Experience Dimensioning Outdoor HVDC Cable Terminations

Cables & Accessories

TenneT, the TSO responsible for the transmission system in the Netherlands and parts of Germany, has been building and operating HVDC offshore grid connection systems with rated voltage up to 320 kV. Now, driven by the European energy transition, 525 kV HVDC onshore and offshore cable connections are projected. Unlike the case for AC cable terminations, the height of outdoor DC termination insulators depends directly on pollution severity of the site (SPS) where the termination is located. Historically, the SPS at TenneT’s HVDC stations was never fully clarified and dimensioning and placement of cable terminations were the responsibility of the supplier. Often, this resulted in an outdoor design but placed indoors.

From a technical point of view, this strategy has proven sufficient since no pollution related terminations failures have so far been encountered. However, in the case of a 525 kV HVDC station, such a policy could have significant impact on costs related to visual impact and land requirements arising from the height of these terminations. In order to limit such impact, TenneT cooperated with the Independent Insulation Group to re-assess the requirements for dimensioning creepage distance of HVDC cable terminations.

As the basis for such an evaluation, an area was chosen in northern Germany close to the North Sea. This area has high practical significance since it will be the location for the northern SuedLink 525 kV HVDC stations and also for a future 525 kV HVDC multiterminal hub. Work to determine the minimum insulator creepage distance required for DC cable terminations was divided into two parts, the main part consisting of pollution measurements at sites of interest near a coastal area of northern Germany. This was followed by utilizing these results for dimensioning. For the purpose of verification, dimensioning was based on already available information from existing overhead AC lines.

Measurement-based dimensioning included the following:

• Analysis of a 12-month pollution measurement program from the end of March 2019 to the beginning of March 2020 using special dust collectors, i.e. Directional Dust Deposit Gauges (DDDG) recommended by IEC 60815-1 as well as specially prepared insulator samples

• Direct measurement of pollution on HVDC and HVAC hollow insulators in service

• Evaluation of dimensioning and service experience of existing in service HVDC cable terminations

• Final dimensioning based on the measurements using recommendations of IEC TS 60815-4.

Three sites in the area of future SuedLink DC terminations and multiterminal hub DC terminations were selected for the 12-month DDDG measurement program. The in-service hollow insulators selected for direct pollution measurement were in the same area, thus adding one more location. Fig. 1 shows these investigation sites.

Fig. 1: Sites selected for pollution measurements.

Two of the DDDGs were specially equipped with composite insulator samples (i.e. insulator sheds) allowing direct measurement of equivalent salt deposit density (ESDD) and non-soluble deposit density (NSDD). One of the samples was fully exposed to precipitation while the other (denoted as the roofed sample) was protected by a circular aluminium plate to limit the effect of natural cleaning. The DDDGs and composite insulator samples were used to determine AC pollution level using the relevant standard, IEC 60815-1. Fig. 2 shows these measuring stand installations.

Fig. 2: Measuring stands at the locations 1,2 and 3.

A standard DDDG device is a special measuring tool to determine SPS and consists of four collection tubes with a jar mounted at the bottom and placed on a support column at some 3m height. Each of the tubes has an opening where the pollution can enter and fall into the collecting jar. The four tubes are placed in the direction of compass points. Each month the pollution deposited in the jars is collected and measurements of the conductivity performed according to IEC 60815-1 by filling with de-ionized water up to maximum 500 ml. Conductivity obtained was normalized to a period of 30 days and for the 500 ml of de-ionized water used. Such normalization was necessary since the measurement period was not always exactly 30 days and because at times different amounts of natural water had already collected in the jars.

The normalized value of conductivity measured each month was denoted as the pollution index (PI). SPS class was determined based on a 12-month measurement cycle of the PI, as per IEC 60815-1. When converting PI measurements into SPS class, the highest SPS class came from either the yearly average or the highest monthly value, as in the standard. IEC 60815-1 was used to re-calculate SPS class into design ESDD level.

As with the DDDG receptacles, insulator sheds were also collected monthly and accumulated ESDD and NSDD determined directly for each shed. Sample insulators were then replaced with clean new samples during the measurement period. This way, every sample held only the amount of pollution accumulated over a single month.

Typically, the problem with most ESDD and NSDD measurements is that it is difficult to obtain reliable results (i.e. maximum level of pollution) over only a short measurement time, especially in areas with frequent precipitation as typical for central and north Europe. Often, insulators are cleaned by precipitation before measurements are taken. This, in fact, was the main driver for performing DDDG measurements in such environments to limit impact of frequent precipitation. A DDDG device, however, does not directly provide results in terms of ESDD and NSDD. As such, DDDG measurements and direct ESDD/NSDD measurements must complement each other so as to obtain the best possible estimation of SPS.

In addition to ESDD/NSDD measurements on the insulator samples mounted on the DDDG devices, pollution sampling was also carried out on hollow insulators installed at the SylWin 1 and NordLink substations. In the case of the SylWin 1 system, sampling was done on the DC cable terminations of the two poles (both made with silicone rubber housings) as well as on the negative pole’s earthing switch in the DC yard, which has a porcelain insulator. In the case of NordLink, two AC cable terminations with silicone insulators were chosen. All insulators sampled were installed outdoors.

Attend the 2022 INMR WORLD CONGRESS to hear a lecture by experts involved in this project. The full research methodology will be explained and findings summarized in terms of dimensioning required of HVDC cable terminations for this service environment.