf, short circuit, arc flash, ect).

 

[Bugman1400]

If you want to learn then additional info is needed about the application:

1.) What is the approx. length of the study line (1 mile, 10 miles, 100 miles)?
2.) What kV are we talking about here?
3.) Does the line go to a gen plant? If so, what kind of plant? Yes, this matters...different methods apply to solar farms than to rotating generation.
4.) 67 directional will only cover you if sufficient polarization is available. The 67G is only for ground faults. What about phase faults? Would this even coordinate downstream or upstream.

Paint us a better picture here. I hope you're not still working on that main & xfer scheme.

 

[mbrooke]

If you want to learn then additional info is needed about the application:

1.) What is the approx. length of the study line (1 mile, 10 miles, 100 miles)?

14 miles.

2.) What kV are we talking about here?

115kv, 1590 (45/7) Lapwing conductor with the following ratings at 100*C conductor temperature:

Summer: 1350 amps normal; 1720amps 3 hour overload; 2000 15 minute emergency overload

Winter: 1500 amps normal; 1820amps 3 hour overload; 2100 15 minute emergency overload

(replacing an old 4/0 cu paralleled circuit line)

Another line (not parallel, separate structure), 8 miles to a different substation:

795 (26/7) Drake conductor with the following 110*C conductor ratings:

Summer: 945amps normal; 1215amp 3 hour overload; 1335 amps 15 minute emergency

Winter: 1045 normal; 1275 3 hour overload; 1395 15 minute emergency



Yes the Drake line is considered slightly over driven by some POCOs, but the sag limits check out ok.

3.) Does the line go to a gen plant? If so, what kind of plant? Yes, this matters...different methods apply to solar farms than to rotating generation.

No, but generating plant a few substations down. It is steam (rotating prime mover) so it does not have the rotor angle stability and short circuit constrains of say a wind farm or solar.

4.) 67 directional will only cover you if sufficient polarization is available. The 67G is only for ground faults. What about phase faults? Would this even coordinate downstream or upstream.

Phase faults will be through an MHO distance setting. Ground might as well be through that as well.

Paint us a better picture here. I hope you're not still working on that main & xfer scheme.

No, different application. The long standing M&T paradox has been resolved by applying SEL400 series relays to all M&T applications. The transfer buss breaker is wired as breaker #2 on the relay's CT inputs. This takes care of switching which use to involve bypassing line protection in order to swap CT circuits to a single CT input relay. For this I owe you and another member on another forum a great deal or respect.

 

[Ingenieur]

I guess my question is why doesn't the relaying/protection group or the line engineering group of a public utility have the basic software tools to do this?

 

[Bugman1400]

Since the other T-line is not in the same ROW then no mutuals should factor. They other parameters that are need to estimate the line impedance are:

1.) What type of static wire is being used and how many (1 or 2)?
2.) What is the avg structure height from the ground up (don't include 10%+2' embedment)?
3.) What is the physical spacing of the conductors with respect to the CL of the structure and the top of the structure?
4.) Give me an idea of the soil impedance (or soil type).
5.) Lastly, can you provide more info on the T-Line length? If you have to give the structure spans (str. by str.), that is good. For a ~14 mile line, a difference of 1.4 miles is 10% error (0.7 mile = 5%) and is considered too much of an error. 10% error could easily cause an over or under trip event, result in mis-coordination, or simply fail to adequately protect the line.