Last month I had the pleasure of visiting Torness nuclear power station in Scotland – many thanks to EDF for hosting me and giving me a detailed tour. I got to both stand on top of one of the reactors and touch the side of it, close to the gas injection pipes. (I joked to an anti-nuclear friend that when I did so I could see all the bones in my hand…obviously I couldn’t!)
When I write my blogs I often go to Adobe Stock to buy the base images I use to make the pages look pretty (I then combine at least two in order to create an original artwork). I try to make these images relevant to the subject, but have noticed when I search for “nuclear” many of the results are scary looking men in hazmat suits, and gas masks, with an evil-looking green miasma swirling around. Clearly for use in anti-nuclear propaganda.
While I know a fair bit about how nuclear power works, reading about it is no substitute for seeing it in practice.
I arrived on site on a sunny but windy day in late February. I had been instructed to leave my phone and other electronic devices in my car, and bring photo ID to the security gate, where I was given my visitor’s pass, had my bag scanned and was subjected to an airport-style security search. I was then met by a nice lady wearing a “Responsible Escort” lanyard, which made me wonder if I would meet an “Irresponsible Escort” at some point in the day and what that might look like(!).
She took me to the office of station director, Paul Forrest, where I was met by Paul and Fiona McCall who is the Senior External Affairs Manager for the existing EDF nuclear fleet. After getting kitted up with PPE (I had my own steel toe-capped boots and safety glasses and was provided with a boiler suit, hard hat, ear defenders and work gloves although I ended up having to take the gloves off since I started getting an allergic reaction to them).
First of all we took a walk past the workshop where some of the 750 on-site staff repair parts or inspect new parts ready for installation, before heading directly to the nuclear part of the site.
Understandably, security was high, both personal and for the site. We crossed over a line which marked the point at which PPE was mandatory and went to collect our dosimeters (radiation detectors) which had to be electronically registered and associated to our security passes. Fiona then went through a full height turnstile to enter the nuclear controlled area and I was directed to enter through a heavy security door. When Paul tried to follow through the turnstile, he found he was unable to enter: his dosimeter and security pass had stopped talking to each other and he had to go away and get them re-connected. Not even the station director gets to by-pass security protocols.
Fiona took me to the charge hall where she explained what it is while we waited for Paul to re-join us. This is the tallest room on the site, named after the machine which loads and unloads fuel into the reactors – charging them up. At each end of the room, at floor level, is the top of the containment for each of the two Torness reactors.
The charge machine, painted a fetching shade of pale green, trundles up and down between the two reactors to carry out its functions (during my visit both reactors were in use, but the charge machine was on a maintenance outage). The machine (and room housing it) is tall enough to accommodate the 10m long fuel rods known as “stringers” – the crane is positioned above the reactor, extracts an old fuel rod, turns round, and inserts a new one, in a seamless motion. Obviously, this takes place during re-fuelling outages and not while the reactors are running!
Reactor 2 is about to go off-line for a re-fuelling outage, and then reactor 1 will go off for a statutory outage, a period of around 65 days when some 13,000 – 15,000 tasks that are completed in order to ensure that the plant remains compliant with all relevant regulations and maintains a safe operating environment. Each statutory outage involves a £20-25 million investment in the plant.
While the tasteful green colour of the charge machine has no particular significance, other colours used on site have more meaning, and not just the company logo! Plant painted pink represents a single point of vulnerability making them easy to identify. Also, in order to avoid removing the wrong kidney /amputating the wrong leg type of mishaps, all the paperwork for reactor 1 is on yellow paper and that for reactor 2 is on blue, so it is immediately obvious to which reactor any given piece documentation relates.
By now, Paul had re-joined us, and we went to stand on one of the reactors, 32 meters above ground (the bottom of the containment is 12m below ground level). Of course, we were not standing directly on the reactor – we’re not idiots – but on top of the containment, in this case, 18 inch thick steel blocks known as the “pile cap” which were designed at the time of initial construction (ie long before 9/11) to withstand an aircraft impact.
Paul explained that the biggest risk is not from a passenger jet, since although they are very large, they travel relatively slowly by aeroplane standards, but from Tornado fighter jets whose heavy, dense engine block could theoretically impact the reactor containment at very high speed. (There are some great pictures in this article, which I won’t reproduce for copyright reasons but are worth looking at, including of the pile cap.)
Reassuringly, there was no green miasma and none of us began to glow as you can see from the photo!
It was also quiet – the only noise was from fans required to cool the building, not from the reactor, but from the sun – the building has little insulation and large windows, meaning that even on cold sunny days, it can become quite warm inside. There were also no vibrations from the reactor. It’s cool, quiet and drama-free.
We then descended several floors to reach the reactor floor, where I got to touch the side of the reactor, next to one of the gas circulators. Also on Friday, the Guardian newspaper published some pictures of nuclear installations under the headline ‘A picture of hell’: inside the UK’s nuclear reactors – in pictures but going on to say the photographer found them “tranquil, beautiful and sinister”, one of which was taken at Torness, at the same spot on which I was standing.
At a basic level, nuclear reactions give off heat which is used to produce steam which turns the turbines which turn the generator which generates electricity. The way that the heat from the reaction is accessed is by passing gas through the reactor – it enters at 250oC (for a cold start this temperature is achieved through pressurising the gas, but otherwise it is warm from its previous passage through the reactor) and reaches 660oC by the time it exits. The big orange circle (in the Guardian picture) is the panel through which the gas (CO2) enters the reactor. You can also see the different phases of the three-phase supply used to power the reactor. A heat exchanger is used to heat water to produce steam.
Leaving the nuclear controlled area was even more involved than entering. This time we had to enter something similar to the full body scanners at airports to make sure we had not received any radiation contamination. The machine had a woman’s voice which initially kept telling me to “get closer” which was tricky since I was already pressed against the edge of the scanner. I had to place my hands and fore-arms inside special slots as well.
The voice then counted down from six (apparently the number is random either four, five, six or sometimes seven. It then comfortingly told me that the radiation detected was zero. In fact there were two such machines, and each time I had to stand facing into and then out of the scanner, and each time the result was the same: zero contamination. Luckily we had not been outside, since the cosmic rays to which we are exposed on a daily basis would result in a non-zero reading!
As we walked to the turbine hall, we discussed ramping and load following for nuclear. While the Torness reactor can be shut down in 4 seconds, it takes 12 hours from start-up to grid synchronisation, making it too slow for practical load following.
The reason for this length of time is the time needed to bring the plant up to temperature for steam production. Paul told me that while Torness could not sit at half-load for extended periods, newer reactors can, and of course, in France, reactors operate flexibly since as close to 80% of the electricity mix they need to be able to vary output in line with demand (load following).
So we then came to the turbine hall. Each reactor has its own dedicated turbine set, each in three parts operating with declining temperature and pressure until you reach the generator itself. The gas which circulates in each reactor heats water to create high temperature, high pressure steam which turns the high pressure turbine.
The steam is then re-cycled before passing through the intermediate pressure turbine, and is again re-cycled before passing into the low pressure turbine which then turns the generator to generate electricity. The casing of each of these turbines gets progressively less warm, as expected (yes, I touched all of them!)
Unlike the charge hall, the turbine hall was LOUD! Very LOUD! But this is conventional plant – turbines are not quiet, and you would not find the turbine halls of gas or coal power stations being any quieter (unless they were turned off!).
After lunch I visited the exhibition centre. The picture below is of a model of the plant, which shows all the different sections of the site and gives a sense of the scale of the operation (the little model people give an idea of how large the plant is).
I had an interesting discussion with the lady there about the control rods, as this was something I had forgotten to ask while I was inside the plant. The control rods are held by magnetic clasps which release instantaneously when needed to drop inside the reactor ending nuclear reactions in 4 seconds once inserted. Both the fuel rods and control rods pass through the core structure which is comprised of graphite bricks.
These bricks are the subject of much regulatory concern since the Office for Nuclear Regulation believes that they may become dislodged during an earthquake inhibiting the passage of control rods into the core. While this may be the case, it currently requires that 100% of the control rods be capable of entering the core in a 1-in-10,000 year earthquake – 10x larger than the largest ever recorded in the UK, despite the fact that only 15% are actually required to terminate the reaction and there are two other shut-down methods that could be engaged in the highly unlikely event that the control rods did not work. I discussed this in more detail in my report on nuclear power published last November.
In the exhibition centre I also learned that the fuel structures are profiled in order to allow the gases to move through the core more efficiently – everything is engineered down to the last detail. This schematic shows everything in greater detail – I suggest downloading it (click on the image) so it can be better magnified to see all of the detail.
People have all kinds of assumptions about nuclear power, most of which I think are mis-placed. The reality is that our nuclear power stations are extremely safe, and entirely drama free. The entire vibe is calm, and the plant was impressively clean for an industrial site.
It was clear that working to a high standard was embedded in the culture – in some organisations, safety rules are met will eye-rolling and a tick-box mentality, but at Torness it was clear that not only was basic industrial as well as nuclear safety taken very seriously, it is part of the operational DNA. There is a clear “take the time to do it right” mentality and a calm efficiency about the operations.
I had a really interesting and enjoyable day at Torness and would like to thank Paul and Fiona for showing me round what is a very impressive site.
“Obviously, this takes place during re-fuelling outages and not while the reactors are running!”
Actually though the AGRs were designed to be re-fuelled while running at 100% power. The reason why this was not possible in practice (although some have managed to do it at part load) is as follows:
1) A dropped stringer could break and obstruct the flow of gas.
2) As long as the reactor is shut down within about 10 seconds, this is fine but any longer and there is a risk of fuel melt and fission product release.
3) To ensure that this emergency shutdown is achieved, there is a weight sensor in the charge machine which can detect a dropped stringer and shutdown the reactor quickly.
4) However here is where the problem occurs because high pressure gas is flowing up through the channel while you lower the stringer in and this pushes the stringer and reduces the weight load on the sensor.
5) They were never able to get to a point where the control system had the required sensitivity to a genuine drop without nuisance tripping from gas flow at any point above 50% power level.
In addition to this, there were turbulence issues which kept it below 30%.
I think even the stations where they got it working for part load don’t do it anymore as it isn’t worth the hassle but it was definitely a design goal.
Pleased you had a good time visiting Torness. I was privileged to work on the design of Torness, alongside many hundreds, if not thousands more, working for the National Nuclear Corporation back in the late 1970s and 80s. The attention to detail in the design is borne out in what you saw and experienced.
When i was in younger members Kent section of the IEE the CEGB were great supporters of providing tours of power stations in the South Eastern Division and that included Dungeness B. It sent quite a shiver down me to be standing on top of the reactor and all that power being contained just below you but as you say no noise and drama to accompany it that you got in the boiler house of fossil fulled stations. The AGRs were a drama initially but they served the UK well and i find it utterly depressing that we are now totally dependent on other countries if we want to build anymore
Thank you. Let’s look forward to Wind, Nuclear and Hydrogen, with fossil fuels for air travel etc.
Good Evening,
It’s been a while since I visited any nuclear power stations or research reactors – but your description is familiar. It all felt very safe. And if we could just keep the existing nuclear fleet going indefinitely I would support that 100%. But we seem to have a problem with building ‘New Nuclear’ and it’s probably for reasons that you appreciate better than I. The capital costs seem to be beyond what any private company can sustain and can only be borne by governments. And the plants are slow to build – with a very high risk of delay and cost overruns.
The prospect of small modular reactors (SMRs) is held out as a lifeline, but I am sceptical. ‘Small’ in the case of Rolls Royce seems to be over 500 MWe which is roughly the size of Torness. The UK doesn’t need that many such reactors and so the advantages of modular designs are unlikely to provide that great a saving. Personally, if they had the plans I would be happy if they just re-built the Magnox or AGRs exactly as they were built last time!
And then there is the question of the safe disposal of the high-level waste. In my understanding – despite having had a solid 60 years to address the issue of safe permanent disposal we have not permanently disposed of a single gram of high-level waste. This is a problem that really needs to be addressed *before* we build a new generation of reactors.
Best wishes
Michael
Security, Security, Security. Is the problem with single SMRs that the cost of security would outweigh the technological benefits?
Or is it that they would most likely be installed as multiple units at a large site, with the same total output as one/two larger reactor(s), such as those at Torness, and hence the cost of security would be no different in magnitude to current sites per GW output?
What are your thoughts?
There are some hopes that Heysham can be re-used for SMRs in part due to lack of space for another GW scale reactor. This would obviously enable scale for security. But down the line there are plans to embed SMRs in industrial sites, and while this makes a lot of sense, it relies in identifying workable security solutions. Building a nuclear reactor next to a chemical plant would increase the danger associated with a terrorist act significantly.
Please forgive the layman’s question. Is the inertia of the masive turbines & generators a factor in varying output? Could the nuclear core ramp up steam production to feed additional smaller ‘throttleable’ turbines in meet peaks in demand above base load?
The inertia of “massive” turbine/generators is a great benefit to the grid in managing variations in grid frequency and is true of nuclear, gas and coal fired plants, but not of solar or wind. And these variations in grid frequency do also require/produce variations in output from the turbine generator. The need to manage grid frequency becomes harder as more wind and solar are used, and although batteries can do a bit of this (keeping it simple) rotating steam turbine generators are actually much better doing this job. There is some feedback of this into the heat exchangers and nuclear reactor, but it is usually limited as most nuclear reactors are not geared to ramp load up and down at the speed associated with managing grid frequency. Some nuclear reactors can be designed to ramp up and down more quickly (submarine reactors come to mind) but if you have spent that much money on a nuclear reactor for electricity production you keep it going at as high a load as practical and safe. Hence they operate as base load – if they could ramp up they would do it 24/7 instead. So in conclusion I don’t believe your idea would work. Until we can overcome the need for electricity to be generated in balance with demand, we will always find a need for generating plant with differing characteristics to suit base load and peak load requirements. That then turns into a question of economics. You don’t want so much base load that you have to turn off a nuclear reactor simply to shed load off the grid. On the other hand you don’t want to be permanently in a position of running diesel generators or open cycle gas turbines to supply all the time (cheap to build, expensive to run). Until we crack the economics of storage of electricity at bulk (ie large enough to take on board all of the load above base load) the need for generators of different characteristics will remain. There are some options out there, but none have yet been sufficient to crack the size of the problem.
Inertia is not a factor in varying output – the aim is to keep the turbine running at a fixed speed (usually 1500rpm) to give a 50Hz output. Output is varied by changing how hard the turbine is pushing against the rest of the grid (trying to lift that 50Hz frequency or allow it to fall) with no significant change in speed. In fact inertia is helpful in lots of ways as the rotational energy in the turbine means that for short periods (seconds) the turbine can deliver more power as it slows down than the nuclear reactor sat behind it, which is handy if another power station suddenly fails as it gives people controlling the grid a little time to react.
As nuclear is run as baseload in the UK then basically the turbine is run as hard as it can be (given the rating of the reactor and the rating of the turbine / generator) and the actual rotational speed (i.e. the grid frequency) is set by other power stations running at part load which are ramped up and down to match the load and keep the frequency constant.
The problem with running traditional nuclear power stations at lower powers is that they aren’t designed for it and the nuclear reactions which occur at low powers create various elements which in turn affect the rate of the nuclear reaction and therefore the stability of the reactor. You can make various design decisions to make low power operation more stable but at the time the AGRs were designed there was no need to.
Torness has two reactors, each capable of generating 660 MW of electricity. The Rolls Royce SMR is (currently) slated as a single reactor unit giving 470 MW of electricity.
Kathryn….when do sleep ?
Phenomenal output in the past few weeks plus Podcasts & a hard hat tour of Torness with excellent commentary. A very interesting informative account that anyone with a modicum of engineering knowledge could appreciate. Great British Nuclear (BNG) are in discussions with EDF over the purchase of 100 acres of the 255 acre site at sister plant Heysham 2. Hartlepool is also being considered. Both are approved for new nuclear & will facilitate Rolls-Royce future SMR & AMR development. In the meantime I look forward to catching up on the latest Watt-Logic (my window on the world) topics in a timely fashion.
Thanks so much
Barry Wright, Lancashire.
Kathryn – an amazing post. Thank you so much. Outstanding detail, in layperson’s language.
A most interesting post. I have always been impressed with the AGRs and their performance.
I was considering that we keep hearing about issues with graphite blocks cracking; if we were re-building would it be a good idea to have thoe graphite material included withing the fuel bundle – so, effectively, reducing the time that the graphite spent in the reactor.
And I would like to point out that modern designs of reactors also include thermal energy stores; such buffering would allow better following of demand. The cost (of rocks) for such would be about £5/kWhr, if operating at 650 degC, as has been proven elsewhere – so really depends upon how long you would want to store energy for however many hours and how many buffers (at £5M/GHr)
Hi all…thanks Kathryn.
This has proved to be a great topic, generating many interesting replies. My thanks to Mathijs Van den berg for his excellent account of problems associated with on load refuelling (OLR) of AGR’s, responsible no doubt for an eye watering loss of revenue over the years.
Heysham 1 on the other hand was never a contender for OLR.
Constructed on a compact footprint with a seismic risk. Emergency reactor shut down coinciding with a power failure was always a major consideration. Clutches release moderator rods which drop under gravity into the reactor core, no power supply required.
Generators normally supply their own auxiliaries via adjacent station transformers, not so at Heysham 1.
Auxiliaries are supplied from a 132kv substation off site a mile away. Loss of this supply is backed up by 4 on site Rolls-Royce Olympus generating sets (3 mins run up time)
My overall view of the postings suggest big nuclear plants remain the favoured way forward in the UK. I would agree but supported by a robust high capacity supergrid network however a massive spend is required as intimated in the media today. I also note the reference to the difficulty of reactors to cope with flexing output to suit demand always was a problem. I assumed that high capacity storage addressed this as well demonstrated with the Dinorwig pumped storage scheme constructed some 40 years ago & still in operation today. The excellent reply posting by David Leigh (10/2/2024) identified over 100 suitable possible sites across the UK; surprised that pumped hydro is not raised in the discussion.
We are a small island with a population approaching 70 million. Energy security is vital moving forward. IMO we have everything that could lead to an independent national energy supply network providing low cost green electricity for all, lifting millions of consumers out of poverty. Independence is the key here, no need for often fragile supplies from Scandinavia & mainland Europe or gas imports from the USA or elsewhere.
Barry Wright, Lancashire.