Circulating Currents in 13.8 KV Parallel Feeders 3

[sishah2011]

We have our two 13.8KV substations (A & B), connected through 2 x cable feeders. Each feeder comprised of 3 x single core power cables per phase.
Electrical power is being fed to Substation B from Substation A (Substation A is receiving electric power from 3 x Steam Turbine driven generators, running in synchronism).
Since commissioning, we are experiencing circulating current in both feeders, which is causing tripping of feeder breakers on Ground Fault protection. The circulating current usually remains around 25 Amps, however, sometime it increases to 38 Amps without any change in load or operating condition. We are not sure what is causing this increase. Appreciate if someone can advise us the possible cause(s) and its remedial action.
Regards.

 

[Mbrooke]

Could be many things from taps differing on the transformers at the supply substations to power flowing through the 13.8kv system due to a high impedance on the 115kv system (ie, ex) to the supply stations not being electrically connected to each other on the primary. Any phase angle differences between the supply stations will cause current on the 13.8kv system.


Do you have a single line?

 

[crshears]

What Mbrooke said; a single line diagram would really help in providing informed responses.

For instance: are the two feeders in galvanic parallel? In other words, do they [a] both connect to a common bus at both ends, or does one end have a common bus while at the other end each feeder connects to an individual transformer winding, or [c] . . .

CR

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

 

[waross]

This sounds like a daisy chain connection.
Generation feed sub A, sub A feeds sub B.
If the power feeds through sub A without a transformer,
then the problem will be in either sub B or in the cable feeding sub B.
Look for a line to neutral load on sub B.
What is the charging current for the feeders?
Do you have wye connected capacitors at sub B?
Failing capacitors on one phase may cause an issue.


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

 

[rcw retired EE]

How is ground fault current measured? Is there a zero sequence CT around all conductors, or is it a residual connection that sums the three (six?) phase CT's?
With two parallel feeders, there are 6 wires (cores) carrying current. All six cores have to go through the same zero sequence CT to get a proper ground fault current reading. (I am assuming that both feeders are connected to the same bus at both ends).

Are these high voltage shielded cables with the shields grounded at both ends? Are your CT's reading current in the shields? Are shields properly routed back through the zero sequence CT's to prevent reading stray ground currents in the shields as ground faults?

Are you sure it is circulating current and not some unusual neutral current circulating between grounded wye transformers, generators, or loads upstream and downstream? If the circulating current is confined to the two feeders, the currents would be 180 degrees out of phase in the two feeders. A properly connected fault sensing scheme would add the two currents to cancel out circulating currents.

Again, it all depends on your one line and how the equipment, cables, CT's, transformers, generators and loads are connected. If there is a grounded Y transformer connected to Bus B and Bus A is fed from grounded wye generators or transformers, circulating ground currents could occur.

Some ideas to think about.

 

[waross]

Are these feeders in series or in parallel?

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

 

[sishah2011]

Thank you Mbrooke for your response. Single Line Diagram of circulating current loop is attached.

Please note that the Generation is at 13.8 KV, which is feeder one Substation A. There are 2 separate feeders connecting substation A and Substation B, forming a ring/loop in the network. Each feeder has 9 single core 630sq.mm. cables (3 per phase).
The currents disappear fully as soon as the loop between 2 parallel feeders breaks. This indicates that these are circulating currents and not actual fault. Further, this is happening at 13.8KV ring only, where there is no direct load, all load is thru DELTA -WYE step down Transformers.

 

[sishah2011]

Thank you rcwilson for your response.
We have 2 separate feeders each having 9 single core 630sq.mm. cables (3 per phase). Ring core or Zero sequence CT is installed on both sides of each feeder that is 02 on each feeder ie 04 in total, at both ends of two feeders. The current unbalance is through phase conductors as the calculated Io (summation of phase currents) is also same as sensed by Ring core Current CT.
Opening of the ring i.e. one coupler breaker or one feeder breaker results in elimination of circulating currents entirely i.e. reading becomes zero. Moreover, the current tends to increase and decrease in proportion to load generally, however, some abnormal behavior ( jumping form 26 A to 37~38 A and vice versa) i.e. sudden increase and decrease has also been noticed.
No, the armour and shields are not through the CT and this current that CT is reading is surely from phase conductors. The armor /shields are connected only at one side directly to ground. The second side is insulated with proper sleeves and not connected to ground
The currents disappear fully as soon as the loop or ring between 2 parallel feeders break as described above. This indicates that these are circulating currents and not actual fault. Further, this is happening at 13.8KV Ring only, where there is no direct load, all load is thru DELTA -WYE step down Transformers. Moreover, there is no reflection of circulating currents either on Source or Generators side or load side.
The generator neutral is connected thru neutral earthing Transformer with 50A rating. The neutral is not used in 13.8KV system.

 

[sishah2011]

Thanks crshears for your response.
SLD is attached, please.

 

[sishah2011]

Thank you waross for your response.
We have 2 separate feeders, each having 9 single core 630sq.mm. cables (3 per phase). Ring core or Zero sequence CT is installed on both sides of each feeder that is 02 on each feeder i.e. 04 in total, at both ends of two feeders. The current unbalance is through phase conductors as the calculated Io (summation of phase currents) is also same as sensed by Ring core Current CT.
The currents disappear fully as soon as the loop between 2 parallel feeders breaks. This indicates that these are circulating currents and not actual fault. Further, this is happening at 13.8KV Ring only, where there is no direct load, all load is thru DELTA -WYE Transformers.
The armour and shields are not through the CT and this current that CT is reading is from phase conductors. The armor/shield are connected only at one side directly to ground. The second side is insulated with proper sleeves and not connected.

 

[RRaghunath]

The cable route length is ~600m from the SLD, which is not much.
The unbalance in phase currents could only be due to difference in laying of cables of one interconnector from the other. This difference gives rise to difference in cable reactances which can cause circulating currents.
Could you review how the cables are laid please, i.e. are they in trefoil formation or in flat formation; is there difference in route or or the two sets of cables following the same route etc.

Rompicherla Raghunath

 

[waross]

I was about to ask the same basic questions as Raghunath. The feeders may be the same length and yet have very different impedances.

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

 

[sishah2011]

Thank you RRaghunath for your comments.

Circulating currents due to difference in cable reactances because of difference in laying condition is understandable. However, our concern is intermittent change in magnitude of current, irrespective of load or operating conditions.

The cables are following same route, generally laid in overhead cable trays, however, a small portion of the cable is directly buried in trench. Within substation and underground, it is in trefoil, whereas, on external overhead cable trays they are laid in flat formation. Please refer attached photographs.

 

[krisys]

Here we should be clear whether the OP has a concern about the three phase balances circulation current or the earth fault current which is circulating.

From the SLD I can make out that the 13.8kV system is quite a big industrial system. In such as large system, a circulating current of the order of 25A (amounting to 0.5MVA! in a system having 100+ MW generation capacity) is negligible. It could be owing to the difference in the cable impedance. When the system is loaded, the circulating current would even out.

But, if there is a circulating earth fault current that is something serious. That needs more due diligence. 13.8kV system neutral earthing, neutral point shift, 3rd harmonics, alternator pole pitch (what fractional pitch the alternator winding is?) etc. etc.

May be even wrong measuring inputs (such as CT polarity) possible.

 

[Mbrooke]

For curiosity, do breakers R11 and R14 need to be kept normally closed?

I agree with Krsiys, is this phase or earth fault current? Are all the generators identical?

 

[waross]

You may be at an operating level very close to the set point of a PF controller. You may be seeing a capacitor bank switching on or off, and changing the PF of the current in such a way as to change the measured ground current.

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

 

[sishah2011]

Yes, both R11 and R14 are bus couplers and remains closed all the time.

Since, the circulating currents become zero as soon as the loop or Ring is opened (by opening one bus coupler or opening one interconnection feeder), therefore, it indicates that these are circulating currents and not actually the ground fault currents.

All generators are identical at 1800 rpm, except their ratings are as follows:

2x 24 MW (STG 1 & 2)
1x 10 MW (STG 3)

 

[waross]

Please verify that each trefoil bundle contains all three phases.
Also verify the position of the phases in the flat configuration.
For flat configuration;
A-B-C---C-B-A is acceptable.
B-C-A---A-C-B is acceptable.
But both feeders must be the same.
A-B-C---C-B-A on one feeder and B-C-A---A-C-B on the other feeder is not acceptable.

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

 

[sishah2011]

We have PF control available on generating sources only. PF is being maintained in the range of 0.83 to 0.87 on all machines. There is no capacitor bank or PF control/ improvement equipment available on load side which is switching ON & OFF at times.

Each trefoil bundle contains all the three phases; but at some points the trefoil arrangement is not maintained due to un-even surface, space limitation or sharp bends on the laying routes. Flat configuration on termination points is A-B-C--or-C-B-A on both sides of same feeder.

 

[stevenal]

Have you performed thermal scans lately? Unbalanced contact resistance or connection resistance at any of the four motor operated disconnects connected to the cables will cause the currents to unbalance. This unbalance will go away when the loop is opened, and could be intermittent as connections chatter. Suggest performing thermal scans with one cable out of service at time.