Safety Aspects in Substation Voltage Uprating
Safety Aspects in Substation Voltage Uprating
Safety Aspects in Substation Voltage Uprating
1250
'
at a higher nominal voltage with the same clearances and changes of only some equipment, for example, transformers or circuit breakers. Voltage uprating of older substations or building new substations with smaller dimensions makes an economic improvement possible, but it also raises several concerns, mainly with personal safety, insulation coordination, insulator performance under polluted conditions, environmental (fields, RI, corona) effects, short circuit withstand, and maintenance. Personal safety concerns are brought up by the decrease in clearances from those presently specified. Personal safety, as the most important issue of substation uprating is the objective of the discussion in this paper. Since basic dimensions within the substation are determined by the insulation coordination aspects, it is helpful to start with a brief comment on design procedures from the viewpoint of insulation coordination. INSULATION COORDINATION Choice of Insulation Level Design of HV substations is usually done according to industry or internal utility standards which directly specify the clearances and BIL's. This bypasses the need for detailed insulation coordination studies to determine these values. While the NEMA and ANSI standards generally [for example, 1, 3, 41 offer a choice of insulation levels for EHV equipment (higher than 242 kV), they do not provide such a choice for the equipment below 242 kV. Some of the standards go even further and tie the BIL and clearances together and associate them directly to the rated voltage. A good example is Table 1, reprinted from NEMA SG6 [l]. The result is just one value of BIL and minimum clearance for a given system voltage below 242 kV. Even though the power frequency voltage stress has to be taken into consideration for insulation design, the equipment BIL and the required substation clearances, are in general, given by the maximum overvoltage, not by the system voltage. Within the surge arrester's zone of protection, the maximum overvoltage is limited by the surge arrester protective effects. Thus, operationally both the required equipment BIL and the substation clearances are determined by the applied arrester. Since it is possible to use various arresters for the same system voltage, various insulation levels should be possible as well. International standards [5] extend the choice down through the HV range. This supports the above reasoning and indicates possible changes in our practices. Indeed, if modern surge arresters are applied and the substation design is based on an insulation coordination study rather than on standardized practices, the acceptable BIL and minimum clearances resulting are substantially smaller than those given in the present standards (see Table 1). This is not only a result of a theoretical study; it is supported by practical experience. Some utility substations have been indeed uprated and successfully operated with BIL's and clearances well below those standardized. For example, by the judicious application of surge arrestors and with the change out of some major equipment, several 115 kV substations were uprated to 230 kV and operated with insulators and disconnects rated 550 kV BIL instead of the usual 900 kV and with clearances to ground corresponding to 115 kV [5, 61. Similarly a 69 kV substation has been operated at 115 kV with 350 kV BIL insulators and
Insulation levels of HV substations per present practices [ l ] appear relatively much higher than those of EHV substations. Expressed in per units of rated voltage (peak value, line to ground), the 550 kV BIL of a 115 kV** substation translates to 5.6p.u. while the 2050 kV BIL of a 765 kV substation is only 3.13 p.u. The reasons for this situation stem more from historical development than from existing design requirements. The practices with HV substations were developed some decades ago when limitations of available knowledge were compensated by higher insulation margins and when economic pressure on substation cost was not as high as it is today. To buy better margins by larger clearances was cheap with 115 kV. The economy is quite different with 765 kV substations nowadays. Even though the comparison of relative insulation levels does not cover all the aspects, (for example, the clearances with EHV are dictated by the switching surge rather than the lightning surge), it serves here to illustrate the possibilities to decrease the insulation in the HV range. Indeed, the application of modem metal oxide arresters could enable a decrease of insulation levels by up to two steps for the substations up to 230 kV while retaining the same or better margins than with 765 kV, as shown, for example, in [2]. A decrease in the insulation requirements enables the substation engineer uprate substations for operation
*Edited by - J. Panek, GE,(W.G. El Chairman); M. Rockwell, B.C. Hydro; A.M. Sahazizian, Ontario Hydro. **For the sake of consistency, normal 60 Hz voltages are used throughout this paper except where a voltage is referenced from another document.
9 1 SM 511-6 PWRD A paper recommended and approved by the IEEE Substations Committee of the IEEE Power Engineering Society for presentation at the IEEE/PES 1991 Summer Meeting, San Diego, California, July 28 August 1, 1991. Manuscript submitted January 25, 1991; made available for printing June 25, 1991.
0885-8977/92/$3.0001992 IEEE
1251 Table 1 Outdoor Substations - Basic Parameters Recommended Phase Spacing Center-to-Center- Inches (Meters) Minimum Bus Supports. Metal-to-Metal Rated Withstand Voltaee Distance Between Vertical Brk Horn Gap Horizontal Disc. Switches Impulse 60 Hz kV rms Rigidly Supported Rated 1 . 2 5 ~0p Wet Energized Ground Clearance Switch & Break Power Fuses Line Max. Volt Wave 10 Conductors Inches (Meters) Expulsion Disc. Non-Expulsion Type SecMXls Inches (Meters) Recommended Minimum T v ~ Buses e Switches Riaid Conductors kV rms kV Crest 6 72.5 350 145 31( .79) 29 ( .74) 25( .64) 84(2.13) 72(1.83) 60(1.52) 7 121 550 230 53(1.35) 47 (1.19) 42(1.07) 120(3.05) 108(2.74) 84(2.13) 8 145 650 215 63( 1.60) 52- 1/2( 1.33) 50( 1.27) 144(3.66) 132(3.35) 96(2.44) 58( 1.47) 168(4.27) 156(3.96) 108(2.74) 9 169 750 315 72( 1.83) 61 - 1/2(1.56) 10 242 900 385 89(2.26) 76 (1.93) 71(1.80) 192(4.88) 192(4.88) 132(3.35) 11 242 1050 455 105(2.67) 90-ln(2.30) 83(2.11) 216(5.49) 216(5.49) 156(3.96) 192(4.88) 12 362 1050 455 119(3.02) 106 (2.69) 84(2.13)* 240(6.10) 104(2.64)* 13 362 1300 525 14 550 1550 620 124(3.15)* 15 550 1800 710 144(3.66)* 300(7.62) 16 800 2050 830 166(4.22)* Recommended Minimum Clearance Between Overhead Conductor and Ground for Withstand Personal Safety S.S. Crest kV Feet (Meters) ll(3.35) 12(3.66) 13(3.96) 14(4.27) 15(4.57) 16(4.88) 18(5.49) 650 759 808 898 982
Note: For insulator data, see the NEMA Standards Publication for High Voltage Insulators, Pub. No. HV 1-1973, * Ground clearance for voltages 362 kV and above are selected on the premise that at this level, selection of the insulation depends on switching surge levels of the system. The values were selected from Table 1 of IEEE Transaction paper T-72-131-6 (Vol. No. 5, page 1924) which is a rep of the Transmission Substation Subcommittee. For additional switching surge values refer to the above noted paper Note: Table reprinted from NEMA Publication SG6, 1977, App A. p. 2.
disconnects instead of 550 kV. M ~ substations ~ ~ are in operation under very adverse conditions: one of the uprated 230 kV substations is at 6000 feet elevation and the 115 kV substation is in an industrial environment with heavy pollution. Table 2 Preferred BIL Series Rated Maximum Voltage kV rms 72.5 121 145 169 242 Preferred BIL's kV Crest 350-250 550-450-350 650-550-450-350 750-650-550-450 1050-900-750-650-550
similar to~Table 1~ should not ~ , be interpreted as restrictive. It is to be noted that the ANSI insulation coordination standard [3] defines just a series of preferred values of BIL for medium and high voltage class (paragraph 5) and does not associate the values with the rated voltages. Such an association is left up to the user.
Table 3 BIL and Minimum Clearance BIL kV Crest 250 Minimum Phase-to-Ground Clearance in (m) 17 (0.43) 25 (0.64) 42 (1.07) 50 (1.27) 58 (1.47) 71 (1.80) 83 (2.1 1)
Note: 1. The clearances are based on NEMA SG6. 2. These are operational or electrical clearances, not safety clearances. 3. The phase to phase clearances make no allowances for devices designed to produce and interrupt an arc in free air such as horn gap switches or expulsion fuses. The standards should reflect a preferred BIL series or range for each nominal voltage as illustrated in Table 2. Each BIL, rather than each nominal voltage, should have an associated set of minimum standard clearances such as those shown in Table 3. Thus, the possibility of choice of various insulation levels for a given system voltage should be open to the substation engineer and the codes and standards which gives values in a manner
1252
L , 1 I
Clearances Margin
Vertical Clearances
Firstly, let us draw a clear distinction between the minimum electrical clearances (line to ground - for example, conductor to structure, and line to line) and the minimum safe clearances. By safe clearances we mean those defined (per National Electric Safety Code [lo]) as clearances from live parts to any permanent supporting surface for workers.. The distinction is apparent from Figure 2. While the clearance c represents the electrical clearance designed not to sparkover, the clearance b is the safe clearance designed for people to move safely in the proximity of energized electrodes. Figure 2 is complemented by Table 4a. which shows the appropriate numerical values as specified by the Safety code (10 Table 124-1) for various nominal voltages. (There are two specified minimum ground clearances with the 230 kV since two values of BIL are recognized with that voltage.)
t
Clearances Insulation
t
Highest Overvoltage Voltage
Margin
Margin
Highest Overvoltage
I/
1b)
la)
b bl Where c b d
= 86+c
= 36 + C = Electrical clearance. = Vertical distance from a grounded support surface
base on an insulator. bl = Horizontal distance from a grounded support surface for workers to a live part. Note: Limits of approach distances of less than b or bl may be applied under restrictive conditions.
1253
"Where surge protective devices are applied to protect the live parts, the vertical clearances, Colwnn 2 of Table 124-1 may be reduced, provided the clearance is not less than eight feet and six inches plus the electric clearance between energized parts and ground as limited by the surge protective devices."
Table 4a Vertical Clearances Nat. El. Safety Code ANSI C2 Section 12 Vertical Clearance b ft in Nominal
Table 4b Horizontal Clearances Nat. El. Safety Code ANSI C2 Section 12 Horizon tal
NEMA SG6
ANSIC37 Table 1 Ground Clearance Winimum)
C
Voltaee
a kV
Clearadce
bl ft
4
NEMA SG6
ANSIC37 Table 1 Ground Clearance (Minimum]
C
Nominal
Voltaee
a kV
10 11 12 12 14 14
5
7 2 10 10
10
ft 2 3 4 4
5
6
in 1 6 2 10 11 11
6 6 7 9 9
in 11 1 8 4 4 4
ft 2 3 4 4
5
6
in 1 6 2 10 11 11
8 8 8 7
0 0
11 11
Ground here means a structure supporting an insulator, not the "supporting surface for workers". Thus, if, for example, a 115 kV station was designed with a safety clearance 8-112 feet + electrical minimum clearance and by application of the arrester we keep the maximum voltage stress such, that the same electrical clearance could hold for 230 kV nominal, then we would be left with the original 8-1/2 feet of safety for workers. From the other viewpoint, we can say that addition of 8-1/2 feet to the minimum electrical clearance, as limited by the arrester application, is in fact more conservative than the present practices. This is apparent from Column " d in Table 4a. The safety clearances, taken as the difference between "b" and "c" values, are generally smaller than 8-1/2 feet. It is concluded that with the vertical clearances the voltage uprating can be done safely in compliance with the National Electric Safety Code.
Horizontal Clearances
Where appropriate, the horizontal clearance requirements should also be taken into account. The NESC does specify the horizontal safe clearances but does not indicate how they can be modified if surge protective devices are applied. It is logical to deduce that the same would apply as with vertical clearances, only the distance would be smaller than 8-1/2 feet. It would be given by the reach, rather than the height, of a person. Table 4b shows in Column "b" the horizontal clearances as specified by NESC and the minimum ground clearances in Column "c" per existing standards. The difference between the two is given in Column "d". From these values and by comparison to Table 4a it appears that the safety clearance based on a reach distance of 31/2 feet would be a reasonable choice. Thus, the following procedure, analogous to the one for vertical clearance from a permanent supporting surface for people, as quoted above is suggested:
This is similar to but not identical with limits of approach which are measured from the person as they move about but which may not involve permanent supporting surfaces. In those cases where the criteria cannot be met, consideration could be given to other precautions (for example, railings) which could be applied to assure safety and prevent accidental contacts by personnel. Even though the above interpretations are straightforward, difficulties have been experienced in the industry. The fact is that the safety rules are meant in the first place to protect personnel who do not have, neither are required to have, a deeper education in electrical engineering. Such personnel have the right to check the safety rules and to understand them. In this context there are some difficulties with the above quoted paragraph on vertical or horizontal clearances. Namely, the end of the paragraph, quote "plus the electric clearance between energized parts and ground as limited by the surge protective devices" defines the safety clearances, somewhat vaguely. The substation maintenance personnel would be much more receptive to a numerically given distance, which can be easily checked by them, than to calculations of clearance from arrester characteristics. Within this context the Table 3 can provide help. The minimum electrical clearance is here given by the BIL, not by the nominal voltage. Thus, for example, if the application of surge arresters enables the use of insulators rated at 350 kV BIL in a 115 kV substation instead of the usual 550 kV BIL, then the minimum electrical clearance is 25 inches instead of the 42 inches. The necessary clearance for personal safety is 8 feet 6 inches + 25 inches = 10 feet 7 inches instead of 11 feet 7 inches shown by NESC for 115 kV, see Table 4a. This example covers a practical case for a 69 kV substation uprated to 115 kV.
"Where surge protective devices are applied to protect the live parts, the HORIZONTAL clearances, Column 3 of Table 124-1 may be reduced, provided the clearance is not less than three feet and six inches plus the electric clearance between energized parts and ground as limited by the surge protective devices."
"The length of break of outdoor break switches, when in full open position, shall be at least 10% in excess of the dry arcing distance over the insulators and shall be such that the open gap(s) will withstand a test voltage which is 10% in excess of the low frequency dry and impulse withstand test voltage given in Table I ."
1254
This sentence is followed by a reference to Table 3 in [4]. This reference is superfluous and could be disregarded. The important fact from the safety viewpoint is that the eventual sparkover is by the design of the switch forced to occur across the insulator rather than across the open break distance. Thus, with the disconnected part grounded, the maintenance personnel are safe independently of the BIL or the rated voltage of the switch. This would become apparent from the applicable standard, if the length of break would be associated only with the BIL and not the rated maximum voltage. There should be a choice of BILs for various rated voltages, as discussed in previous paragraphs and shown in Table 2. Table 4c
Outdoor Air Switches
Rated
Rated
Length of Break
IEEE/PES 1989 Summer Meeting, Panel Session Paper, Long Beach, CA, July 11,1989. [ 81 J.R. Stewart, Ongoing Work with Substation Voltage Uprating . IEEE/PES Summer Meeting, Panel Session, Long Beach, CA, July 11, 1989. [9] H. Elahi, J. Panek, J.R. Stewart, H.R. Puente, Substation Voltage Uprating: Design and Experience, Paper 496-0, IEEE/PES Summer Meeting, Minneapolis, Minn., July 1519, 1990. [ 101 ANSI C2: National Electrical Safety Code, 1990. [ 1 13 AIEE Committee Report: A Guide for Minimum Electrical Clearances for Standard Basic Insulation Levels, AIEE Transactions, Part I11 (PAS) Vol. 73, June 1954, pp. 636 -
641.
Line X L
6
7 8 9 10 11
Maximum Voltage
(kV rm&
72.5 121 145 69 242 242
Wave
kV Crest
350 550 650 750 900 1050
Disrance (inches)
32 (0.813 m) 50 (1.27 m) 60 (1.52 mj 68 (1.73 m) 84 (2.13 m) 104 (2.64 m)
Distance (inch)
22 (0.559 m) 32 (0.813 m) 38 (0.965 mj 4 4 (1.12 m) 50(1.27 m) 57 (1.45 m)
CONCLUSION
In conclusion, a simple summary can be made: If a substation met the safety criteria before uprating, it would also meet the criteria after uprating. Uprating by application of surge protective devices would reduce the electrical clearances, line to ground and line to line, but will leave the safety clearances unchanged and therefore satisfactory. The above, of course, assumes that the same criteria is used and that it remains valid This wording really does not cover newly constructed substations. However, the newly constructed substations can be designed to smaller BILs and clearances than those presently standardized and still provide personal safety corresponding to present requirements. These requirements, embodied in the National Electrical Safety Code and some other standards, have to be interpreted with care, but also with due flexibility, as discussed in the above paragraphs. These interpretations should be made easier, particularly for personnel without deeper knowledge, by appropriate changes in standards. The most basic change would provide some freedom of choice o f BIL for various rated voltage in the HV range. This is embodied in the insulation coordination standard [l], but not in some other, generally older standards. Updating of such standardsis desirable so that they do not limit the applications of new technologies and corresponding economies. Changes in standards generally require considerable length of time. Until it is accomplished, this paper can provide support for the engineer striving for economy with safe design in mind.
REFERENCES
NEMA SG6-1977: Power Switching Equipment. Panek, J, Elahi, H.: Substation Voltage Upgrading, IEEE Transactions on Power Delivery, July 1989, Vol. 4,No. 3, pp. 1715 - 1724. ANSI C 92.1: Insulation Coordination. ANSI C 37.32-1972: Schedules for Preferred Ratings, and Application Guide for High-Voltage Air Switches, Bus Supports and Switch Accessories. IEC Publication 7 1: Insulation Coordination. W.J. Lannes, K.W. Priest, et al., 230 kV Operation of a Substation Designed for 115 kV by Controlling Voltage Transients, IEEE Transactions, Vol. PAS-90, pp. 16981718.
wish to congratulate Dr. Panek and his coauthors for an interesting paper. and have A study of the safety aspects Of uprating is preceded the building of uprated substations. The authors response to the following comments and questions would be appreciated: The last sentence of the first paragraph of the CONCLUSION of the paper needs to be emphasized: safety requirements are not automatically met in the uprated substation unless the voltage withstand criteria have not been changed. In this connection, the following is noted: (a) Same compact designs involve a reduction in voltage withstand criteria. For example, the 500 kV compact line designs reported in Ref. 12 are based on allowing one flashover per 10 switching surge (SS) operations while the typical industry standard is one flashover/lOO SS operations. (b) Some of the uprated substations have experienced flashovers which remain unexplained [ 131. In view of the above, there may be locations in some uprated substations where safety clearance requirements are not met and access needs to be restricted by use of barriers. As pointed out in the first sentence of the second paragraph of the CONCLUSION, the safety criterion given in the paper does not cover newly constructed substations. To cover this latter important case, there is a need for providing two things: (a) An equation for determining the limit of approach as a function of the overvoltage limits guaranteed by the applied arresters. Presumably those overvoltage limits can be correlated to the ratings of the arresters. The methods for calculating safety clearances reported in [14] and [I51 may be useful in that respect. (b) A guide regarding the amount of room which should be provided for the mans body in the different situations. A basis for such a guide is already available in the material given in [16]. As an example, [16] specifies the vertical reach of the worker as 2.25m as compared to the 8.5 ft (2.59m) given in the paper. In a traditional substation design, three different insulation levels often exist: A BIL for the transformer, a somewhat higher BIL for the post insulators, disconnect switches and circuit breakers, and a third voltage withstand level higher than the above two applying to the air gaps to grounded structures. The above practice (reflected in Fig. Ib) may be justified by the fact that the overvoltage level appearing at a bus point increases with increase in separation between that point and the nearest arrester. In the traditional layout, surge arresters were usually used only at the terminals of the transformers. Thus the existence of 3 insulation levels may have been justified. The new compact designs involve use of more sets of arresters. Nevertheless, some of the points of the strain bus may be far enough from the arresters to justify having two insulation levels. In fairness to the pioneers of traditional substation designs, it should be mentioned that they used higher insulation levels not because of limitations of their knowledge about insulation coordination but rather because of the limitations of the equipment available them. Insulation levels have dropped in 3 distinct steps when better equipment became available: (a) The traditional gapped arresters of today represent a huge improvement over the arresters used 40 years ago. Insulation coordination levels in ANSI Standards have undergone at least one major overhaul to reflect that fact. (b) With introduction of EHV systems, switching surges became the governing parameter. Use of breaker pre-insertion resistors controlled the overvoltages at their generation point thus keeping insulation levels from rising in proportion 60 Hz levels. (c) The introduction metal oxide arresters opened a new era by controlling both lighting and switching surge levels. Other exciting developments are destined to happen in the future and metal oxide arresters will probably become obsolete. Hopefully, the engineers of the future will look back with kindness at the less knowledgable engineers who pioneered metal oxide arresters!
[I21 E. J . Yasuda and F. B. Dewey, BpAs New Generation of500 kv Lines, IEEE Trans., Vol. PAS-99, No. 2, pp. 616-624, 1980. [13] H. Elahi et al., Closure of Substation Voltage Uprating: Design and Experience, IEEE Trans. on Power Delivery, Vol. 6, No. 3, pp. 1049-1057, July 1991. [ 141 IEEE Working Group; Live-Line Maintenance Methods, IEEE Trans., Vol. PAS-92, pp. 1642-1648, 1973. [I51 P.C.V. Esmeraldo et al., Calculation of Minimum Safety Distances for Live-Line Maintenance-A Statistical Method Applied to 765 kV AC Itaipu Lines, IEEE Trans., Vol. PWRD-1, No. 2, pp. 264-271, 1986. [I61 CIGRE Working Group 06 of Substations Committee, The Effect of Safety Regulations on the Design of Substations, Electra, No. 19, pp. 79-102, November 1971. Manuscript received August 22, 1991.
Working Group El: RECOMMENDED MINIMUM CLEARANCES IN SUBSTATIONS: The authors wish to thank Mr. Mousa for h i s interest in the paper and for his comments. Below is our reply: 1.a: Flashover criteria for station and line insulation ought to be different. T h i i is because the line insulation, having a self restoring insulation can accept a higher number of flashover occurrence than the insulation of station equipment. Also, for substation faults, no high speed reclosures are allowed; while for lines they are often implemented. Also reliability requirements influence the criteria, loss of a line and loss of the whole substation have different impact on the system. The voltage withstand criteria for the insulation of the substation equipment have not been changed. 1.b The failure of 2 cap and pin insulators remains unexplained. It is not clear if the failure had an electrical or mechanical cause. As described in [13), both insulators were 20 to 30 years old, i.e, they were exposed to transportation and handling during the new installation as w e l l as to temperature variations during the years of service. A number of failures of the same insulators has also occured in other,- not uprated- substations. The utilities concerned did not change their safety criteria, they replaced the insulators with other units, often substation post. We agree that substation uprating necessitates careful examination of safety clearancerequirementsat all locations. 2. The principles of voltage upgradings are the same as for new substations except for the latitude of providing additional margin in new construction. Reduction of overvoltage stresses can always be accomplished by additional arresters as in the case of uprating. Thus the paper is fully applicable also to new, compacted, substations. Different levels of insulation within substation is an approach adopted 3. by many in the past. With the present surge arresters, protective levels have been significantly reduced. The range of protective levels allows one to choose the most economical solution. - Prescribing the number of insulation levels within the substation is not necessary. This paper recognizes the technological changes permitting upratiag. 4. These changes include protective equipment, switching surge and temporary overvoltage suppressing schemes (i.e.,closing resistors/reactors) and analyticaltools (i.e.,EMTP).
In conclusion the safety aspects raised in this paper show the need of updating and coordination between the provisions of some standards, related to the choice of the BIL for a given rated voltage. There is a need of providing a guide for the designers regarding the minimum safety clearances i n relationship with the BIL and the available protective level, rather than only to the rated voltage, as it is now. Manuscript received February 6 , 1992.