Wikipedia article on estimating costs of transmission upgrade to deal with renewables

Below is an extract from the excellent Wikipedia’s article on National Grid which references Bernard Quigg’s paper at the last Claverton Conference.

Can anyone help on this – it seems to me that the Costs of Transmission derived from Bernard’s’ paper are too high when compared to the method derived from Triad charges,  Is this because Bernard is including generator connection charges which it could be argued are the costs of connecting the generator, not transmission.  Any thoughts?

Best to comment and then edit the wikipedia article directly (anyone can do this of course)

National Grid (UK)

From Wikipedia, the free encyclopedia



 Triad demand

Triad demand is measured as the average demand on the system over three half hours between November and February (inclusive) in a financial year. These three half hours comprise the half hour of system demand peak and the two other half hours of highest system demand which are separated from system demand peak and each other by at least ten days.

These half hours of peak demand are usually referred to as Triads.

In April of each year, each licensed electricity supplier (such as Centrica, BGB etc.) is charged a fee for the peak load it imposed on the grid during those three half hours of the previous winter. Exact charges vary depending on the distance from the centre of the network, but in the South West it is £21,000/MW for one year, or £7,000/MW for each of the three half hours, for convenience assuming they were identical, (which is unlikely however they will be close). The average for the whole country is about £15,000/MW per year. This is a means for National Grid to recover it charges, and to impose an incentive on users to minimise consumption at peak, thereby easing the need for investment in the system. It is estimated that these charges reduced peak load by about 1 GW out of say 57 GW.

This is the main source of income which National Grid uses to cover its costs and these charges are commonly also known as TNUoS – Transmission Network Use of System charges. (Note this is for high voltage long distance transmission and the lower voltage distribution is charged separately). The grid also charges a fee to generators to connect.

[edit] Estimating Costs per kWh of Transmission

If the total number of units delivered by the UK generating system in a year, are divided into the total TNUoS or Triad receipts, then a crude estimate can be made of transmission costs, and one gets the figure of around 0.2p/kWh. This is calculated by taking the total annual Triad charges, which are say £15,000/MW/year x 50,000 MW = £750 million/year and dividing it by the total number of units sold – say 3.6 trillion kWh.

However, according to [11] which is probably more accurate…”According to the 06/07 annual accounts for NGC UK transmission, NGC carried 350TWh for an income of £2012m in 2007 i.e. NGC receives 0.66p per kW hour. With two years inflation to 2008/9 say 0.71p per kWh”. However this figure will include connections fees paid by generators, so is probably on the high side.

[edit] Upgrading National Grid to deal with Renewables

Questions have been raised about supposedly enormous costs to upgrade the UK or other National Grids to deal with more renewables. However on hte basis of the above figures, this is unlikely to be true. For example if the UK Grid were to be hypothetically doubled, this would only add about 0.71p/kWh which is about 7% of the domestic charge for power in the UK.

[edit] Generation charges

In order to be allowed to supply electricity to the Transmission system, generators must obtain permission to do so from NGET. This permission is supplied in the form of Transmission Entry Capacity (TEC). Generators contribute to the costs of running the System by paying for TEC, at the generation TNUoS tariffs set by NGET. This is charged on a maximum-capacity basis. In other words, a generator with 100 MW of TEC who only generated at a maximum rate of 75 MW during the year would still be charged for the full 100 MW of TEC.

In some cases, there are negative TNUoS tariffs. These generators are paid a sum based on their peak net supply over three proving runs over the course of the year. This represents the reduction in costs caused by having a generator so close to the centre of demand of the country.


  1. Reply

    It seems to me that neither revenue stream (£2 or £7/MWh as per the wiki) will properly reflect the costs of the required physical capital investment to extend the grid.

    Correct me if I am wrong but I would have thought that the numbers given in the wiki would largely reflect revenue and capital maintenance costs with system replacement being only a relatively small proportion of the total income.

    Therefore the capital cost to extend the grid itself would be in addition to the running cost and could be quite a large number.

    Regards, Bob.

  2. Reply

    Dear Bob, my understanding from Bernard Quigg, is that the charges are meant to reflect the replacement capital costs AND the operational costs.

    Hence they should reflect the costs of instantly replacing and operating the grid as it is now.

    Can anyone comment definitively?

    Dave A

  3. Reply

    The comments above are a bit off the mark.

    The point is that National Grid is a regulated business which is allowed by the regulator Ofgem)to cover its costs and make a certain rate of return on its assets. The amount of capital expenditure that National Grid can make also has to be approved by the regulator. The upshot is that the total of National Grid’s income from transmission charges each year is more or less fixed/guaranteed in advance. National Grid can propose and consult on the way in which the charges are imposed on users (and this has to be approved by the regulator), and is expected to do this in a way that is reasonably cost reflective. There is no one right way to do this and the charging principles have changed several times since the industry was privatised, with lots of arguments about the way it should be done. Under the present system the charges are partly related to long run marginal costs, but you can’t really work back from current charges to work out what you think the marginal costs are.

    As a rough rule of thumb, 400kV overhead lines are about £1milion per mile to build, underground cables are around 10-20 times that, 400kV substations cost £20million upwards etc.

    Last time I asked, the total Gross Asset Value (on paper) of the National Grid system was about £11billion. I have no idea what the notional replacement cost would be



  4. Reply

    Dear David,

    this is great stuff – could you give us a one liner on your background, so we can cite it authoritatively in wikipedia?

    Taking your figures, if we assume the national grid is roughly 1000 miles long, and 500 miles wide, as a box, and we duplicated it – then that is say 6000 miles of line. Round that up to say 10,000 miles for spurs to Dinorwic / Cornwall etc.

    Then that gives from your figures say ?10 billion.

    Assume 200 miles of hvdc – say from your figures another ?4 billion.

    Assume say 70 GSP substations cost say ?1.4billion.

    This gives a total Capex of ?15.4 billion.

    A rule of thumb is that financing costs for this would be 10% per annum say ?1.54 billion.

    Opex and maintenance for this kind of plant are reckoned to be between 2 and 4 % per annum – lets assume 4% of capex ie ?0.6 billion.

    Total annual cost therefore of replacing the grid is around ?2 billion.

    Total Annual power sales are say Annual power used in the UK is around 3.6 ×10exp11 kWh.

    This gives say an annual kWh charge if we had to replace the entire national grid of say:

    ?2 billion/3.6 ×10exp11 = 0.55 p/kWh or ?5.5 / MWh which is remarkable close to the number they first thought of, unless there is a mistake in my maths.

  5. Reply

    My original comments were intended to outline the methods used to develop a cost reflective structure and I avoided putting numbers on the items.
    I think it is very easy to underestimate the costs which many of us would regard as optional. Examples which I think we might find hard to price are the damage to the ecconomy caused by transmission systems. Wayleave charges are probably far too low when one considers the damage to tourism,property etc. Business rates based on the value of assets are identifiable in NGC accounts.
    The entry and exit charges are not small items. Grid transformers and switchgear are very expensive and include the lengths of transmission line regarded as service lines to individual ‘customers’ such as spurs to power stations. I did some rough figures a few years ago and ended up with a number not far off a penny per kWh for new assets served by high load factor generation. The ‘nasty’ in the mixture for intermittant renewables is that when the total grid costs are averaged over the units the transmission cost can be double or more that of any form of generation with a high load factor. Most power stations have to make a lot of money in the first few years to justify their existance on later low load factors. Don’t ignore intake and exit charges. Small scale generation in distribution doesn’t need them!

    Bernard Quigg

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these <abbr title="HyperText Markup Language">HTML</abbr> tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>