Taslink or Moonlink?

A proposal for an HVDC link between NZ and Australia hit the news last week. Here are 10 reasons why this is a particularly daft idea.

Ideas matter. One such idea is that interconnections are a good thing. Be they international trade, the internet, municipal water or electricity, or personal relationships, we all seem to benefit from being well connected. However, there are limits to how connected we can be. The limits differ depending on the domain but sooner or later some kind of physical or practical limitation is reached. In the domain of infrastructure such as electricity, that limitation is often a commercial one. In other words, the benefits of further interconnections, at some point, don’t outweigh the cost of creating those connections.

Think of your personal relationships. You can only have so many high quality ones. This is a simple fact due to the limited time we have available. All personal relationships require time to nurture, and inevitably, at some point, you have to make a choice about nurturing one at the cost of another.

Last week a proposal for an HVDC link between NZ and Australia, an interconnection between the two country’s electricity grids, hit the news. Technical details of the proposal can be found at: http://www.taslink.com/, although the website is very light on information.

I must admit that I nearly fell off my chair and spat out my coffee when I saw this come to light, and indeed the rather credulous treatment that it got from the mainstream NZ media has now prompted me to write this post.

Could it be done? For sure, I’ve no doubt, that given enough money and time, something like this could be built. It’s not out of the bounds of technical possibility. However, just like hydrogen transportation and fusion reactors, the commercial reality is pure fantasy. Should it be built? No, for reasons I’ll articulate below.

Reason #1

It would become the longest HVDC submarine cable link in the world by more than 1800 km. The world’s current record holder is the 765 km Viking Link between Denmark and the UK. Viking Link was constructed for £1.7bn. True, someone has to push the limit and stretch the boundary of what is possible, but a link almost four times the length of the current record holder between two sparsely populated countries is maybe pushing things a bit far.

Reason #2

It would be the world’s deepest submarine HVDC cable. The Tasman sea is deep, over 5000 m at it’s deepest point, with large stretches deeper than 4000 m. The current world HVDC cable record holder for depth is in the Mediterranean Sea at a depth of just over 2000 m. As for reason number one, it’s good to push boundaries, but this is likely to add significant cost to the project.

Reason #3

The NZ and Australian grids are small (from an energy perspective). HVDC links only work when you have a large amount of potential energy transfer over long distances, generally to connect a large generator (or generators) to a large load. For example, the inter-island link in NZ connects NZ’s abundant hydro generation in the South Island to the larger population centre of the North Island. The Basslink between Tasmania and mainland Australia connects Tasmania’s abundant hydro and wind resources to the mainland. Neither NZ or Australia currently have a surplus of energy. In fact, neither country can build generation fast enough to meet growing electrification demand and retirement of existing fossil fuelled generation. Taslink would be built on the premise that one of countries would have a surplus to trade whilst the other has a deficit.

Reason #4

The NZ interconnection point does not have the required grid stability. If one assumes that the link will use conventional thyristor based converter stations, which are also the lowest cost for a link of this size, then grid stability becomes a key consideration. One way to talk about this is what is known as short circuit ratio (SCR). This is the ratio of the fault level (a proxy for grid strength), to the size of the development or link. If we are overly generous to the development and assume that it is connected to Otahuhu substation, one of the strongest buses in the region, then the SCR would be around two, which is right on the minimum level of feasibility for a thyristor based link. At a minimum, grid stabilisation plant such as synchronous condensors would be necessary, which would increase the capital cost.

Reason #5

The investment money would be better spent on local generation and transmission projects within both countries. In today’s money, $12bn would buy around 7000 MW of solar generation, which could supply 20% of NZ’s electricity needs.

Reason #6

It’s not the best interconnection option between the two countries. A link between the South Island and Tasmania would be about 20% shorter and would connect into areas of both grids that are near to strong sources of hydro generation, providing grid strength.

Reason #7

The NZ battery project looked at this as an alternative “dry year” solution for NZ. It was thrown out at the pre-feasibility stage for being too expensive, too risky and not delivering on the right outcomes.

Reason #8

The costs are not credible. Perhaps the closest comparable recently constructed link is Viking Link which cost £1.7bn (NZD 3.84bn). Viking Link was constructed in Europe, the “backyard” of many key component suppliers like Siemens, and with extensive offshore cable laying capability. Based on other projects, we could expect a “NZ/Aussie” premium of at least 20%. With this project being 3.4 times longer and much deeper, plus higher capacity, I would guess the actual cost is more likely to be around 6x Viking Link, or $23bn.

Reason #9

Transpower’s existing asset base is ~$6.5bn. This link on it’s own would nearly quadruple that. Yes, the development company is looking for private investors, although they haven’t ruled out government investment. Presumably it would make sense for Transpower, given their expertise operating the inter-island HVDC link, to have some ownership stake in this proposed project. However, even a modest percentage stake by Transpower would significantly increase their regulated asset base, which would in turn flow on to higher transmission prices.

Reason #10

It will require higher electricity prices in both countries to pay for it, which would need to be astronomical to justify the investment. Some back of the envelope math assuming a 2 GW link that is 60% utilised, results in energy flows of 10.5 TWh pa. Assuming a 10% return on capital invested and a $23bn investment, the link would need to generate free cash of $2.3bn pa. Working backwards to revenue, assuming the project is debt financed with 20% equity on a 6.5% loan, debt servicing costs would be around $1.2bn pa. I don’t have a feel for what OPEX would be, but possibly in the tens of millions annually. Even ignoring that, you are still looking at revenue of $3.5bn just to provide a basic return to investors and service the debt. This works out at an outrageous $333 / MWh. Even if we are generous to the project and assume it can be delivered for $12bn, is 100% equity financed and investors only want a return of 7%, then the cost is still $80 / MWh. Remember the link is only shuttling energy around, so this is the premium that has to go on top of the existing energy price, or put another way, is the arbitrage needed between NZ and Australian energy prices.

As much as the engineering nerd in me would love to see such a link constructed, no sane investor is going to put money into this project. It belongs in the dumpster of bad ideas, hopefully never to be spoken of again. We should recognise it for what it actually is, a publicity stunt. Let’s not pander to the proponents by giving it any more air time than it deserves.

*This author has been known to make mistakes. If you think anything I’ve written above is incorrect, get in touch, and I’ll gladly correct any errors.