Doing a bit more digging, the British built a gaseous diffusion Uranium enrichment plant at Capenhurst that started producing LEU from 1952, HEU from 1954 and then back to LEU for power reactor fuel from 1962. That would be where the fuel for the DIDO research reactors would have come from.

I have no idea where they got the heavy water moderator and coolant for these reactors. Had supplies become available by the early-mid 50s that the British could import?

Heavy Water was being produced in Canada since 1943 (originally intended for the British nuclear programme before being diverted to the Manhattan project) with the ZEEP reactor in Chalk River, Ontario going critical in 1945. High quality Graphite was originally being used as a regulator for the British atomic program in Harwell in the absence of sufficient quantities of heavy water being available while several test reactors were built in Ontario that proved the concept of Plutonium production, producing a few milligrams which was transformed into full scale production at Harwell producing 200g in its first month (as well as validating the material properties of graphite inside a reactor). We were also importing Heavy Water from Norway such that by the end of the 1950's we actually had a surplus and so agreed to sell Israel 20 tons in 1959.
 
I’ve looked more into the tender process that saw the AGR selected, as it seems to be the key event.

Atomic Power Constructions submitted a tender for the AGR to have 408 fuel channels with a lattice pitch of 394mm2, which IIUC was twice as many fuel channels (in a given area?) as the much smaller Windscale AGR prototype. Also IIUC there were other AGR tender submissions as well as submissions for license built US light-water reactors. IIUC this tight packed fuel rod arrangement meant that calculations could say the British AGR is the clear winner on economic grounds and Atomic Power Constructions got the contract to build Dungeness B. Nuclear Power (Dungeness B Station) - Hansard - UK Parliament

It is important to note that the 660MWe output of these AGRs allowed them to use standard coal and oil power station turbine-generators whereas the 300MWe Magnox could not.

Later versions of the AGR that were built by the other consortia had somewhat larger lattice pitch of 460mm2 in more or less same size reactor and about ¾ of the fuel rods (308, 324 & 332), the larger fuel spacing which gave longer slowing down distance for the neutrons. My guess this is the difficulty in scaling up the design that sources mention without providing detail. I’d also guess that the less dense reactor arrangements of the follow-on to Dungeness 2 were an evolution of this proposal and the other AGR designs had the half density rod arrangement of the WAGR prototype.

The announcement in Parliament only mentions the SGHWR to ask if that programme will be reduced in priority as a result of this decision.

Without the dodgy APC tender submission is there a chance the Government rejects the submissions and ask for a re-tender, possibly with the SGHWR in the mix?
 
After looking into it some more it appears the difference is in the fuel source.

Any liquid moderated reactor will have a void coefficient of reactivity (the amount the reactivity changes when voids (bubbles in this case) form. In a Pressurized Heavy Water reactor like CANDU this is negative and could apparently be fairly easily dealt with. But a light water cooled reactor like SGHWR has a positive void coefficient which was difficult to deal with (don’t ask me why, I am learning this as I go).
Okay, I've got this one. As per one of my college classes:

A negative void coefficient of reactivity means that as the reactor gets bubbles in it, the number of reactions goes down. You need the water to slow the neutrons down so they can react with uranium atoms. No water, no slow neutrons. No slow neutrons, no uranium atoms splitting.
This means that as you increase the electrical draw on the turbines, you need to pull the control rods OUT in order to maintain the increased power (increase power setting, then pull rods out). Makes a mostly self-regulating setup, once you're critical and making steam in the secondaries.

A positive void coefficient of reactivity means that as the reactor gets bubbles in it, the number of reactions goes up. Which makes more voids, which makes the reactivity go up more, which makes more voids... Until you get a Chernobyl.
This means that as you increase the electrical draw on the turbines, you need to push the rods in. But if you have bubbles in the control rod channels, you will have to push hard on those rods. Worse, if a graphite core gets too hot, the control rod channels may close down or shift and bind your control rods. Chernobyl again.​



That seems like an odd design decision for an oil-fired plant, when you've got combustion heat right there. Might make more sense for a LWR, if it can be made energy-positive.
As I understand it, the boilers had put all the heat into the steam that they could get out of the oil.


It is important to note that the 660MWe output of these AGRs allowed them to use standard coal and oil power station turbine-generators whereas the 300MWe Magnox could not.
Kinda depends on how many turbines you're spinning at your existing combustion plants. I mean, if a standard coal or oil plant uses some multiple of 300MWe generators (or some combination to get 300MWe), it's not a problem.

But yes, being able to use standard existing turbine designs would be a significant money saver.


Without the dodgy APC tender submission is there a chance the Government rejects the submissions and ask for a re-tender, possibly with the SGHWR in the mix?
Eh, maybe?
 
Kinda depends on how many turbines you're spinning at your existing combustion plants. I mean, if a standard coal or oil plant uses some multiple of 300MWe generators (or some combination to get 300MWe), it's not a problem.

But yes, being able to use standard existing turbine designs would be a significant money saver.

The vibe I get from offhand references is that Britain had a standard turbine-generator setup they used for coal plants. Interestingly enough the SGHWR that won the 1970 Australian tender was 650MWe, so my guess is that would have the same setup.

Eh, maybe?

The reason I bring it up is because not only was the coal industry fighting back so not only did they have to try to get electricity generated cheaply but also sell the design on the export market. If the AGR is beaten by a US light water reactor there is no possibility of export sales, so they might try for a 2nd round.
 
The vibe I get from offhand references is that Britain had a standard turbine-generator setup they used for coal plants. Interestingly enough the SGHWR that won the 1970 Australian tender was 650MWe, so my guess is that would have the same setup.
Exactly so.

In fact, there were a couple of generations of it. A standard 500 MWe turbine-generator set went into 10 coal and 4 oil fired plants ordered from 1961 onwards. From 1967, an enlarged 660 MWe set went into one coal, two oil, and the AGR plants. There seems to have been some talk of a 900 MWe supercritical unit in the 1980s, but that never came to fruition.

One of my idle fantasies is the six CEGB oil fired plants being built as Magnox (in place of Kingsnorth, Fawley, Pembroke, and Ince B) and AGR (in place of Grain and Littlebrook)
 
Exactly so.

In fact, there were a couple of generations of it. A standard 500 MWe turbine-generator set went into 10 coal and 4 oil fired plants ordered from 1961 onwards. From 1967, an enlarged 660 MWe set went into one coal, two oil, and the AGR plants. There seems to have been some talk of a 900 MWe supercritical unit in the 1980s, but that never came to fruition.

One of my idle fantasies is the six CEGB oil fired plants being built as Magnox (in place of Kingsnorth, Fawley, Pembroke, and Ince B) and AGR (in place of Grain and Littlebrook)
Would the 4 additional MAGNOX stations be copies of Wylfa? And if they were do you think they'd be operational before the Oil Crisis?
 
Would the 4 additional MAGNOX stations be copies of Wylfa? And if they were do you think they'd be operational before the Oil Crisis?
It would be seven Wylfa-sized stations. Almost certainly not copies, because of the competing consortia. I'd expect them to come online about a year later than the oil fired plants they're substituting for (Kingsnorth, Fawley, Pembroke and Ince) - noting that the delays to Ince probably wouldn't be subject to knock-ons so that would still come online about 1984.

Different sites would probably be needed for a couple of them, though all were considered for nuclear plants at various times.
 
Exactly so.

In fact, there were a couple of generations of it. A standard 500 MWe turbine-generator set went into 10 coal and 4 oil fired plants ordered from 1961 onwards. From 1967, an enlarged 660 MWe set went into one coal, two oil, and the AGR plants. There seems to have been some talk of a 900 MWe supercritical unit in the 1980s, but that never came to fruition.

One of my idle fantasies is the six CEGB oil fired plants being built as Magnox (in place of Kingsnorth, Fawley, Pembroke, and Ince B) and AGR (in place of Grain and Littlebrook)
As far as I can tell from the internet.

oil fired power stations.png

9 out of 11 MAGNOX stations had 2 units (including Wylfa which were 500 MWe) and all 7 AGRs had a pair of 660MW units. The first AGR was Dungeness "A" which was begun in 1965 and the next pair was Hinkley Point "B" & Hunterston "B" begun in 1967.

In that case wouldn't Kingsnorth have been a "Super Wylfa" MAGNOX station with four 500 MWe units instead of 2 and Fawley onwards have been AGR stations with between 2 and 4 units rated at 660 MWe?
 
In that case wouldn't Kingsnorth have been a "Super Wylfa" MAGNOX station with four 500 MWe units instead of 2 and Fawley onwards have been AGR stations with between 2 and 4 units rated at 660 MWe?
It depends how gung-ho you want to be about building AGR stations. I went with Magnox largely to be conservative, and to match the 500 MWe generator set capacity.

Grain was actually built as 3,300 MW from five units; I'm not sure when it was reduced to two units. In addition, Littlebrook is within the boundaries of Greater London; if a nuclear plant was substituted, a different location would definitely be needed. That totals four twin-unit plants, plus Torness and Heysham 2 - which might be AGRs or SHGWRs.

There were also Inverkip and Peterhead oil-fired stations in Scotland, under the SSEB and NoSHEB respectively. Inverkip had three 660 MWe units, with provision for a fourth, but barely ran between 1976 and 1988 when it was decommissioned. You could conceive of an AGR plant or two taking its place, but they might well wind up being white elephants.

Peterhead had two 660 MWe units; an alternative SGHWR plant at Stake Ness was seriously considered for a time. It had the significant advantage of being located close to the landfall of most of the North Sea pipelines, making it somewhat comparable to the pithead coal fired plants. Stake Ness was probably a hard sell in comparison.
Half as a joke and half seriously . . . how many CVA.01s and TSR.2s would the money spent on the AGR programme have bought?
You'd have to allow for the cost of providing alternative generating capacity, whether coal, oil or light-water reactors. A 1977 paper argued that the AGR programme had cost £2.1 billion in then-year pounds more than an equivalent LWR programme. Whether that's capital expenditure, or expected lifetime operating costs, I'm not sure.

In 1980, Torness was estimated to cost £1.1 billion to build - suggesting that the £2.1 billion figure might be capital expenditure. Some very rough estimating suggests that the difference in cost between an AGR plant and an LWR plant is comparable to the cost of one aircraft carrier. And there were seven of them.
 
In almost all nuclear-power countries there was in the 1960s a hughe and sometimes very heated, if generally muffled debate within the nuclear industry on what kind of 2nd-generation power reactors to built in large numbers in the '70s and beyhond. It was especially the case in countries that had started to create a national deterrent and supposedly would need large quantities of WG fissile material and had a nascent nuclear power industry (UK, France), countries which had made the political decision to produce nuclear weapons but not yet implemented it (Switzerland), countries which were seriously considering to do it in a pre-NPT era (Sweden, Italy, Japan), and those which had hiden nuclear weapon aspirations (Spain, Argentina). Everywhere the confrontation was between:
- on the one hand, the proponents of (very) large static-moderator, natural U reactors of national design, led by the atomic research authority, made of the military (interested in WG material production), radiochemistry industry, construction industry (those reactors were first and foremost large concrete structures) and mechanical industry manufacturers that had in their catalogues relatively small conventionnal steam-turbines for thermal power-stations ;
- on the other hands the proponent of (mostly) US technology LWR : utilities and power-generation companies, heavy industries able to produce large forged components (pressure vessels, steam generators) and electromechanical industry. They wanted LWR using low-enriched fuel because they were safer, could be built faster, and were less expensive, both to built and operate. Having to stop operations for refuelling was not an issue for power companies because they would need to do it anyway for maintenance, as they did with thermal plants. Refuelling under power was in fact a nuisance to them.
Everywhere, by 1970, they had won, and imposed (mainly US technology) LWR, "helped" by some minor accidents in national-technology natural-U reactors (Sweden, France) or a serious one (Switzerland, Lucens 1969: worst reactor accident in Europe before Chornobyl).
Everywhere, except in the UK, where Tony Ben overruled the CEGB, prevented LWR under US licences, imposed the AGR, and led to the demise of English nuclear industry.
In the USSR the debate was surprisingly similar, the proponents of the natural U reactor RBMK calling disparagingly the Soviet PWR, the VVER, "amerikanskii reaktor", except that in the end due to power-games in Moscow, it was decided to go ahead with both types. In communist Czecoslovakia Skoda had designed and built a national natural-U reactor. It suffered a major accident and Skoda became a VVER licensee.
The only exceptions were of course the US where AEC had decided already in 1957 that future commercial reactors would only be allowed to be LWR; Canada which had chosen CANDU in 1954 and had no industrial capability to manufacture large forged components; and of course Federal Germany. FG was the main proliferation concern of the US administration in the 60s and American pressure led to the early development of national LWR which were often superior to their US counterparts (Siemens Typ 72 BWR vs. GE BWR or KWU Konvoy PWR vs. Westinghouse "F" PWR).

PS : please note that in a CANDU reactor using the original natural U fuel, the void coefficient is indeed positive, albeit compensated by a negative temperature coefficient, and that BWRs have a single steam/water loop (hence, no "primary loop" there).
 
My version of that question is if SGHWR was the only reactor built on time and budget how does the public purse look if it was selected instead of the AGR,

Earlier in the thread someone posted that Dungeness B was to cost 89m GBP, and a design change to another plant was to cost 25m GBP. That said when they selected the SGHWR in 1974 it was subject to cost increases, but given 1974 was the era of stagflation that isn't surprising.,
 
How long would it take to properly investigate the validity of Atomic Power Constructions' bid with the tightly packed power rods and come to the proper solution that was used by the rest of the AGRs? Would it be long enough for the prototype SGHWR to come online and be part of the solution?
 
I don't think we should romanticize the Magnox, they weren't built to a standard design and had their own cost and schedule blowouts. The AGR was a logical development of the Magnox to meet the need for a more economical reactor type. My issue with the AGR is the implementation of the programme: the dodgy APC tender and all subsequent garbage. I'm starting to get the impression that a well executed AGR programme might be OK, and there's no need for the SGHWR. However I think that to properly roll out the AGR programme would take time and money which would erode its lead over the SGHWR.
 
I don't think we should romanticize the Magnox, they weren't built to a standard design and had their own cost and schedule blowouts
Magnox also had the 'delightful' feature that its spent fuel elements couldn't be safely stored long term, requiring reprocessing. There's a reason the UK had the largest, longest running civilian nuclear reprocessing programme.
 
Do you mean THORP at [Windscale] Sellafield?
B205, the Magnox Reprocessing Plant, which ran from 1964 to 2022. THORP only ran 1994 to 2018.
And the 2 fast reactors at Dounraey.
The Dounreay site had its own small scale reprocessing facility to handle fuel from the breeder reactors.
 
B205, the Magnox Reprocessing Plant, which ran from 1964 to 2022. THORP only ran 1994 to 2018.

The Dounreay site had its own small scale reprocessing facility to handle fuel from the breeder reactors.

A major reason why this is such an interesting topic for me is because of the completeness of Britain's nuclear industry by the early-mid 60s. The reason for this thread is to see if they could maintain this domestic progress into the 70s and beyond.
 
I found this picture on Reddit from Nuclear Power magazine, it's the Windscale AGR prototype. The description mentions 250 fuel channels in a reactor 14' (4.2m) x 15' (4.5m).

The Atomic Power Constructions AGR tender had 408 fuel channels in a reactor about 9m x 8m, which apparently was an untested proposal.
 

Attachments

  • 1730961440401.png
    1730961440401.png
    3.6 MB · Views: 16

Attachments

  • FB_IMG_1730965214451.jpg
    FB_IMG_1730965214451.jpg
    34.6 KB · Views: 14
  • FB_IMG_1730965218406.jpg
    FB_IMG_1730965218406.jpg
    35.8 KB · Views: 11
This cutaway clearly shows the arrangement of early Magnox designs which had the primary heat exchangers, and associated connecting pipework, outside the containment. 'Gamma shine' from activated coolant hadn't been anticipated in the design, so these plants had higher than expected radiation emissions.
 
I would have thought earlier than 2005, given the pre-70s styling of the signpost on the road.

You're very optimistic about the rate of road sign renewal in rural Scotland - there are signs of that vintage still doing sterling work in 2024!
I grew up in the area and can vouch for it being a bit of a backwater. Especially when you turn off the main roads onto ones like this photo was taken from. Once you leave the beaten track, not much changes down there!

Note, no steam emanating from the cooling towers as was usually the case when it was operating. So after June 2004 when it stopped electricity generation and before May 2007 when the cooling towers were demolished.

Chapelcross was built on the site of the old WW2 RAF Annan airfield that was used as a fighter OTU base. Some of the old runways and taxiways are still visible on Google Maps.
 
Chapelcross was built on the site of the old WW2 RAF Annan airfield that was used as a fighter OTU base. Some of the old runways and taxiways are still visible on Google Maps.
I'm absolutely convinced that there was provision for another four reactors at Chapelcross for a total of eight, plus the four at Calder Hall, running military fuel cycles. They would certainly fit within the footprint of the site.

Can I find the reference to this? Heck no!

There was also a map in New Scientist at some point in the 1970s or early 1980s showing sites identified for potential future reactors. There were... a lot.
 
I'm trying to come up with a realistic alternative path for the British nuclear industry.

The best I can come up is when Atomic Power Constructions comes up with their bid for the AGR with 408 fuel channel instead of 250 in the WAGR the British Government does not award them the tender. Instead, in light of APCs perilous financial position, the theoretical nature of APCs bid, the likelihood of a US light-water reactor design winning the competitive tender and the desire of the British Government to build a design they can export they put the tender on hold in early 1965 to undertake research on the expanded fuel channel AGR concept.

This research takes some time, and it is found that APCs proposal is flawed, the fuel channels are too close together and the optimum number is in the low 300s. During this period Atomic Power Constructions was absorbed by British Nuclear Design and Construction Co., leaving only 2 nuclear power construction consortia in Britain. Also during this time the Winfrith SGWHR construction is progressing extremely well, the only nuclear reactor being on time and on budget, and this reactor type was seen favorably by the 2 Scottish power generating boards.

In late 1966 the tender to build Dungeness B to a 'compromise' AGR design with 308 fuel channels was given to British Nuclear Design and Construction Co.. In 1968 Nuclear Power Group was given the contract to build the SGHWR the Scottish boards preferred design at Hunterston B. In 1969 a further AGR power station was ordered, from British Nuclear Design and Construction Co., identical to the Dungeness B plant.

In 1970 the Australian AEA selected the SGHWR design for its first power reactor at Jervis Bay and work began leveling the site. This export success led the British Government to decide that all future British power stations were to be SGHWR. A change in Australian PM in March 1971 caused a loss of support for the nuclear power plant on economic grounds, so the British government intervened to sweeten the deal and the project continued.

The rest writes itself.
 
Last edited:
Yet again, trawling through my vast numbers of image and documents downloaded over the years, found this....
I know it's Naval based, but anyway, here goes....
 

Attachments

  • 4301321.pdf
    8.1 MB · Views: 9

Similar threads

Please donate to support the forum.

Back
Top Bottom