Different British nuclear reactors?

[Rule of cool]

I keep bumping into this from different angles, its technically very interesting to me so I'll ask the question here and see what happens.

In the late 50s Britain was looking for a more commercially oriented nuclear power reactor type to follow on from it's operationally successful but not very commercially successful 'Magnox' series reactors, the first of which had been the worlds first nuclear power station. To this end they built 4 prototype nuclear reactors, 3 of which were connected to the grid.

Name	Location	Coolant	Neutron Moderator	Energy-thermal/electric	start operations	cease operations
Prototype Fast Reactor (PFR) loop-type Fast Breeder	Dounreay	Sodium–potassium alloy	no moderator	60 MWt and 14 MWe	14-Nov-59​	Mar-77​
Windscale Advanced Gas-cooled Reactor (WAGR)	Sellafield/Windscale	Carbon dioxide cooled	graphite moderator	100 MWt and 30 MWe	9/08/1962​	1982​
Dragon high-temperature gas-cooled reactor (HTGR)	AEE Winfrith	Helium Cooled	graphite moderator	20 MWt	1/01/1964​	1976​
Steam Generating Heavy Water Reactor (SGHWR)	AEE Winfrith	light water cooled	heavy water moderator	100 MWe	1967​	Oct-90​

In 1965, using reasoning that if 'refueled online', so called 'British conditions' it would be cheaper to generate electricity than coal, the Central Electricity Generation Board selected the (Magnox derived) Advanced Gas-cooled Reactor design for full scale roll-out. In the event this was badly botched and the AGR proved complex, difficult and expensive to build, not competitive with US PWRs and BWRs on the export market nor meet its commercial promises.

In contrast the South of Scotland Electricity board (as opposed to North of Scotland Hydro-electric Board) preferred the Steam Generating Heavy Water Reactor type, which incidentally was the only nuclear reactor in the period built on time and on budget. In 1970 Australia selected the SGHWR for its Jervis Bay reactor project, only to cancel it a year later. In 1974, looking to build another batch of power stations and the AGR programme mired in problems the SGHWR reactor was selected for future builds, but this decision was reversed in 1976 and the AGR was persevered with for the next nuclear power station.

Given Australia and the SSEB preferred the SGHWR and the AGR was chosen under dodgy pretenses would Britain have been better off to bite the bullet with the SGHWR back in the 60s?

 

[Grey Havoc]

A bit of background on AEE Winfrith for those who may not be familiar with it:

Winfrith - Wikipedia

en.wikipedia.org

 

[Rule of cool]

I couldn't believe how many nuclear reactors the UK built in the 50s and 60s. They had something like 40 research reactors, those 4 prototypes, another Fast Breeder Reactor as well as 26 Magnox power reactors.

 

[Grey Havoc]

It was the Cold War, they needed all the edge they could get, both military and civilian.

 

[Yellow Palace]

North of Scotland Hydro-electric Board

The Hydro Board also preferred the SGHWR, oddly enough. They planned to build one at Stake Ness near Banff in... well, Banffshire. Instead they built an oil fired plant at Peterhead.

AFAIK, there was a certain amount of politics in the selection of the AGR for the third round of power plants. UKAEA was heavily pushing the AGR, but the CEGB was looking at open tenders and licence-built BWRs were very attractive. The UKAEA heavily pushed the advantages of on-load refuelling, which the AGR could do but the BWR couldn't, making their candidate look better. Of course, the AGR's on-load refuelling capability turned out not to be the great advantage that was anticipated.

The same dance repeated itself with the second phase of AGRs. The CEGB actually liked the SGHWR a lot, but the AGR was still the UKAEA's preference.

Unfortunately the timelines don't quite work out for the SGHWR to replace the first round of AGR construction. Construction started on Dungeness B, the first production AGR, in 1965 - two years before the prototype SGHWR came online.

Giving the SGHWR the nod for Torness and Heysham 2 is distinctly possible - in fact, it was expected. Sizewell B was also initially planned as a SGHWR plant. If these are successful, the proposed Hinkley Point C, Wylfa B, and Sizewell C plants also may get built as SGHWRs rather than the unbuilt PWRs of OTL.

One interesting wrinkle in all of this is that the AGR was designed to work alongside a Commercial Fast Reactor. While construction of this wasn't planned until after sufficient experience with the Prototype Fast Reactor at Dounreay, design studies had been carried out and discussions about suitable sites were beginning. The CEGB was unwilling to fund the project or provide a site, but was willing to buy any power generated; Winfrith was looking likely.

 

[PMN1]

An interesting read is the 1985? 'Going Critical: An Unofficial History of British Nuclear Power' by Walter C. Patterson

 

[zen]

Question
In AGR use, the need for enriched uranium would tap into mass enrichment....commercially justifying such large scale enrichment that could also feed into.....further enrichment for 'other uses'.
Does weapons production influence this decision?

Ironic that PWR could have become the SSN choice earlier and fed back into civilian decision making.

 

[ArtosStark]

Interestingly, I believe the Winfrith site was one of only two SGHWR reactor in the world that could claim a measure of success. Despite experience with the CANDU HW Cooled and Moderated design, the Quebec attempt with Gentilly-1 was shut down after having only 180 on-power days over 7 years of operation. The Italian CIRENE pushed past the end of Italian Nuclear use and was completed but never commissioned. The Advanced CANDU design from the 2000’s never got enough traction to be built. The only other one that had a similar lifespan was the Japanese Fugen Advanced Reactor. Though it had some fairly high profile safety issues come up in the 90’s and 2000’s.

Perhaps there is scope to work with Canada on development? Though expensive to initially fill with heavy water, CANDU-6 definitely worked, and it could possibly de-risk the project slightly to have a multi-reactor facility with a CANDU and the new SGHWR? That way even if one fails you still get a useful amount of power added to the grid. And, if the SGHWR design works, Canada may be willing to buy that in exchange, since their own attempt at Gentilly failed.

 

[Rule of cool]

One interesting wrinkle in all of this is that the AGR was designed to work alongside a Commercial Fast Reactor. While construction of this wasn't planned until after sufficient experience with the Prototype Fast Reactor at Dounreay, design studies had been carried out and discussions about suitable sites were beginning. The CEGB was unwilling to fund the project or provide a site, but was willing to buy any power generated; Winfrith was looking likely.

What was the role of the commercial fast reactor in the AGR system and was that system doomed to fail without it? Would a system built around SGHWRs need anything like that?

 

[Rule of cool]

Interestingly, I believe the Winfrith site was one of only two SGHWR reactor in the world that could claim a measure of success. Despite experience with the CANDU HW Cooled and Moderated design, the Quebec attempt with Gentilly-1 was shut down after having only 180 on-power days over 7 years of operation. The Italian CIRENE pushed past the end of Italian Nuclear use and was completed but never commissioned. The Advanced CANDU design from the 2000’s never got enough traction to be built. The only other one that had a similar lifespan was the Japanese Fugen Advanced Reactor. Though it had some fairly high profile safety issues come up in the 90’s and 2000’s.

Why is that? Was it because the British were starting from scratch in 1946 and built so many research reactors in the 50s?

 

[Grey Havoc]

What was the role of the commercial fast reactor in the AGR system

A combination of nuclear waste recycling and Plutonium production for both military & civilian purposes.

 

[Rule of cool]

The Hydro Board also preferred the SGHWR, oddly enough. They planned to build one at Stake Ness near Banff in... well, Banffshire. Instead they built an oil fired plant at Peterhead.

AFAIK, there was a certain amount of politics in the selection of the AGR for the third round of power plants. UKAEA was heavily pushing the AGR, but the CEGB was looking at open tenders and licence-built BWRs were very attractive. The UKAEA heavily pushed the advantages of on-load refuelling, which the AGR could do but the BWR couldn't, making their candidate look better. Of course, the AGR's on-load refuelling capability turned out not to be the great advantage that was anticipated.

The same dance repeated itself with the second phase of AGRs. The CEGB actually liked the SGHWR a lot, but the AGR was still the UKAEA's preference.

Unfortunately the timelines don't quite work out for the SGHWR to replace the first round of AGR construction. Construction started on Dungeness B, the first production AGR, in 1965 - two years before the prototype SGHWR came online.

Giving the SGHWR the nod for Torness and Heysham 2 is distinctly possible - in fact, it was expected. Sizewell B was also initially planned as a SGHWR plant. If these are successful, the proposed Hinkley Point C, Wylfa B, and Sizewell C plants also may get built as SGHWRs rather than the unbuilt PWRs of OTL.

Does the AGR decision need to be made and new generation power reactor ordered in 1965? What would the impact be holding off until 1968 when the 4th prototype was running, after all its not as if they WAGR had been running for long in 1965?

 

[Rule of cool]

A combination of nuclear waste recycling and Plutonium production for both military & civilian purposes.

I understand that the Magnox reactors were plutonium producers primarily, that they produced electricity was a side benefit. However IIUC by the early 60s the British were looking for reactors that were better commercial electricity producers than plutonium producers. Fast Reactors have a role to play in a large nuclear industry like Britain's in the 60s, but it sounded like the decision to go with the AGR fleet required the Commercial Fast Breeder as an essential components of that fleet's successful operation.

 

[Grey Havoc]

Does the AGR decision need to be made and new generation power reactor ordered in 1965? What would the impact be holding off until 1968 when the 4th prototype was running, after all its not as if they WAGR had been running for long in 1965?

I suspect that the newly established Ministry of Technology, and the Wilson government's policies in general (including related political infighting), had a lot to do with the rush.

 

[ArtosStark]

Why is that? Was it because the British were starting from scratch in 1946 and built so many research reactors in the 50s?

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).

Gentilly 1 was designed to be fueled by natural uranium, like all Canadian reactors. Canada did not have an enrichment industry and even if it wanted to set one up the U.S. likely would oppose it to avoid proliferation and a challenge to an established U.S. industry. Apparently the use of enriched Uranium in the British reactor allowed them to better deal with the void coefficient (again, don’t ask me why that is).

Which means that my suggestion probably wouldn’t work. The British were looking to move from a natural uranium reactor to more advanced enriched uranium concepts now that they were sure they would have enough supply and Canada was unlikely to want to invest in developing their own enrichment industry.

 

[Zoo Tycoon]

I understand that the Magnox reactors were plutonium producers primarily, that they produced electricity was a side benefit. However IIUC by the early 60s the British were looking for reactors that were better commercial electricity producers than plutonium producers. Fast Reactors have a role to play in a large nuclear industry like Britain's in the 60s, but it sounded like the decision to go with the AGR fleet required the Commercial Fast Breeder as an essential components of that fleet's successful operation.

Magnox primarily plutonium producer, Yes and No, depending which side of the fence you’re on. In hindsight the Magnox stations provided many times there worth in power than PU239, but agreed, that may not have been the original intention. The last Magnox’s to be built didn’t have dedicated PU239 channels and most that did used them infrequently in their later operations. It was quite a clever scheme to support both and barring a few teething troubles worked out quite well. Magnox was needed quickly so they went for very conservative (low risk) secondary circuit steam turbine technology from the marine sector. This drove a lower temperature and a smaller pile. It was always the plan (from 57) to moved more slowly to higher efficiency purpose designed steam turbines, with larger piles not complicated with the dedicated PU239 channels. These of course were the AGR’s…. Gee to think the UK use to be able to plan and execute strategy like that.

 

[Rule of cool]

I'm learning a lot. The SGHWR was to use enriched uranium which appears to have something to do with why it could use a light water coolant when the CANDU used heavy water coolant.

In the earliest days Britain was faced with an array of Hobson choices; it could get enriched uranium, so had to use natural uranium. Natural uranium dictated either graphite or heavy water moderators, but Britain couldn't get heavy water so graphite it was. That said the British had the first of three 26MWt DIDO research reactors go critical in 1956 with HEU and heavy water coolant and moderator, so it appears that within a few years of their ~1947 decision that Britain was able to get hold of enough HEU and heavy water to decide to build 3 quite large research reactors that used them.

 

[Rule of cool]

Magnox primarily plutonium producer, Yes and No, depending which side of the fence you’re on. In hindsight the Magnox stations provided many times there worth in power than PU239, but agreed, that may not have been the original intention. The last Magnox’s to be built didn’t have dedicated PU239 channels and most that did used them infrequently in their later operations. It was quite a clever scheme to support both and barring a few teething troubles worked out quite well. Magnox was needed quickly so they went for very conservative (low risk) secondary circuit steam turbine technology from the marine sector. This drove a lower temperature and a smaller pile. It was always the plan (from 57) to moved more slowly to higher efficiency purpose designed steam turbines, with larger piles not complicated with the dedicated PU239 channels. These of course were the AGR’s…. Gee to think the UK use to be able to plan and execute strategy like that.

I'm a believer in the whole of government benefit of Magnox, not just plutonium and electricity production but also strategic value of an alternative energy source not beholden to established interests like the coal industry or to international shocks like the oil industry using the Suez canal.

Presumably by the 60s Britain had sufficient sources of fissile material for it's nuclear weapons programme that it wouldn't miss the AGR (and presumably the SGHWR) inability to inherently produce it? The breeder reactors at Dounreay would be a big part of this?

 

[Rule of cool]

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?

 

[starviking]

These of course were the AGR’s…. Gee to think the UK use to be able to plan and execute strategy like that.

Though, IIRC, the AGR build was significantly slowed down by industrial action - and the use of multiple prime contractors for various power plant builds.

 

[Zoo Tycoon]

Magnox PU239 production generally went well and smoothly after a few initial problems, indeed so much was available, several station with the necessary modifications were never used for this purpose. However I’ve been told (normal caveats apply) Tritium production was a different story. Capenhurst had the only two Magnox piles modified for Tritium production together with the supporting CXPP Tritium separation plant. A very long time was spent getting it working , which to be fair, it eventually did. Making it work and operating it had a big impact on the power production. I understand the much easier option was to overproduce Pu239 and swap it for American Tritium which itself actually came from Canada’s CANDU reactors…. where the US had an exclusive supply contract.

 

[Rule of cool]

Does anyone know why the prototype SGHWR was the only one built on time and budget?

 

[zen]

Heavy water supply....Norway maybe?
Infamously they had a facility for HW production in the 40's.....

 

[Rule of cool]

I found a reference in a 1934 edition of Nature that said I.C.I. could make heavy water at Billingham, but I've found nothing else.

It looks like by the mid 50s there was an active export market for Heavy Water. Norway imported US HW for its little research reactor in 1957, US HW was subject to end use inspection but Norway's wasn't so was more expensive. In 1959 Britain sold 20 tons Norwegian HW they no longer needed to Israel, who used it to fill Dimona in 1963, so Britain had HW to spare!

 

[aim9xray]

FWIW, Charles Stross has this fascinating account of a tour that he had at the AGR facility at Torness:

https://www.antipope.org/charlie/blog-static/rants/nothing-like-this-will-be-buil.html

 

[Scott Kenny]

Given Australia and the SSEB preferred the SGHWR and the AGR was chosen under dodgy pretenses would Britain have been better off to bite the bullet with the SGHWR back in the 60s?

Having worked with USN PWRs, I'd say yes they should have.

PWRs are a fancy coal-fired steam plant. Simple enough that a high school student can operate them.

 

[Rule of cool]

PWRs are a fancy coal-fired steam plant.

On that, I seem to recall reading that the steam from AGRs is just right that it can be used by standard coal type turbines. Obviously steam can be hotter or cooler, but apparently it can be wetter or drier and coal makes a certain type of steam and AGRs make the same but other reactors make different steam that would require different turbines. Or am I delusional?

 

[Yellow Palace]

What was the role of the commercial fast reactor in the AGR system and was that system doomed to fail without it? Would a system built around SGHWRs need anything like that?

The CFR was intended to burn plutonium produced in thermal reactors, thereby closing the fuel loop. In the 1960s and 1970s there was concern about running out of uranium if nuclear power continued to grow. In addition, certain design choices in the MAGNOX reactor meant that the fuel had to be reprocessed, creating an attitude in the UK's nuclear industry that reprocessing spent fuel was just what you did. That created the problem of what to do with the resulting reactor-grade plutonium.

Can't put it in weapons (for political and technical reasons. Wasteful to store it forever. So build fast reactors which use all the fissile material in nuclear fuel, and 'breed' fissile material from natural uranium to boot. Solves two problems for the price of one.

Also introduces about twenty million other problems which run up the cost to a ludicrous level, but that's life.

Does the AGR decision need to be made and new generation power reactor ordered in 1965? What would the impact be holding off until 1968 when the 4th prototype was running, after all its not as if they WAGR had been running for long in 1965?

Realistically, it probably means that a few more big MAGNOX plants, along the lines of Wylfa, get built in place of the first AGRs.

Capenhurst had the only two Magnox piles modified for Tritium production together with the supporting CXPP Tritium separation plant.

Chapelcross, surely? Capenhurst didn't have any reactors.

On that, I seem to recall reading that the steam from AGRs is just right that it can be used by standard coal type turbines. Obviously steam can be hotter or cooler, but apparently it can be wetter or drier and coal makes a certain type of steam and AGRs make the same but other reactors make different steam that would require different turbines. Or am I delusional?

Light water reactors (PWRs and BWRs), by the nature of their design, can only produce saturated (wet) steam in the primary loop. For reasons of thermodynamics, that means you're stuck with saturated steam in the secondary loop.

Gas-cooled reactors inherently have a much hotter primary loop, which means you can set up the heat exchangers to produce superheated steam in the secondary loop. This is more thermodynamically efficient in the turbines, and avoids blade erosion problems from suspended water droplets.

Coal and oil fired plants have used superheated steam for a very long time for these reasons. PWRs and BWRs can get away with having to use saturated steam because burning a few more neutrons isn't too big of a deal, and the additional turbine maintenance is partly offset by the pipework and heat exchangers running at more manageable conditions.

 

[NOMISYRRUC]

Does the AGR decision need to be made and new generation power reactor ordered in 1965? What would the impact be holding off until 1968 when the 4th prototype was running, after all its not as if they WAGR had been running for long in 1965?

Realistically, it probably means that a few more big MAGNOX plants, along the lines of Wylfa, get built in place of the first AGRs.

Would that be a good thing in itself? Because as I understand it the MAGNOX plants took a lot less time to build, cost less to build and weren't any more expensive to run than the AGRs.

 

[Scott Kenny]

Light water reactors (PWRs and BWRs), by the nature of their design, can only produce saturated (wet) steam in the primary loop. For reasons of thermodynamics, that means you're stuck with saturated steam in the secondary loop.

Gas-cooled reactors inherently have a much hotter primary loop, which means you can set up the heat exchangers to produce superheated steam in the secondary loop. This is more thermodynamically efficient in the turbines, and avoids blade erosion problems from suspended water droplets.

Coal and oil fired plants have used superheated steam for a very long time for these reasons. PWRs and BWRs can get away with having to use saturated steam because burning a few more neutrons isn't too big of a deal, and the additional turbine maintenance is partly offset by the pipework and heat exchangers running at more manageable conditions.

You can add electrical superheaters to get dry steam in the secondary. Don't know if anyone did with nuclear plants, but the USN did that a lot with their oil-fired steam plants.

 

[Yellow Palace]

Would that be a good thing in itself? Because as I understand it the MAGNOX plants took a lot less time to build, cost less to build and weren't any more expensive to run than the AGRs.

Depends on your point of view, really.

MAGNOX was seen, AFAIK, as being compromised by its history as a plutonium production plant, and it was thought that an optimised power production plant would be more attractive.

If the theoretical benefits of the AGR had materialised, then it would have been a sensible decision. With the economics as they were... building more MAGNOX, or just building BWRs, would have been more sensible.

Given the UKAEA's blatant preference for gas-cooled, graphite-moderated thermal reactors - see also Dragon - I'm not sure why the SGHWR got as far as it did.

You can add electrical superheaters to get dry steam in the secondary. Don't know if anyone did with nuclear plants, but the USN did that a lot with their oil-fired steam plants.

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.

Of course, the Soviet Union put oil fired superheaters on the secondary loop for the KIROV class.

 

[NOMISYRRUC]

Part of Post 18.

I'm a believer in the whole of government benefit of Magnox, not just plutonium and electricity production but also strategic value of an alternative energy source not beholden to established interests like the coal industry or to international shocks like the oil industry using the Suez canal.

All of Post 12.

Does the AGR decision need to be made and new generation power reactor ordered in 1965? What would the impact be holding off until 1968 when the 4th prototype was running, after all its not as if they WAGR had been running for long in 1965?

Part of Post 28.

Realistically, it probably means that a few more big MAGNOX plants, along the lines of Wylfa, get built in place of the first AGRs.

All of Post 29.

Would that be a good thing in itself? Because as I understand it the MAGNOX plants took a lot less time to build, cost less to build and weren't any more expensive to run than the AGRs.

Part of Post 31.

Depends on your point of view, really.

MAGNOX was seen, AFAIK, as being compromised by its history as a plutonium production plant, and it was thought that an optimised power production plant would be more attractive.

If the theoretical benefits of the AGR had materialised, then it would have been a sensible decision. With the economics as they were... building more MAGNOX, or just building BWRs, would have been more sensible.

I had to resort to Wikipedia for the following so the following may be wildly inaccurate.

The 11 MAGNOX power stations (with 26 reactors) took an average of 6 years to build. They were in operation for an average of 38 years. The shortest was Hunterston "A" at 26 years (1965-1991) and the longest was Calder Hall at 47 years (1956-2003). Wylfa (which was by far the largest) took 8 years to build (1963-71) and was in service for 44 years (1971-2015) Oldbury (1968-2012) was the joint third longest period in service. The 26 reactors could produce a total of 3,314 MWe. Wylfa's 2 reactors produced 980 MWe, which was 30% of the total.

A total of 7 AGR power stations (with 14 reactors) were built in two batches. The first batch of 5 stations was begun 1965-70 and completed 1976-83. The second batch of 2 stations was begun in 1980 and completed in 1988.

• The contract for the first station (Dungeness "B") was let in 1965 for completion in 1970. Construction began in 1965, but it wasn't completed until 1983. Therefore, it took 18 years to build instead of 5 and was 13 years late.
• The next two (Hinkley Point "B" & Hunterston "B") were begun in 1967 and completed in 1976. Their shorter building time (9 years) was due to being built to a different design.
• Hartlepool and Heysham 1 were begun in 1968 & 1970 respectively and both were completed in 1983, but they didn't supply electricity to the National Grid until 1988, so depending upon how you measure it they took an average of 14 years to build or an average of 19 years to build.
• Construction of the final pair of AGRs (Heysham 2 & Torness) was begun in 1980 and both were completed in 1988. They "only" took 8 years to build, but that was still 3 years longer than Dungeness "B" should have taken to build, when usually the more you build, the less time it takes to build & the building cost reduces due to the "economies of scale" and "learning curve" effects.
• The next nuclear power station to be built in the UK was Sizewell "B" (a PWR that produced 1,198 MWe) which was begun in 1988 and completed (7 years later) in 1995.

The average building time for the 5 AGR plants begun 1965-70 was 13 years, which increases to 15 years if the completion date for Hartlepool & Heysham 1 is counted as 1988 instead of 1983. The 3 stations so far decommissioned were in operation for an average of 43 years and the 4 still operational have been producing electricity for an average of 36 or 39 years depending upon whether you count the completion date of Hartlepool & Heysham 1 as 1983 or 1988.

I found the articles on the AGR power stations confusing. All 7 stations seem to have been designed to produce 1,320 MWe, but what they could put into the National Grid seems to be about 1,200 MWe, which meant they could have produced 14,400 MWe when all four were operational. [See Spot the Mistakes No.1]

Which brings us back to what @Rule of cool wrote in Post 18 . . .

If the first 5 AGRs had taken 5 years to build and worked as advertised the amount of nuclear generated electricity produced in 1975 would have been increased five-fold to about 15,000 MWe and 4 of them would have been operational when the Oil Crisis of October 1973 happened. [See Spot the Mistakes No.2]

Only one AGR (Dungeness "B") would have been operational during the 1972 Miners' Strike (09.01.72 to 28.02.72) but 4 out of 5 would have been operational during the economic & political crisis of November 1973 to March 1974.

• The NUM would have been in a weaker negotiating position.
• Would it's members reject the OTL pay offer in November 1973 and introduced an overtime ban?
• Would we have had the Three Day Week which was in force from 01.01.74 to 07.03.74 IOTL?
• Would the miner's have gone on strike from 05.02.74 to 06.03.74?
• Would Edward Heath have called the General Election held o 28.02.74 and would he have won?
• Would Margaret Thatcher have become leader of the Conservative Party in 1975 and Prime Minister in 1979?

IOTL the prototype AGR (at Windscale) was completed in 1962 and the first production station (Dungeness "B") was ordered in 1965. Therefore, my guess is that if they waited until 1970 (i.e. 3 years after Winfrith the prototype SGWHR was completed) then Heysham 1 would have been the first production SGHWR and the 4 power stations begun 1965-68 would have been copies of Wylfa or enlarged Wylfas producing about 1,200 MWe each. None would have been operational before the 1972 Miners' Strike if they took as long to build as Wylfa (8 years) only Dungeness "B" would have been operational by the end of 1973, with the other 3 coming into service 1975-76 so their effect on events between late 1973 and early 1974 are likely to have been negligible.

Spot the Mistakes No.1

I found the articles on the AGR power stations confusing. All 7 stations seem to have been designed to produce 1,320 MWe, but what they could put into the National Grid seems to be about 1,200 MWe, which meant they could have produced 14,400 MWe when all four were operational.

Should have read.

I found the articles on the AGR power stations confusing. All 7 stations seem to have been designed to produce 1,320 MWe, but what they could put into the National Grid seems to be about 1,200 MWe, which meant they could have produced 8,400 MWe when all 7 were operational.

Spot the Mistakes No.2

If the first 5 AGRs had taken 5 years to build and worked as advertised the amount of nuclear generated electricity produced in 1975 would have been increased five-fold to about 15,000 MWe and 4 of them would have been operational when the Oil Crisis of October 1973 happened.

Should read.

If the first 5 AGRs had taken 5 years to build and worked as advertised the amount of nuclear generated electricity produced in 1975 would have been increased nearly three-fold to about 9,000 MWe and 4 of them would have been operational when the Oil Crisis of October 1973 happened.

 

[Hood]

The £89 million contract for Dungeness B was awarded in August 1965 to Atomic Power Constructions, a consortium backed by Crompton Parkinson, Fairey Engineering, International Combustion (Holdings) and Richardsons, Westgarth & Co.

This consortium suffered considerable losses on the project when in 1969 the management passed to the British Nuclear Design and Construction Co., a consortium of English Electric, Babcock and Wilcox and Taylor Woodrow Construction - although the project was still nominally that of Atomic Power Constructions. This seems to have been the result of a reorganisation of the nuclear industry, with Fairey Engineering and International Combustion leaving the industry.

British Nuclear Design and Construction built Hartlepool; construction was delayed in 1970, when the Nuclear Installations Inspectorate declared that they were unhappy with part of the boiler design, that added £25 million to the price tag. They also built Sizewell A.
English Electric Co. and Whessoe were involved in the Dounreay Fast Breeder construction. Babcock and Wilcox were involved with Winfrith.

The other construction group was the Nuclear Power Group, a consortium of Nuclear Power Plant Co. Ltd. and AEI-John Thompson Nuclear Energy Co. Ltd. They built 6 of the 11 initial power stations built, including Berkeley, Bradwell and Dungeness A. They also built Hinkley Point B and Hunterston B and Heysham 1 (Interestingly Heysham 1 and 2 have different turbine sets; 1's is GEC, 2's is NEI-Parsons).

In 1972 British Nuclear Design and Construction and Nuclear Power Group were merged as the National Nuclear Corporation. Shareholdings were: GEC 50%, UKAEA 15% and the remained held by British Nuclear Associate on behalf of Babcock and Wilcox, Clarke-Chapman John Thompson, Head Wrightson, Sir Robert McAlpine and Sons, Strachan and Henshaw, Taylor Woodrow Construction, Whessoe.
National Nuclear Corporation was responsible for Heysham 2 and Torness.

So the AGRs began in a period of buggins turn between the NDC and NPG with the upstart APC forced out altogether. Then the industry from 1972 was effectively unified but by then most of the contracts had been let and the AGR programme was winding down, in terms of new-builds. You would have thought that construction might have been quickened by experienced gained but that doesn't seem to be the case.


A diagram of the SGHWR at Winfrith

 

[Yellow Palace]

I found the articles on the AGR power stations confusing. All 7 stations seem to have been designed to produce 1,320 MWe, but what they could put into the National Grid seems to be about 1,200 MWe, which meant they could have produced 14,400 MWe when all four were operational.

This confused me for a while. It seems to be a difference between gross and net power - that is, the turbines were designed to produce 660 MWe, but the plant itself consumed 45 MW of that, leaving 615 MWe to go to the grid.

And, if anyone's keeping track, the actual reactor was nominally 1500 MWth, for an overall efficiency of 44% gross or 41% net.

You would have thought that construction might have been quickened by experienced gained but that doesn't seem to be the case.

Standardising the blasted things from Day One, instead of each plant being different, would have been a good idea. For both Magnox and AGR. Although inevitably they'd have standardised on the worst iteration of both designs.

 

[Zoo Tycoon]

Depends on your point of view, really.

MAGNOX was seen, AFAIK, as being compromised by its history as a plutonium production plant, and it was thought that an optimised power production plant would be more attractive.

The last Magnox stations built were incapable of producing weapons grade plutonium, had concrete pressure vessels featuring inherent flow convection cooling in case of recirculating fan failure, permitted almost Candu levels of access to critical core components, all of which enable on line times well in excess of any of the AGR’s. Hence they gave the lowest cost electricity of any nuclear station in the UK. I understand many in the business considered the mistake was not building more of the type.

 

[Yellow Palace]

I understand many in the business considered the mistake was not building more of the type.

I think this may be a case where the operator and the designer perspective differ. CEGB presumably held the opinion you state, and just wanted something that turned neutrons into electrons cheaply and reliably.

UKAEA held the opinion that a shiny new design incorporating all the new ideas they'd had would obviously be much better at it than that old rubbish.

UKAEA seem to have had more influence.

 

[Rule of cool]

The £89 million contract for Dungeness B was awarded in August 1965 to Atomic Power Constructions, a consortium backed by Crompton Parkinson, Fairey Engineering, International Combustion (Holdings) and Richardsons, Westgarth & Co.

This consortium suffered considerable losses on the project when in 1969 the management passed to the British Nuclear Design and Construction Co., a consortium of English Electric, Babcock and Wilcox and Taylor Woodrow Construction - although the project was still nominally that of Atomic Power Constructions. This seems to have been the result of a reorganisation of the nuclear industry, with Fairey Engineering and International Combustion leaving the industry.

British Nuclear Design and Construction built Hartlepool; construction was delayed in 1970, when the Nuclear Installations Inspectorate declared that they were unhappy with part of the boiler design, that added £25 million to the price tag. They also built Sizewell A.
English Electric Co. and Whessoe were involved in the Dounreay Fast Breeder construction. Babcock and Wilcox were involved with Winfrith.

The other construction group was the Nuclear Power Group, a consortium of Nuclear Power Plant Co. Ltd. and AEI-John Thompson Nuclear Energy Co. Ltd. They built 6 of the 11 initial power stations built, including Berkeley, Bradwell and Dungeness A. They also built Hinkley Point B and Hunterston B and Heysham 1 (Interestingly Heysham 1 and 2 have different turbine sets; 1's is GEC, 2's is NEI-Parsons).

In 1972 British Nuclear Design and Construction and Nuclear Power Group were merged as the National Nuclear Corporation. Shareholdings were: GEC 50%, UKAEA 15% and the remained held by British Nuclear Associate on behalf of Babcock and Wilcox, Clarke-Chapman John Thompson, Head Wrightson, Sir Robert McAlpine and Sons, Strachan and Henshaw, Taylor Woodrow Construction, Whessoe.
National Nuclear Corporation was responsible for Heysham 2 and Torness.

So the AGRs began in a period of buggins turn between the NDC and NPG with the upstart APC forced out altogether. Then the industry from 1972 was effectively unified but by then most of the contracts had been let and the AGR programme was winding down, in terms of new-builds. You would have thought that construction might have been quickened by experienced gained but that doesn't seem to be the case.

.....Snip.....

My imperfect understanding is that Atomic Power Constructions threw in a dodgy tender for Dungeness B because they were in trouble and were going under. They could never have done what they offered and their failure as a company reflects this.

Another thing I've heard is that going with 3 building consortia was fundamentally the wrong approach, as not only did they use their own designs but they competed with each other for a finite number of suitable personnel and other resources. The 1972 situation where there is a single nuclear constructor would have been the best solution in retrospect, but it was felt in 1965 that competition in the sector would produce the best results.

Getting rid of APC doesn't seem like it would be too difficult, a bit of aggressive vetting of the company during he tender process could have eliminated them or at least put them last, likely getting Britain down to 2 consortia and reducing some of the problems.

Does the SGHWR design lend itself to bespoke, unique to the builder, modifications?

 

[Rule of cool]

All 7 stations seem to have been designed to produce 1,320 MWe, but what they could put into the National Grid seems to be about 1,200 MWe, which meant they could have produced 14,400 MWe when all four were operational. [See Spot the Mistakes No.1]

Yes, they've doubled up on the double up. Each reactor makes 600MWe, and each power station has 2 reactors so each station makes 1,200MWe. But the author of the article has mixed up 'reactor' with 'station' and doubled the power output again.

 

[Rule of cool]

This is a mashup of NOMISYRRUC's and Hood's post about construction times and construction companies, because I think it's important to paint the picture of a major reason behind the AGR's relative failure.

The contract for the first station (Dungeness "B") was let in 1965 for completion in 1970. Construction began in 1965, but it wasn't completed until 1983. Therefore, it took 18 years to build instead of 5 and was 13 years late.

The £89 million contract for Dungeness B was awarded in August 1965 to Atomic Power Constructions, a consortium backed by Crompton Parkinson, Fairey Engineering, International Combustion (Holdings) and Richardsons, Westgarth & Co.

This consortium suffered considerable losses on the project when in 1969 the management passed to the British Nuclear Design and Construction Co., a consortium of English Electric, Babcock and Wilcox and Taylor Woodrow Construction - although the project was still nominally that of Atomic Power Constructions.

The next two (Hinkley Point "B" & Hunterston "B") were begun in 1967 and completed in 1976. Their shorter building time (9 years) was due to being built to a different design.

Nuclear Power Group, a consortium of Nuclear Power Plant Co. Ltd. and AEI-John Thompson Nuclear Energy Co. Ltd. ............ built Hinkley Point B and Hunterston B

Hartlepool and Heysham 1 were begun in 1968 & 1970 respectively and both were completed in 1983, but they didn't supply electricity to the National Grid until 1988, so depending upon how you measure it they took an average of 14 years to build or an average of 19 years to build.

British Nuclear Design and Construction Co., a consortium of English Electric, Babcock and Wilcox and Taylor Woodrow Construction........ built Hartlepool; construction was delayed in 1970, when the Nuclear Installations Inspectorate declared that they were unhappy with part of the boiler design, that added £25 million to the price tag.

Nuclear Power Group, a consortium of Nuclear Power Plant Co. Ltd. and AEI-John Thompson Nuclear Energy Co. Ltd. ..... built ..... Heysham 1

Construction of the final pair of AGRs (Heysham 2 & Torness) was begun in 1980 and both were completed in 1988. They "only" took 8 years to build, but that was still 3 years longer than Dungeness "B" should have taken to build, when usually the more you build, the less time it takes to build & the building cost reduces due to the "economies of scale" and "learning curve" effects.

In 1972 British Nuclear Design and Construction and Nuclear Power Group were merged as the National Nuclear Corporation. .......
National Nuclear Corporation was responsible for Heysham 2 and Torness.

(Interestingly Heysham 1 and 2 have different turbine sets; 1's is GEC, 2's is NEI-Parsons).

I think there's a much better path to AGR (and presumably SGHWR) construction in there, APC has to be eliminated for starters.

 

[Rule of cool]

Has anyone else seen this YouTube documentary?

At minute 14 it says that the almost bankrupt Atomic Power Constructions was employed to do a consultancy, they threw in a proposal based on doubling the number of rods from the original AGR design. It was this proposal from a company that couldn't physically deliver it that caused the AGR to be selected over a US light water design, according to this video.

This was a couple of years too early for the SGHWR.

View: https://www.youtube.com/watch?v=p-wrruwzASc