The case for converting gas power stations to run on oil

The UK energy crisis (Part 1)

The subtleties

About half of the UK's electricity generation comes from gas turbines

All the UK gas power plants are of the "combined cycle" variety, in which there is a gas turbine generating most of the power, and then the hot exhaust is used to boil water for a steam turbine. This is a clever trick to make them more efficient, but only the fact that they use gas turbines is relevant to this post. The UK power grid requires about 30GW of power on average, of which about 15GW (on average) comes from these power plants.

"On average" is doing a lot of work here because the system relies on the mix of highly variable renewables, with gas power to make up the gap. This is a sensible system because gas plants are cheap to build but expensive to operate so it doesn't cost that much to build a lot of them even if you expect them to sit idle a lot of the time. In fact the fleet of gas power stations has almost enough capacity to run the entire electrical grid on its own, 28GW, but hardly ever is more than 20GW used. Look at this not very clear graph to see for yourself:

Having all this extra capacity will turn out to be good for my plan.

Gas turbines can happily run on either liquid (oil) or gaseous (natural gas) fuel

The "gas" in "gas turbine" refers to the fact that it’s hot gas that turns the turbine, not that it has to run on natural gas. It just so happens that in power stations the hot gas is usually generated by burning natural gas, which is slightly confusing.

Now I think about it it's stupid that the specific type of gas that comes out of the ground and is good for burning is simply referred to as "gas", or even "natural gas" for that matter. There are many different gases and loads of them are natural. Imagine if petrol was called "liquid" or "natural liquid". Quite silly.

But anyway gas turbines can famously run on Just About Anything, including hydrogen, unrefined crude oil, and even pulverised coal. Turbine engines are less pernickety about the fuel they burn than piston engines because a piston engine has to burn the fuel very quickly at a specific part of the engine cycle, and not before or after. In a turbine the fuel just burns continuously in the combustion chamber so anything that can be pumped in and sustain a flame will work fine.

In reality burning fuel that produces a lot of soot is bad for the turbine and may eventually lead to it working not fine, but a thing they can definitely run on without problems is kerosene (aka jet fuel aka diesel). We know this because that's what planes use. In fact there are far more gas turbines on aircraft burning kerosene than there are burning gas in power stations, probably.

I furthermore claim that it's "not that hard" to convert a turbine that is set up to run on gas to one that can run on oil, I am basing this on the following:

GE, which makes a lot of the turbines in UK power stations, has marketing material about just how Fuel Flexible their turbines are. Including saying they can cope with worse fuels than kerosene (such as unrefined crude oil) for power generation
For a few turbine models they offer a standard upgrade to convert a turbine to "dual fuel" (so it can run on either gas or kerosene), and it only takes 10-20 days. They even claim it can switch fuel during operation. Also for the Alstom GT26 turbine specifically (which is used in a lot of the bigger plants) it can definitely support dual fuel operation
I used to watch a youtube channel by someone who worked in a jet engine repair shop and they would run them on propane for testing even though they usually ran on jet fuel, although now I can't find it

From this I conclude that this is a standard bit of maintenance that happens sometimes and probably takes under a month to complete, and not some wartime "melt down the railings to make spitfires" sort of thing.

Oil is now cheaper than natural gas

This is the wholesale price of gas vs crude oil in the UK. I have also included the retail price (minus tax) of normal diesel as a sort of upper bound since this also includes the cost of trucking it around and selling it to individuals.*

Note that since the war in Ukraine started gas has been above oil for a lot of the time, but not all the time. But remember:

Only about half of the UK gas power stations tend to be in use at any given time, so you could convert half of them and only occasionally end up burning the more expensive fuel
For many of them it will be possible to convert the turbines to "dual fuel" so a single turbine could switch between the fuels while running

So it would be practical to burn whichever was the cheapest fuel at the time. All the area between the graphs is up for grabs as potential savings if you were only buying fuel at the spot price and you could switch between them instantly.

How much money would this save

If we assume the thing I just said above, that power companies only buy fuel at the spot price and instantly switch to whichever is the cheaper fuel, and also that demand is constant (to make things easier), then since the beginning of the year the average cost of generating electricity would be (assuming 50% efficiency):

- For gas only: £168/MWh

- For gas or retail diesel: £142/MWh (15.5% cheaper, overall saving of £260M/month)

- For gas or crude oil: £93/MWh (44.6% cheaper, overall saving of £730M/month)

These are pretty respectable savings, but the really big benefit is that it protects against massive spikes in the gas price. The current strategy is to just pray that the price comes down. It did fairly quickly with the last big spike, and maybe it will again this time (it’s already fallen a bit since I started writing this), but it’s not good for our whole energy system to be beholden to the whims of an extremely volatile market. Oil is much easier to transport and store than gas and so we can expect it to be a stable, if relatively expensive, option to fall back on.

Would this solve the energy crisis

No, because power generation only accounts for about (27%) of gas usage in the UK. The rest is basically used for heating (I'm assuming a lot of "Other industry" and "Other final customers" is also heating)

In general I think people focus too much on clever ways to generate electricity when thinking about how to fix the gas crisis. For example this post does just such a thing. It's an easy trap to fall into because electricity is fun and complex whereas heating is boring and simple. The only way to solve the whole thing quickly is to get a lot of cheap gas from somewhere.

*These are the sources of data I’m using:

Gas, this is the spot price to extract gas from the network. It’s not that clear that this is definitely what this refers to from the website, but I’m pretty sure it is. Worst case it’s a month ahead future or something which will be almost the same as the spot price
Brent Crude, chosen because it’s a light crude oil that you could probably directly burn in a turbine, and it currently comes from Norway which is where the UK gets most of its oil so I think this is close to the actual price a large company would pay
Diesel retail prices, with 20% VAT and fuel duty subtracted. Go ahead, tell me it’s a loss leader, I dare you
USD to GBP exchange rate for the crude oil which is indexed in dollars

Postscript: Possible objections that I haven't already covered

Burning oil is worse for the environment

This is true, but it's not that much worse if you burn something that is closer to diesel than crude oil. The CO2 content of diesel per MWh is only 12% higher than gas. It does also emit more particulates, but there are way more gas turbines way closer to large groups of people at airports, and you don't see people dropping dead left and right.

Power companies are hedged so they won't actually pay the high wholesale price

This is an objection that I expect from the midwits among you, who know enough to know that hedging exists, but not enough to know that it doesn't magically prevent any unfavourable financial situation. It is true that power companies will buy futures (and possibly options) contracts for the gas that they expect to need a few months in advance so that on the day they are not actually paying the spot price for gas. But consider the net effect of having the option to switch to oil on a hedged vs unhedged company:

Suppose the price of gas was around £1/MWh 6 months ago and the price now is £10/MWh, suppose the price of kerosene is currently £2/MWh. Company A decided to play it fast and loose and didn't hedge so now they can only buy gas on the spot market, whereas Company B hedged and bought a future contract 6 months ago to buy gas at £1.10/MWh today. Both companies have agreed with their customers to sell electricity for £2/MWh, and assume the power stations of both companies are somehow 100% efficient.
In the case where they can't switch to burning kerosene:
Company A is forced to buy gas on the spot market for £10/MWh. Profit: -£8/MWh. The company goes bankrupt and the CEO kills himself
Company B takes delivery of its contract to buy gas at £1.10/MWh. Profit: +£0.90/MWh. The company has a subdued Christmas party where they celebrate their prudent planning
In the case where they can switch to burning kerosene:
Company A buys kerosene instead of gas for £2/MWh. Profit: £0/MWh. The CEO is relieved to have got through this crisis and vows to give up his cocaine habit and devote his life to charitable works
Company B sells the future contract that it bought 6 months ago for £10/MWh (making £8.90/MWh profit) and buys kerosene on the spot market instead for £2/MWh. Profit: +£8.90/MWh. The company has a has a wild Christmas party during which the CEO dies of a cocaine overdose
The net gain to both companies from switching to kerosene is the same as the difference in price between gas and kerosene, the hedging just changes the starting point.

It’s also not the case as far as I can tell that power companies use privately stored gas when prices are high, or that there is some kind of below market rate mechanism to get gas from national storage.

Extra infrastructure would be required to supply the liquid fuel to power plants

This is true but I don't think it would turn out to be that much compared to the wholesale cost of the fuel itself, and also when comparing to retail diesel that sort of thing is already priced in. But anyway let's think, what would you need?:

A big tank at each power plant to store about a day's worth of fuel
To bring the fuel in on trucks

Cost of the big tanks:

Here is a website that says you can install a small (2000L) home oil tank for £2800, this is only about the same as the cost of the fuel you could fit inside it. You only pay for the tank once where as you pay for that amount of fuel every day, also massive tanks would be much cheaper than this per unit volume, so this cost would be negligible.

Cost of trucking the fuel in:

A fuel truck carries about 30 tons of fuel, so around £30,000 worth. I would guess the cost of driving the truck from the oil terminal to the power station would be way less then this, maybe a few hundred pounds.

metakuna's blog

Discussion about this post