Cablejoint and termination failures are always a controversial issue. The cable jointing kit manufacturer is generally the first to be called to account. In most cases the issue ends up in accusations and counter accusations involving the cable jointing kit manufacturer, the cable manufacturer, and the installation contractor. All three parties end up pointing the finger at others. The end user who has many different network items to look after is often unable to devote sufficient time and attention to go into the matter in sufficient depth, and sometimes may not have the in-house knowledge to correctly do a root-cause analysis to determine the preventive steps. As hok Saigal
-
Fit1J Failures of
MV Cable Terminations: A Study
j
52 11 September 2012 11 Electrical India
M
ost Cable manufacturers today do not have in-house construction teams that have accumulated knowledge on the practical usage of their cables and can analyze the field conditions that could have led to the failures. The larger Contractors mostly sub-contract the cable laying and jointing, and the smaller subcontractorsthey employ neither have neitherthe skill, nor organisational strength, nor the inclination to go into technical issues. They are more like labour providers for technical installations. Given the generally acknowledged constraints of appropriately skilled workers in India, they make-do with what is available. Frontec has supplied many cable jointing kits which are in use in different types of projects throughout the country. In some cases the performance has been faultless. In some cases the failures have been very high and heart- wrenching. In all reported cases of failures, Frontec, as a responsible manufacturer, attempts to analyse, learn, and improve frsmthese experiences.The action taken is not only limited to -- - product improvement, but is targeted at enhancing the overall customer satisfaction from a reliable electrical network, encompassing the areas of cables, their installation and selection and installation of accessories. This study aims to share some of the findings of cable termination failures.
c MV Cables
I
Failures due to poor materials and design
I
I
r:
I
I
I
L
t
4 I
I
~ ~ ~ m d u c -ah
d designs are few,Mlrdfmd~ 'con(bmlto low !:prkedand gtW h r k e t ktts.
-
*-b these are a h ~onwtbne5
, "
COUt8tddt
Failures due to poor materials and designs are few, and mostly confined to low priced and grey market kits. Unfortunately these are-also sometimes counterfeit and carry the names of better quality brands. Some defects are easily apparent lon visual examination before installation, or after installation and before commissioning. For example many components are without adhesive coating at crucial parts. In some cases the adhesive layer is too thin. In others it is a hard and inflexible coating that cracks on the slightest flexing. In such instances the environmental seal against moisture and water ingress is non-existent. Entry of water into the higher electrical stress tareas o can result in.trackina over the cable insulation or the stress control tubing. In case of outdoor terminations the skirts do not adhere to the tracking resistant covering.They may slide out of position.~his reduces the arcing diitance. In addition, the creepage and arcing paths are no longer over the surface of the skirt, but the currents find a path between the neck of the skirt and the track resistant tubing. This considerably shortens the creepage 81flashover distances. In case of rain these can particularly quick11 lead to termination failure. The lack of adhesive seal at all external
have a common colour and texture. However after a few years certain components, specially breakouts and skirts, start showing a lighter colour, a rougher surface and a drastic reduction in mechanical strength. In some cases it has to been found that a skirt could be torn apart with bare hands as easily as tearing a piece of chapati!
Failures due to poor cable preparation Failures caused by poor cable preparation form the largest percentage of failures observed by us. Fn small concentrated installations like power stations the concentrated location allows closl
d
interfaces can be visually checked before installation. The quality of the adhesive bonding can be equally easily detected by trying to rotate an installed skirt. A good adhesive should not allow the skirt to rotate, while a poorly bonded one will turn with little reistance. In poorly designed low priced kits, thl length of track resistant tubing is often shortened to enable lower prices. As a result, the length of tubing is not sufficient to cover the individual cores from the breakout to the lug. A length of bare XLPE insulation is left exposed. This will crack under UV and allow entry of water into the cable and termination. Another cause of-failure observed in some competitive products i s the poor material quality of some of the heat shrink components. Initially all components appear to be made from the same material, and
I
Fig. 2: Insufficient length of track resistant sleeve. Also note the scrape marks on the sleeve
Electrical India I1 September 2012
k5
MV Cables
&
supervision and check on the quality of jointing and the skill levels of the jointers employed. In disbursed locations like state distribution utilities and rural electrification projects such close monitoring is generally absent. Jointers who lack the training and skill levels are often allowed to do jointing due to the non-availability of properly trained jointers. Removal of semi-con is a highly skilled and often tedious technique. This is specially true when the semi-con is removed by scraping with a strip of glass, as is the most common practice in Distribution and Rural electrification projects and maintenance. The termination failures have been seen to be caused by three common defects in semicon removal. First and most obvious in a visual examination is the failure to smoothen the insulation with Alox abrasive tape after removal of semicon.This leaves air gaps between the stress control tubing and the cable insulation, which in turn
I
r ~ g3: . Rough surface of insulation after scraping semicon with glass
I
-I
Fig. 4: Peaks left while removing semicon. Also note the good adhesive bonding for sealing against water ingress
4 I1 September 2012 11 ~lecticalIndia
causes high levels of partial discharge that over time damage the cable insulation. Secondly the semicon is not removed in a level circle, but jagged peaks and deep valleys are left. The peaks become points for stress concentration, again leading to high partial discharges. a Thirdly, knife cuts and nicks are often made in the cable insulation. These cuts result in reduced insulation thickness, and are specially critical and common at the edge of the semicon. This being the highest electrical stress area of the termination, it results in very early failure.
Failures due to wrong installation of jointing kit components Of all causes that lead to pre-mature failure of cable joints the most in-excus~bleis the failure to read and follow instructions provided in each kit. Most installation instructions of reputed manufacturers depict the procedure in easy to understand pictures along with the description. So there is no reason that even a not well educated person should not refer to them. And it is not understandablewhy Supervisors should not read them to learn the procedure, and to check the workmanship of the jointers. Yet in many instances of failure it has been easy to establish that components of the kit were installed incorrectly. One of the common and fatal errors is to position the stress control tubing incorrectly. In many cases it has been installed over the metal tape screen, down towards the cable jacket instead of over the cable insulationtowards the lug. Naturallythe stress control function is completely lost. Furthermore, in case of single core cables, the red tubing is then installed with the stress control tubing4rotruding out below the end which has the sealant adhesive. This leaves a clear entry path for water between the uncoated stress control sleeve and the cable screen.
Another common mistake, as discussed earlier, is to leave the cable core longerthan the length of red outer sleeve provided. The lug seal is then installed overlapping the lug barrel, and a length of XLPE cable insulation is left exposed to the environment.
1
Fig. 5: Stress control sleeve positioned wronalv towards cable iacket
Failures due to improper earthing of cable screens (at screen, at pole) MV XLPE Cables are screened for many reasons. A major reason is to confine the electrical field to within the insulationand prevent voltages outside of the screen. This is achieved by earthing the metallic part of the insulation screen. When the screen is not earthed the electric field is not confined t o within the screen. Capacitive coupling occurs between the cores and voltages are induced on each and every insulating surface between them and near them. These would be on the surface of the cores and the inner and outer sheaths. In the case of 3 core cables there would be potential differences between the cores that would lead to discharges between them. In simple terms, these discharges are electrical sparks. In practice the magnitude of these sparks is considerable. Chattering is often heard. In some instances, particularly during humid and wet weather visible sparking has been observed at night on MVaerial bunched cables.This sparking has been observed to lead to damage to the core covering, progress to the metallic screen, and eventually to damage the insulation itself. In early cases of detection of poor
.
5
earthing through the sound of chattering it has been found that proper earthing has eliminated the problem. Poor earthing of the screen arises at two locations. First is the connection between the metallic part of the cable screen and the screen earth lead. Often an earth braid is used. The convention in India is to connect these braids to the metal tape screen using binding wire & recommend soldering. Unfortunately soldering is most often not done by the jointer. Further in many cases the binding wire connection is also left loose. A good quality electrical connection is made more difficult when the cables are provided with aluminium screens and not copper ones. In all cases there are two consequences. One is that the screen can become free floating electrically and develop the induced voltages described above. The second is that the loose connection leads to localised sparking that damages the insulation below the connection and leads to a puncture directly below the connection between the earth braid and the metallic screen. The problem is further aggravated in case of fault conditions where high levels of current pass through the screen and the.high current combined with the high contact resistance burns a hole through the insulation. Another source of poor earthing of the screen occurs outside of the termination. This is the connection from the earth braid to the external earth. In some instances the earth lead is simply not connected to anything. In many cases it is connected to the commonly used 8 SWG GI earth wire by simply threading the earth wire through the eye of the thimble of the earth lead. This leads to a limited line contact, with high possibility of the connection being intermittent. In other instance an under sized binding wire of 1.5 or 2 sq mm is used to make the connection between the earth lead of 4, 6 or 12 sq mm and the metal earth
cable terminations. The Jointers do not have the requisite skill levels. While installation of modern accessories has been made very much easier compared to the days of hot pouring compounds and paper insulated cables that required testing for moisture absorption, the skill required to prepare cables has not diminished. Surgical precision and neat working methods are still essential parts of a Jointer's attributes. These have to be developed through proper training, and honed through practical experience. While every erection contractor or utility engineer cannot be expected to have the resources to carry out such training, they should have the resources to check and evaluate the skill level of the Jointer before engaging or allowing him to do an installation. If required they can take help of the Jointing kit supplier to carry out such evaluation and prequalification. Implicit in this is the requirement for the Supervisory staff (which includes not just the Field Supervisors, but also more senior Project Engineers) to be themselves conversant with the requirements of ensuring good quality cable joints. Too often the lack of knowledge of the subject by the Supervising Engineers allows the work to be passed at sub-standard levels. Therefore the training of Engineers, in a different manner and with different content and style, is equally essential. Investment in training and qualification of jointers is a good value
Fig. 6: Loose connection of earth braid tn to met; ' -
'ig. 7: Good connection of earth braid to metal tape screen
Fig. 8: GI earth wire loosely looped thmmnnh l m n n nf m = d h hr=ir(
0
strip. This often fuses out under high currents, again leaving the earth floating.
Role of training, evaluation and Supervision Irr many of the failures studied by Frontec, it has been found that the human factor has often been neglected during the installation of
6 11 September 201 2 II Electrical India
.
Pi.
.;
. 1
,,
.
,*\ ,
.
I
.T .. . .. .. ., .
I..-
.
I .-
Fig. 9: On site field training of Jointers
-
. ~. . .-
. i
:.
.,
.
..
,
,
..
, .-y
:
-.
. - - . ~ - . .- ---~-
~-
The very best
proposition, and i s strongly recommended as the easiest way to avoid failures, and the most cost-effective way of doing so.
...Snap-on mesh cable tray ever developed
Conclusion
by OBO engineers. Lot of practical advantages make the GR-MagicB more than just fast:
When a Joint or termination fails in the field, there is no one willing to take ownership of i t s mortal remains. The cablecompanies by and large do not consider the issue at all involves them. The Electrical Consultant limits his responsibility to a good set of specifications, and steers clear of taking any stand on the merits of the particular brand or manufacturer selected by his client on the basis of a technical cut-off in line with the specification, and the lowest bid criteria.TheContractorfeelsthat his responsibility ends with the standard 18/24 month guarantee period. In casea fewjointsorterminationsdofail during the guarantee period, he will first blame the kit supplier, then the cable quality, and if these excuses are not accepted, finds it simpler to replace the failed product.The end user who has the most to gain as he has to live with his network for the next 30 to 40 years continues to give prime importance to the up-front cost of the kits, and feels comfortable that he has passed on the responsibility to the Contractor or kit supplier through the guarantee clauses. In this scenario it is the more responsible cable jointing kit manufacturers who try to protect their reputation by providing cable jointing services, training in cable jointing, evaluation of Jointer skills, and field services to analyse field failures, and thereby minimise future complaints and customer dissatisfaction. All too often this value addition is not appreciated by the various segments that benefit from it. More importantly the lessons that could be learnt from it are not learnt and their significance under-estimated and the opportunities to eliminate the possible causes of future failure are lost. These subjective criteria need to be given due importance along with the bare price quote in choosing not only Cable Jointing kit suppliers, but any product or service supplier, to partner or associate with for any project.
1 Ashok Saigal is an Electrical Engineer from IIT Kharagpur,
4
Available In Electro Galvanized, Hot Dip Galvanized & Stainless Steel
and an MBA from a leading US Ufli~ersity. He is ths co-founder and Managing Director of Frontier Technologies, a small and specialised manufacturer of heat shrink products and cable jointing k i . In the past he has alsD been associated with L&T Raychem, and Sicame. He has been associated with Cables and accessories for over 30 years. He has played an important role in the introduction of several new concepts in lndii, including the introduction of !$at shrink technology, Low Fire hazard cables, and ma& recently Aerial Bunched cables. He has served on several Sub-committees of the Bureau of Indian Standards related to Cable Jointing kits, and helped many Utilities in developments of their Standards and Practices related to --tab IjBS.
