×
INTELLIGENT WORK FORUMS
FOR ENGINEERING PROFESSIONALS

Log In

Come Join Us!

Are you an
Engineering professional?
Join Eng-Tips Forums!
  • Talk With Other Members
  • Be Notified Of Responses
    To Your Posts
  • Keyword Search
  • One-Click Access To Your
    Favorite Forums
  • Automated Signatures
    On Your Posts
  • Best Of All, It's Free!
  • Students Click Here

*Eng-Tips's functionality depends on members receiving e-mail. By joining you are opting in to receive e-mail.

Posting Guidelines

Promoting, selling, recruiting, coursework and thesis posting is forbidden.

Students Click Here

Jobs

Questions from reviewing SEL recording of fault

Questions from reviewing SEL recording of fault

Questions from reviewing SEL recording of fault

(OP)
This is a fictional event at a fictional generating plant, invented for training purposes. Any resemblance to real events, past or present is purely coincidental.

System Description:
345KV switchyard ---- 2 parallel GSU’s----- GCB---- Generator

GSU’s are each 750MVA each, 345kv/25kv.... grounded wye / delta.
Phasor diagram shown on slide 1. Phasors rotate CCW. CBA sequence.
Note the ordering of bushings is opposite ordering of the phases:
H1 GSU bushing corresponds to C phase and H3 to A phase.
X1 GSU bushing corresponds to C phase and X3 to A phase.

GCB is Cogenel low-voltage pressurized air generator circuit breaker with 250,000A interrupting capacity.

Aux transformer also taps off isophase bus between GSU and generator.

Generator neutral is grounded thorugh a single phase neutral grounding transformer 25kv/240vac with 0.154 ohm resistor on the secondary.

Event Description:
Plant trip, loss of auxiliary buses – fire in GSU 2A (extinguished quickly). Reliefs lifted. Tank ruptured in certain places.

Relays received: 87T1 (differential on main transformers)

87T1 relay results in lockout which tripped 345kv breakers in switchyard on transformer high-side and generator circuit breaker on low side.

DC Electrical testing of transformer (determinated on hi/low side bushings) showed ground on low side (high side is already grounded through H0 bushing). TTR applied to hi-side phases A and B was normal, TTR applied to high side of phase C was indeterminate (could not get a reading).

Internal inspection of transformer showed most damage on C phase... lesser damage on B and no damage on A. The damage on B appears to be mechanical (the entire transformer shook during the event as evident by external inspection). Damage on C appears electrical, including several locations where high side coil is pushed up above the rest of the winding to expose bare copper (although no obvious arcing at these locations).

Slide 2 shows switchyard recording. Voltage collapses on C phase but relatively unchanged on A and B (suggests C phase to ground fault). It seems like C phase current increased by approx factor of 10, B phase by factor of 2 and A phase remained constant (until fault was cleared on high-side 4.5 cycles later, at which time all go to zero). The working hypothesis is failure on C phase high side winding of GSU.
Question 1: Why would phase B phase current double on a C phase high side fault ? What other fault scenarios would match these measurements? Could an H0 neutral voltage shift resulting from the fault cause this?
Question 1A:3 numeric values are listed next to the vertical axis for each plot of slide 2... the relationship of these numbers to the plot is not clear. Does anyone recognize what these numbers are describing?

Slides 3 thru 6 are extracted from an SEL300G relay sensing generator voltage (on generator side of generator circuit breaker) and generator currents (at generator terminals).

Slide 3 is labeled with some of the more obvious aspects.
Note from slide 1 phasing diagram we can infer that phases transformer B and C low side terminals are both attached to the leg of the low-side delta winding which is associated with the assumed fault on the high side phase C winding. So in general for fault on high side C phase, we expect collapse of Vbc (confirmed slide 4).
Labeled times on slide 3:
T0 = Time of fault initiation
T1 = Time of opening of A phase GCB main contact at current zero. We expect A to interrupt first, and we expect it to interrupt at current zero (it checks out). Note that a much smaller A phase current persists after this due to current in interrupting contact/resistors as expected.
T2 = Time of opening of B and C phase GCB main contacts at current zero (as expected). Note that a much smaller B and C phase current continues due to current in the interrupting contacts/resistor.
T3 – Current in B/C phase interrupting contacts stops (at current zero as expected).

Now it is interesting to look at generator voltages (phase to neutral on slide 3 and phase to phase on slide 4) during the period between T0 and T1. Phase to neutral voltages (slide 3) show at least 4 sharp jumps of Vcn toward zero and perhaps one sharp jump of Vbn towards zero. Whenever these jumps on Vbn or Vcn occur, another similar jump occurs on the other one (Vcn or Vbn) such that the phase to phase voltages (slide 4) do not show any sharp jumps (except Vbc sharp jump down upon fault initiation and sharp jump up to normal after final fault clearing).

I suppose these jumps of Vbn and Vcn toward zero indicate a short to ground occurring intermittently somewhere on the 25kv system or transformer low-side winding. The currents are not significantly effected (as expected for single ground since generator neutral grounding resistor has high impedance when transformer to generator side...I think it should limit ground current to around 9A). The logical assumption is that this corresponds to intermittent ground fault on GSU due to lots of stuff going on in transformer (result of initial fault, not cause).
Question 2a: Do you agree these collapses of Vbn, Vcn indicate intermittent ground faults?
Question 2b: Does intermittent collateral-damage ground faults of GSU low-side this seem like a reasonable explanation for the intermittent ground faults, or do you see any reason to suspect a ground fault may have occurred elsewhere such as generator circuit breaker? (we will do electrical testing of other components but not sure what degree of inspections).

Now lets examine neutral/ground current. Slides 5 and 6 are “neutral” and “ground” currents reported by the relay. I have questions about each.

For “ground” current – I know exactly what it represents because I can recreate the ground current trace by simply adding together the individual reported phase currents. I expect this may not be particularly accurate for due to CT errors as well as the low sample recording rate (16 samples saved per cycle). I attribute the ac present on Iground prior to the fault to CT errors. Question 3:But what about the dc present after the fault? To what do we attribute that?

For “neutral” current – I don’t know what it represents. As shown in slides 7 and 8, the return current from the wye neutral point does not even pass through the relay. The neutral current input to the relay is unconnected. I can think of two possibilities for what the In signal represents:
A – calculated from phase current using more precise info than 16 samples per cycle
B – Garbage based on reading an input that has nothing connected to it
Question 4: How do you think the SEL400G generates this neutral current signal (A or B or something else)? If A is the case, then why doesn’t the apparently slowly decaying DC show here that showed in the ground current?

Question 5:I assume the SEL relay current input simply has a current sensing resistor and just measures voltage across it... correct?

Slide 8 and 9 are my attempt to prove to myself why we expect higher current on GSU low side C phase than on GSU low side B phase during the period between T0 and T1, given that they both are associated with presumed fault on C phase high side. To me it seems a logical result if the fault current is much more inductive than the normal current. No questions on this aspect, but open to comment.

=====================================
(2B)+(2B)' ?

RE: Questions from reviewing SEL recording of fault

(OP)

Quote:

or “ground” current – I know exactly what it represents because I can recreate the ground current trace by simply adding together the individual reported phase currents. I expect this may not be particularly accurate for due to CT errors as well as the low sample recording rate (16 samples saved per cycle). I attribute the ac present on Iground prior to the fault to CT errors. Question 3:But what about the dc present after the fault? To what do we attribute that?

For “neutral” current – I don’t know what it represents. As shown in slides 7 and 8, the return current from the wye neutral point does not even pass through the relay. The neutral current input to the relay is unconnected. I can think of two possibilities for what the In signal represents:
A – calculated from phase current using more precise info than 16 samples per cycle
B – Garbage based on reading an input that has nothing connected to it
Question 4: How do you think the SEL400G generates this neutral current signal (A or B or something else)? If A is the case, then why doesn’t the apparently slowly decaying DC show here that showed in the ground current?
For question 4- I am leaning toward B (IN signal here is garbage) basedon the fact that we have nothing wired to that input and the fact that there is no change in neutral current behavior before/during after the fault.


For question 5 - I have attached additional graph to this message showing that each of the phase current traces from the SEL has a slowly decaying dc component after the event. The weird thing is they don't seem to be decaying toward zero, but instead toward some negative dc value.

=====================================
(2B)+(2B)' ?

RE: Questions from reviewing SEL recording of fault

In SEL raw events you get the currents without null error correction being applied. Capture a raw event with no current and you'll find three distinctly different current values.

Not that I can say I'll have time to thoroughly look at it, but can you post the EVE or CVE files from the SEL relay?

RE: Questions from reviewing SEL recording of fault

(OP)
Thanks. I'll check if files can be posted.
Can anyone explain briefly what is this null error correction?

=====================================
(2B)+(2B)' ?

RE: Questions from reviewing SEL recording of fault

Actually null error correction is perhaps not the most accurate description. In the raw event you are seeing what the sensors in the relay measures. (I believe that the current channels have an aux CT that is connected to a resistor and the A/D reads the voltage across that resistor, the impedance between CT terminals of the same phase on the relay should measure close to a dead short.) In the raw events, each current channel is the unfiltered output of the A/D. Before the relay does anything with those values they are run through a filter that eliminates all of the dc, including standing offset, so that the filtered events show no dc offset.

Red Flag This Post

Please let us know here why this post is inappropriate. Reasons such as off-topic, duplicates, flames, illegal, vulgar, or students posting their homework.

Red Flag Submitted

Thank you for helping keep Eng-Tips Forums free from inappropriate posts.
The Eng-Tips staff will check this out and take appropriate action.

Reply To This Thread

Posting in the Eng-Tips forums is a member-only feature.

Click Here to join Eng-Tips and talk with other members!


Resources