circulating currents in the screen of MV cables

[tanaru]

Hi,

I find out that one of the older wind farms from one of our customer has some serious issues with the circulating currents in the screens of the MV XLPE cables between the transformer and the MV cell. the values for the current are extremely high and they are having serious heating issues. We established that one of the issues was the way they laid the cables. the connection between the trensformer and the MV cell is done with three paralleled cable per phase and they laid them in RRR GGG BBB configuration instead of the RGB recommended formation. This determines variations between the cable length from 54 m on one of the phses to aprox 47 m on other phase.the cables are 33 kV XLPE of 500 mmp copper with 35 mmp copper screen and the distantce between phases is 25 cm.
The screens are both end bonded and the transformer is 110/33 kV of wye-wye configurations.
My question is that what other elements can drive those currents in the screens to be that high beside the laying configuration which is wrong?
Thank you,
Best regards,

 

[Skogsgurra]

That problem is quite common when you use separate cables. Putting phases in separate cables (RRR GGG BBB) is, as you say, not a good idea. The reason for the high screen currents is mostly inductive coupling from the magnetic field surrounding each cable. Cables running 50 m and close to each other have a mutual inductance that can't be ignored. Put three phases (RGB) together to balance out the magnetic fields. That should reduce your problem an order of magnitude. And that is usually sufficient.
Another "fix" is to increase the impedance of the "shorted secondary" by installing a reactor in each connection from screen to Ground. Either end works.

Gunnar Englund

http://www.gke.org

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

 

[7anoter4]

The right configuration for 3 cables per phase –three phases system- has to be in triangles as Skogsgurra said.
[See:Frequent errors in cable run[Prysmian-Argentina]:

http://www.cooprg.org.ar/documentacion/Errores en el trabajo con cables.pdf

If you have to set them in a flat configuration a minimum unbalance you’ll get in the following configuration:
RGBBGRRGB. See:
“A CABLE CONFIGURATION TECHNIQUE FOR THE BALANCE OF CURRENT DISTRIBUTION IN PARALLEL CABLES”
by: San-Yi Lee

http://www.google.co.il/url?sa=t&rc...qoGIBQ&usg=AFQjCNEO9c8xtuL5SrAoblBF8_rq8XomrA

But, nevertheless this configuration will reduce the main conductor losses, the shield losses will be still elevated. You have to cross bond the shield, and better, to cross bond the conductors too. See:
ANSI/IEEE Std 575-1 988 “IEEE Guide for the Application of Sheath-Bonding Methods for Single-Conductor Cables and the Calculation of Induced Voltages and Currents in Cable Sheaths”
And: “HV Cable Earthing Link Boxes and Cable Sheath Voltage Limiters”

http://www.google.co.il/url?sa=t&rc...&cd=1&cad=rja&sqi=2&ved=0CBwQFjAA&url=http://

http://www.akpowersolutions.com/upl...2oGoAQ&usg=AFQjCNELJQt3d3qNXD1FPD5wO6ZID6boRA

 

[tanaru]

@ Skogsgurra and 7anoter4 thank you for your answers and for the references but I was wondering if other factors can contribute to extremely high currents in the screens of the paralled MV cables: 400A, 350A and almost 500 A in the screen of the last phase.
we know that the configurations is a one of the issues but here are more facts:
- the same configuration was set in two substations and all the laying conditions are the same, and the same cables and sections are used. The older substation is experiencing the issues of heating and the newer one is not.
I tried to see what is different between the two substation is the way the 110 kv underground cables that go out of the substations is. On the newer one a trefoil arangement ABC with adequate cross bonding instead of the older substation where the 110 kV arrangements is flat and from what I could see on the screen bonding scheme, they tied the screens only at the ends to the substations earth grid and no other earthing on the route. On the newer substation with trefoil configuration of the 110 kV cable and with the screns tied to the ground after the major sections there are no issues.
I want to stress out that the problem is on the screens of the MV cable that link the transformer to the MV cell and only on the substations abovementioned ant in the other one here is no heating . is it just a matter of time? And beside the laying configuration which is def wrong could the way the HV cables are laid in can factor in, or other issues related to the wye wye transformer that they choose? i want to know if we should pursue other "problems" beside the configuration.
skogsurra : On another note one of the solution proposed was to untie the screens of the MV cable at one of the ends but that at fault conditon it will create high induced voltage at the ungrounded end.
@7anoter4 thank you for the refs it will def help, I know the best configuration is the one you suggested, unfort this an older substation and I dont know how they engineer was done. The issue is on short runs max 48-54 meters, i dont see how you can cross bond sheath and conductors on such a short run
Hope i was not to confusing and got my quetions through/
Thank you,

 

[tanaru]

I am new to this forum. Can I go back and edit a post?
I submitted the post and didnt edit it.
Thank you

 

[ScottyUK]

Hi tanaru,

No, unfortunately it isn't possible to edit posts once they are submitted. Hopefully one day it will.

Welcome to the forum.

 

[davidbeach]

Nope, no edit capability.

 

[ghostbuster7]

Maybe this will give you a different perspective:

I just finished correcting a problem with 750 MCM. 1 single conductor per phase laid out in a flat configuration(not triangular).

The ground cable was located beside the outer phase only for approx.200' run.

Phase currents were 400 amps and the ground current was 110 amps