Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations IDS on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

Should 15KV system ground be connected to 280v ground

Status
Not open for further replies.

Paulpower

Electrical
Jul 19, 2008
10
Hi,

I am working on a substation project. The transformer is 13.8kv/208v DYn1, secondary solidly grounded. The primary is tapped from 13.8kv overhead line. 13.8kv system is high resistance grounded system.We have ground grid installed around the substation building. Metal clad Substation ground bus and secondary 208V neutral are bond to ground grid. The 13.8 kv overhead phase conductors are accompanied by a ground wire on the messenger line.

The directly buried 3c 13.8kv cable includes an integral bare ground conductor. If I use an outdoor cold shrink termination with a boot of 3 breakout for 3 phase conductors to tap from overhead line, shall I weld the cable shield to the integral ground conductor, then bond them to ground wire on overhead line? Generally, what shall I do on the other end of the cable which enter the metal enclosure of substation? Do I have to bond the ground conductor in the 13.8kv cable to the unit substation grond grid? Do I need to install a separate bare copper ground conductor with the cable when I bury the cable in a PVC conduit?

I am puzzled by the groud concept, please recommend some reference document also.

Thanks in advance.


 
Replies continue below

Recommended for you

The important things to consider are whether a fault on the high voltage equipment will cause an unsafe voltage to be exported onto the LV system, and also to avoid any differences in touch potential in the substation.
If you were in the UK (which I know you are not), the HV and LV earths would only be connected together if there is 1 ohm or less between them. If it's more than 1 ohm then the HV and LV systems have separate earthing with at least 10m separation. You have made the decision to connect them together in the substation. If the substation feeds a single customer you could alternatively earth the neutral at the service termination. To avoid any risk of the HV transformer and cable sheath earthing having differing potential you should bond all the HV earths to the substation main earth. Also the cable screens and separate earth conductor should be bonded. You should only need to run an additional earth conductor if the prospective earth fault current exceeds the rating of the integral cable earth conductor. There is no need to weld anything. With Coldshrink terminations the cable screens are brought out of the bottom of the insulator boot as wires for wire screened cable, or by attaching an earth cable to the copper tape screen using a constant force spring, if the cable has tape screens. All the various earths can be marshalled together on the pole using lugs to a bolt, or compression connectors/line taps.
Regards
Marmite
 
Hi Marmite,

Thanks for your detail explanation.

We can not earth the neutral at service termination because we have ground fault protecion by zero sequence CT at 208v bus which is required by CODE. We plan to run a separate bare copper cable from the substation neutral to the building as protection line, this line is bond to the substation grid and building grid. In fact, substaion grid and building grid are bond together since the substaion is only 2m away from the building.

Your reasoning for separating HV ground from LV ground is correct for it can prevent ground fault on HV side from affecting LV side. However, I do not have an idea how to implement it on this project. In this metal clad substation, the 15kv enclosure, transformer enclosure and 208 switchgear are metallically connected and a copper ground bus extend through the whole structure. This ground bus are bond to transformer secondary neutral, substation ground grid and all the non current carrying metal enclosure, transformer metal case, 15kv CB case, surge arrestor support etc. So HV ground and LV ground are bond together in the substation. When ground fault happened in transformer MV winding or in 15kv circuit breaker, the 208V and 15KV side metal enclosure, substation grid and building grid suffer from same ground potential rise.

My concern is the ground potential rise of the building ground grid. That means steel structure, metal pipe ,and equipment metal case in the building have a voltage over earth and this is undesirable.

I may make mistake somewhere. Your comments, suggestions are really appreciated.

Steven


 
From what you've described it sounds as though all metalwork in the building and all metalwork in the substation is bonded together. This is what you need to avoid touch potentials between metallic objects. If there is a fault, the whole metalwork in the building would theoretically rise above true earth potential for the duration of the fault, but as everything is bonded together the danger to personnel is averted.
Regards
Marmite
 
When a ground fault occurs on the HRG 13.8 kV system, the grounded phase becomes close to ground potential. There is a high voltage between the source neutral and ground, but I don't think you would see a high GPR at the substation. The ground fault current will be very low.

If the 13.8 kV and 208 V grounds are bonded together, there will be no voltage difference between equipment bonded to the 13.8 kV side and equipment bonded to the 208 V side. If the 13.8 kV and 208 V grounds are not bonded together, there will be a voltage difference between equipment bonded to the 13.8 kV side and equipment bonded to the 208 V side. If a 10 meter distance is maintained between anything bonded to the 13.8 kV side and equipment bonded to the 208 V side, then this may not be a problem. The voltage difference may be low for a 13.8 kV ground fault because the fault current is low.

I don't see the reason for isolating the two ground systems. I'm not used to dealing with HRG MV systems, so I may not be taking all issues into account. I'd be interested to know the reasoning behind separate grounding if there is a 1 ohm or more difference.

 
I would think that the higher the ohmic difference between the systems, the higher the voltage difference would be during a fault.

Bill
--------------------
"Why not the best?"
Jimmy Carter
 
the HV and LV earths would only be connected together if there is 1 ohm or less between them.
Actually, I don't know what is meant by "1 ohm or less between them."
 
jghrist & waross, you are right. I phrased it very poorly. What I meant to say was that the HV and LV earthing systems can be connected together if their combined resistance to earth doesn't exceed 1 Ohm. I've been trying to find where the 1 Ohm figure comes from. I think it's our Electricity Association Technical Std 41-24. The 1 Ohm figure and the separation distances are aimed at keeping the rise of earth potential on the LV system for a fault on the HV system to less than 430V. The maximum figure increases to 650V where the protection clearing time is 200mS or less.
My comments are general and not related in particular to HRG systems. Like you, I'm not used to dealing with these systems. However,the system as described by Paulpower sounds perfectly safe, especially so with the low ground fault current of the HRG system.
Regards
Marmite
 
Our practice is as follows:
1) If the 15 kV cables are three-phase cables you have to bond the both ends terminations to ground.
If the 15 kV cables are single-phase cables you have to bond the termination at the transformer end.
If you will bond both ends to ground you have to check if the shield induced losses do not produce cable overheating.
If you keep one end disconnected you have to check shield induced voltage build-up on the other end [usually the limit is 30 V].
2) You cannot maintain an isolation of the low-voltage grounding as the transformer enclosure is bonded to grounding grid and also to the star point of the transformer.
All around the substation, any equipment enclosure has to be bonded to grounding grid and so, the maximum fault voltage will be the touch potential [and as jghrist indicated, is low enough].
But you have to lay a grounding cable along and close to all supply cables in order to reduce the loop impedance, in order to facilitate the protection to clear-up as fast as possible, the short-circuit and also to reduce the residual grounding voltage on low-voltage side.
Regards
 
 http://files.engineering.com/getfile.aspx?folder=c3b9977a-28e7-463e-bda3-64538114a303&file=HV_LV_Grounding.jpg
Status
Not open for further replies.

Part and Inventory Search

Sponsor