Title: Capacitor Switching
1Capacitor Switching continued
- A 13.8 KV, 5000KVAR, 3ph bank,NGr
- Source Gr, inductance1 mH
- Restrike at Vp
- 1- c5/(377x13.8?)69.64µF
- 2- Zv1000/69.643.789O
- 3-Ip2v2x13.8/(v3x3.789)5.947 KA
- 4-f0603 Hz
2Subsequent Occurences
- C.B. Interr. H.F. current at its zero
- 3Vp on Cap.at most 4Vp across C.B.
- If 2nd B.D. occur, 2nd Osc.
- I doubles, Vc from 3Vp to -5Vp
- If C.B. opens, at most 8Vp across C.B.
- further ESCALATION possible
- Rs seq.Restrikes, Cs subseq.Clearing
- Variation of VCB shown in FIG
3Other Restriking Phenomena
- Seq. restrik. clearing of C.B.
- O.V.s (even inductive load)
- low p.f. results in Vp at zero current
- Dominant fTRV?0/2? 1/2?vL2C Fig.
- Several KHz if C stray Cap. small
- reigniting effect heuristic approach
- Superposition of Vs(0) effect Vc(0) effect
- IVm/?(L1L2) sin?t
- I1(0)I2(0)I(0)It(during short duration)
- I Vm/L1L2
4 Restriking cct
reignition after isolating inductive load
- Equivalent CCT
- Short interval ?t0
- source sub. Battery
- I1I2 rising ramp
- as current restab.
- Ramp component1 Vs(0).t/L1L2
- f011/2? x vL1L2/L1L2C
5Discussion Continued .
- Now superposing effect of the Capacitance
- Initial Voltage Vc(0) at reignition
- Surge impedance
- Z0vL1L2/c(L1L2)
- The Osc_component2
- IcVc(0)/vL1L2/c(L1L2)
- fraction L1/(L1L2) of it pass L2
- fraction L2/(L2L2)of it pass L1
6Formal Solution
- Two Loops Equation
- L1 dI1/dtVc(0)1/c?(I1-I2)dtVs(0),
(1) - Differ.(1) d?I1/dt?I1/L1C-I2/L1C0,
(2) - Vc(0)1/C?(I1-I2)dtL2 dI2/dt, (3)
- L.T. of Eqs (2) (3) respectively
- (s??1?)i1(s)-?1? i2(s)s I1(0)I1(0)
(4) - ?2?i1(s)-(s??2?)i2(s)sI2(0)I2(0)
(5) - I1(0)Vs(0)-Vc(0)/L1, I2(0)Vc(0)/L2
- ?1?1/L1C, ?2?1/L2C
7Discussion of Formal Solution
- Solving Eqs (4) (5) simultaneously
- yields current 2 comp.s
- 1- a ramp component
- 2- a damped oscillating component with
v(?1??2?) v(L1L2)/L1L2C - In first method assumed I2(0)0
- true if Reignite at peak of TRV
- sys Gr Neutral, this fault High Cur. cause
Damage - All sys Gr directly or through some stray C
- So ARCING Gr Next Discussion
8ARCING GROUND
- L-Gr faultarcph Gr stopReig.repeated
- Fig of simple model without sources
- C1 ph to ph cap.
- C0 ph to Gr cap.
- N at Gr potenial
- A to Gr ?shift V Ep1pu
- VA neg.peak at F. instant
- Shift shown in phasor D.
9Discussion of Arcing Gr. results
- C0 of A dischargeN rise to Ep B,C rise
- C1C0 share charges at B and C
- not at once to Diagram values
- Charge Conservation
- V(C0C1)1/2C0Ep3/2C1Ep
- VEp/2(C03C1)/(C0C1)
- VBVC rise to 3/2Ep osc. C0,C1Ls
- CCT shown in Fig.
10Discussion Continued
- Equivalent CCT
- f01/2?v3L(c0C1)
- Z0v3L/4(C0C1)
- VBVC? above 3/2 Ep
- Ip(3/2Ep-V)/Z0
- subs. for VZ0
- Ip2Ep(C0/C0C1)x
- v(C0C1)/3L
- ? IB pass out of node
B?dividedIc1,Ic0 - ? Ic1 pass the arc its frac.
- C0/(C0C1)
- ? Also a 60Hz current due to VA through CN
11What Happens Afterward?
- Depends on Arc behavior
- 1-Iarc till next p.f. zero in 1/2cycle
- 2-Iarc immed. Ceased in 1st zero of the IHF
- If 1, occur after ½ cycle
- since A is still at Gr levelN in -1pu
- 3ph Phasors in next Fig a
- N keeps -1pu and after 1/2cycle
- VA rise to -2pu Fig b
12Phasor Diagram after interrupting a reignition
current
- Fig a b
- If Reignite again
- similar shift ?V2pu
- Rather than 1pu
- Transients higher
- N -1pu? 1pu
- Swings of A B
- Then seq. can repeat
13continued on case 1 then Case2
- after 1/2cycle A,Gr however at peak
- BC instant. -1.5pu to Gr
- If now arc interrupts VN change -1pu
- After ½ cycle exactly as last fig b
- Case 2 If interrupt at 1st zero of IHF
- occur at point P of curves
- 1-Vc1,Vc2 attain Vp
- 2-Arc extinguish C1 s rise to VLL(VACVAB1.5pu)
- Vp1.5Ep(1.5Ep-V)3Ep-V
14Continued on Case 2
- i.e. Vp-3/2Ep3/2Ep-V across arc path
- ? N has corresponding Disp.
- ?Phasor Diag.Fig.a?
- ?different from 1st fig
- ?Then arc interruption
- at p.f. current zero
- when VmaxVAG,
- 1/2cycle later, situation of figb
15Discussion
- transient following next reignit. Is greater
- Arc is interrupted in 1st H.F. zero I
- neutral displacement increases
- increase in energy trapped on zero seq Cap
escalate the voltage - How to suppress arcing Gr OVs
- 1- an appropriate reactance in neutral
- 2- Peterson coil , sensitive for fault
detection -
16Assignment No. 3
- Ques. 1
- C12µF,C2.38µF
- L800µH,R5O
- VC1(0)75 KV
- S closes, compute
- 1-Max energy in L?
- 2-t0 instant current flow in c2
- 3-Vc1(t0)?
- 4-Max of Vc2 ?
17Solution of Question 1
- Z0vL/C1 v800/220O
- S close Ipeak(undamped)75/203.75KA
- ?Z0/R4(fig 4.4)?
- Ip0.83x3.75 3.112 KA
- Emax0.0008x3.1122/24.12KJ
- I flows in C2 when reverses?fig4.4 at t3.15
- or t3.15vLC1126µs, (T40µs)
fig4.6 ? Vc1-0.69x75-51.75 KV
18Eq. CCT (Diode goes off)
- series RLC CCT
- At this instant
- Vc1(0),Vc2(0) required
- CCT diff. Eqs employed
- Solved for Vc2
-
19Solution of Question 1 continued
- 2nd1/2cycle,Vc1(0)51.75KV,Vc2(0)0,I(0)0
- C1C2 now in series CeqC1C2/(C1C2)
- Z0v(800x2.38)/(2x0.38)50.05O ?50/510
- Now for series C1,C2,R,L we have
- Vc1Vc2RIL dI/dt
(1) - I-C1 dVc1/dt-C2 dVc2/dt (2)
- Vc1Vc1(0)-1/C1?IdtVc1(0)C2/C1Vc2 (3)
- Solving Eqs 1,2,3 for Vc2
- d?Vc2/dt?R/LdVc2/dt(C1C2)/LC1C2Vc2
-
-Vc1(0)/LC2 - s?vc2(s)-sR/LVc2(0)-R/LVc2(0)sR/Lvc2(s)-v(c1C2
)/LC1C2xVc2(s)-Vc1(0)/sLC2
20Solution of Q1 ..continued
- I(0)0?Vc2(0)0,TvLc1c2/(c1c2)
- vc2(s)
- -Vc1(0)/LC2 x 1/s(s2Rs/L1/LC)
- where CC1C2/(C1C2)
- ?10?fig4.7 Vpeak1.855
- 1pu-T2.Vc1(0)/LC2
- -C1/(C1C2)Vc1(0)-2/2.38 x (-51.7)43.48KV
- Vpeak1.855x43.4880.7 KV
21Question 2 , C.B. opening resistor
- C.B. clear 28000 A sym fault
- R800O, Cbus0.04µF,
- Vsys138KV,f50Hz
- 1-Peak of TRV?
- 2-energy loss in R (it is 2 cycle in)?
22Q2, Solution
- X138/(v3x28)2.8455, L9.1 mH
- Without R, TRV2x138xv(2/3) 225.4KV
- Z0vL/Cv0.0091/0.04x10-6477O,
?800/4771.677 - fig4.7? 1.38x 225.4/2155.5 KV
- Energy dissipated in 2 cycles
- VTRV112.71-exp(-t/2?)x sin.. cos
- Integrating PlossVTRV?/R over ?t0 to ?t4?
- for a (1-cosine) wave(1-cosx)?1cos?x-2cosx
23Q2 continued
- ?(1-cos?t)? dt
- 3/2t1/(4?) sin2?t-2/?
sin?t6?/? - t0,4?/?
- Energy loss if TRV was a 1-cosine wave
- V?/Rx6?/?112.7?/800x6?/3140.9526MJ
-
(952.6KJs) - However (1-cosine) is deformed and assuming
- to be halved Energy loss952/2476 KJs
24Question 3
- 246 KW Load on a 3ph S.G., 13.8 KV
- p.f.0.6 lag(load parallel RL) C?pf1
- loadcap switch as a unit
- S opened, Vpeak across C.B.?
- Zs negligible
25Q3, solution
- V?/R246 KW, R(13.8/v3)?/246/3774O
- p.f.0.6, F53.13
- tanF1.333R/X? X580.6O,L1.84 H? Xc580.6
- C1/(314.15x580.6)5.48µF
- switch open at Is0, Vs0, when ILIC
- are at peak (opposite sign)
- Ipeak13.8v(2/3) /580.619.41 A
26Q3, solution continued
- CCT a parallel RLC
- -cdVc/dtVc/R
- 1/L?VcdtIL(0)
- d?Vc/dt1/RCdVc/dt 1/LC 0
- s?s/RC1/LCvc(s)(s1/RC)Vc(0)Vc(0)
27Q3 continued
- Vc(0)-Vc(0)/RC-IL(0)/C, Vc(0)0
- ? vc(s)-IL(0)/cs2s/RC1/LC
- without R ?-IL(0)/c x 1/(s2?02)
- and Vc(t)-IL(0)/C?0 sin?0t
- -IL(0) vL/C sin?0t
- vL/Cv1.84/5.48579.5O,
- IL(0)19.41 A? Vpeak11.27KV
- ?R/Z01.333? fig 4.4 0.6x11.276.76KV
28Transformer Magnetizing current
- Mag. Inrush ? transient
- Im0.5 to 2 rated, non-sinusoidal
- Distortion B in core
- Instant of energizing Residual Flux can
cause Inrush - Continue several sec, small Transf.
- several min, large Transf.
- 1000KVA, 13.8KV load 42 A, Inrush peak 150 A
- A DC declines and finaly mag. current
29Ferroresonance
- series Resonance
- Very H.V. across CL (series LC)
- if excited Natural Freq.
-
30Example of Transformer Ferroresonance
- simulate nonlinearity of core
- LdF/dtA exp(-I/B)I0, LA IB, LA/e
- A 13.8 KV step down T. resonance
- with cables in primary CCT, A8H,B1.76 or
LdF/dt8exp(-I/1.76) - if 60Hz res. Occur in L4H,
- C1/??L1.75µF
- neglecting resistance
- L dI/dt1/C?I dtV sin(?tF)
- 8 exp(I/1.76) dI/dt
- 106/1.75?I dt13.8v2/3 sin(377tF)
- A nonlinear Diff. Eq. solved for I and then V,
F0,45