navitron
 
Renewable Energy and Sustainability Forum
Welcome, Guest. Please login or register.

Login with username, password and session length
News: Anyone wishing to register as a new member on the forum is strongly recommended to use a "proper" email address - following recent spam/hack attempts on the forum, all security is set to "high", and "disposable" email addresses like Gmail, Yahoo and Hotmail tend to be viewed with suspicion, and the application rejected if there is any doubt whatsoever
 
Recent Articles: Navitron Partners With Solax to Help Create A More Sustainable Future | Navitron Calls for Increased Carbon Footprint Reduction In Light of Earth Overshoot Day | A plea from The David School - Issue 18
   Home   Help Search Login Register  
Pages: [1] 2 3 4 5 ... 10
 1 
 on: Today at 08:36:51 AM 
Started by dan_b - Last post by M
Small modular units are being considered as part of the economies of scale spoken about earlier but they still have the reprocessing element of cost to overcome.

In the current climate I cannot see how they would ever be viable economically.

Moxi

Yep, they seem to be well behind the curve already. Rolls Royce suggest that costs could eventually fall to ~65/MWh, around 2035, so higher than RE today. And that around 5GW of deployments would need subsidisng to get costs down to HPC prices (100/MWh), form an orderly queue for that first 5GW's! They also estimate the worldwide market at approx 65-85GW max, which sounds big, but current PV deployments in the world would match that generation today, and we can probably roll that much (400GWp) out every 3yrs or so going forward, and in some countries PV contracts are being issued today at around 80% cheaper than that hoped for 2035 figure from RR.

If 65/MWh SMR's were available today, I could certainly see some market, but as I said, they are behind the curve, and RR's figures suggest they'll never catch up - So expect the UK government to support it with subsidies (but not cheap PV and on-shore wind).   Wink

 2 
 on: Today at 08:27:39 AM 
Started by dan_b - Last post by M
I think a lot of the problems the French have had with the EPR is simply one of time - only 2 reactors started work between 1984 and 2007 when they started work at Flamanville, and there was a 5 year gap between when the last reactor being built (Civaux 2) entered commercial operation and the start of the new EPR. That's an awfully long time to keep people on gardening leave, and a hell of a lot of perishable skills will have been lost. One of the fundamental economic problems with nuclear is that the high safety standards and construction costs drive you to build very big reactors - of which you don't need very many, so there isn't much opportunity to learn how to do them better and as such you never get to move down the cost curve as you find out how to actually build it. In the 1970s and 80s the French were building a lot of reactors, so learnt how do do them well and they will have been pretty cheap. Once they stopped building lots of new reactors - because they didn't need the power any more - the wheels started to come off the economic model.

I appreciate that we can come up with lots of reasons and excuses why nuclear is still so expensive, but eventually, logic has to creep in and conclude that it's simply a very expensive form of generation, and always will be.

My negativity towards nuclear is simple, I supported it till the early years of this decade as it's cleaner than coal, low carbon, and cheaper than the cost of FF's when you include externalities.

But, my opinion was open to new information and a changing world, and during this decade we've seen the UK turn away from coal, we've seen an acceptance of externality costs and the slow start to carbon pricing, we've seen a public acceptance (massive support and minimal opposition) of RE, and we've seen costs for RE fall below nuclear.

So my position changed, and with relief I could happily withdraw my reluctant support for nuclear.

Since then, the myth of 'cheap' nuclear has been well and truly busted, with high prices, and companies withdrawing from negotiations because they can't get a high enough price to meet costs, and RE costs have continued to tumble, and now we see a rollout of more and more storage ideas, growing efficiencies, and falling costs. All the while, we still don't know how to deal with nuclear decommissioning. Why create a clean up problem for folk 100yrs down the line, when RE is already a far better option today.

I accept that their are many that argue that nuclear can still be cheaper (not cheap compared to where RE is already at), but after 60yrs of vast support for the industry, isn't there a point where we have to conclude it's a myth?

Fool me once, shame on you, fool me for 60yrs ...... shame on me.

 3 
 on: Today at 08:18:43 AM 
Started by dan_b - Last post by Moxi
The problem with nuclear isn't just the economics of the reactors though, the fuel reprocessing part of the fuel cycle is just a huge money pit and that needs to be rolled in alongside the generation otherwise you don't get the full picture.

Nuclear power doesn't work well in the free market, it barely worked under massive state sponsorship for the French, Japanese and us (ultimately China also ?) and now technology has advanced and consumer products are more energy efficient and we have industrial scale renewables the Nuclear fleets look expensive dirty and to a greater proportion of the general public dangerous.

Small modular units are being considered as part of the economies of scale spoken about earlier but they still have the reprocessing element of cost to overcome.

In the current climate I cannot see how they would ever be viable economically.

Moxi

 4 
 on: Today at 08:15:27 AM 
Started by M - Last post by M
The issue is that the current CFD process already provides for this. So far no detail has been provided that indicates that a new CFD would cause lower bids to be made.

Of course they would be lower. PV and on-shore wind are cheaper, but are excluded from the current CfD mechanism. Having a net-zero subsidy scheme would not encourage higher prices. CfD auctions include a max value for bids to be eligible, so you'd just set that at the max desirable price, and see how much lower the bids are.


For PV and on-shore wind, the real issue seems to be political, not financial - our current government seems to want to make these options impossible regardless of cost. Even if our government changes or at least changes its view the current CFD process provides a valid assurance as the strick price of any CFD can only be at or above the project costs + profit margin. To lower that strike price you increase the number of companies (additional competition) bidding and/or lower the cost base - planning, surcharges (rates), etc. You don't do it by offering additional profits in exceptional circumstances as these cannot be quantified and so cannot be priced.

Yes, our government doesn't support on-shore wind and PV, hence why they should. You can't use the status quo to argue against a change. I also find it strange to argue against a supportive scheme on the grounds that the government isn't supportive, that's not a valid argument.

Yes, to lower prices you need more competition, such as a net-zero subsidy CfD scheme for on-shore wind and PV.

No, you do increase competition by offering profits, especially if a base income (below the average wholesale price) is guaranteed so investment can be found at lower costs.


The current CFD process allows for the provider to make a profit, the issue is that the new structure would also allow a provider to make additional profits if the market was to go badly wrong (inflated prices) without any real downside as they keep the excess. The term 'windfall profits' is often used in such situations. This is where the detail in the report is to vague to know what they mean by their CFD floors and separate reconciliation periods.

I think you are getting confused here. If prices go higher, all companies make higher profits. The only reason why the existing CfD beneficiaries don't is because they have to pay subsidies back, as they are part of a subsidy scheme where it's accepted that they will receive net subsidies over the CfD period. So it's only fair that having made their acceptable profit (off the back of subsidy support), they don't make extra.

This is not a net subsidy scheme as the strike price would be below the average wholesale price, so would be net-zero, simply providing assurance. Companies not receiving subsidies are allowed to make profits?

You are also missing the benefit of such schemes within your example of large increases in prices, since more supply, including supply form net-zero subsidy schemes would act to reduce prices (supply and demand), so even if prices rose, they would rise less than otherwise, thanks to more RE generation.


The example I gave here was to show that both the old and new type of CFD would result in the same level of protection to the provider. So the new type of CFD has no downside, but the opportunity for additional profits if the opposite situation was to happen.

I think your use of additional profits is misleading. We want RE to be profitable, it's not a bad thing, and if prices rise, then RE without subsidies would benefit exactly the same, no cost to the subsidy pot, so why not allow for this with a net-zero subsidy scheme that simply provides a low floor price, below the average wholesale cost.

Regarding your example of the wunder tech lowering our prices, and therefore the 'cheap' PV and on-shore wind getting more subsidies, well so would off-shore wind and nuclear with subsidies above the current average. But have you 'tested' your idea?

I'm actually a supporter of taking an idea to the extreme to test it, so let's test yours.

We give on-shore wind a 35/MWh CfD, against current predictions of an average wholesale price of 50/MWh. So over a 6 month period we don't expect it to receive any net subsidies.

For the average to drop to 35 for 50 what would it take, what would this wunder tech have to deliver within the 15yr period?

1. 30% of leccy generation at 0/MWh. [(70% x 50) + (30% x 0)]

2. 50% of leccy generation at 20/MWh. [(50% x 50) + (50% x 20)]

3. 100% of leccy generation at 35/MWh.

Do any of those sound likely? If not, then whilst testing the idea (net-zero CfD) to the extreme is worth it, but raises no real concerns.


Please take a step back and have a think. Don't worry about profits, we want RE to be cheap and profitable. Don't worry about 'losing' if leccy prices drop significantly, it won't happen to that extreme, and even if it did, we'd have cheaper leccy. So the question is quite simple, should we help support more RE, by providing a net-zero subsidy scheme to give PV and on-shore wind a boost?

 5 
 on: January 17, 2019, 09:00:22 PM 
Started by paul149 - Last post by nowty
And I think in 2018 most of us got a little more sun than average too. fingers crossed!

 6 
 on: January 17, 2019, 08:50:06 PM 
Started by stannn - Last post by stannn
https://www.reddit.com/r/Morocco/comments/agur9w/solar_energy_in_morocco_dw_documentary_renewable/

 7 
 on: January 17, 2019, 08:20:13 PM 
Started by dan_b - Last post by RIT
I think a lot of the problems the French have had with the EPR is simply one of time - only 2 reactors started work between 1984 and 2007 when they started work at Flamanville, and there was a 5 year gap between when the last reactor being built (Civaux 2) entered commercial operation and the start of the new EPR. That's an awfully long time to keep people on gardening leave, and a hell of a lot of perishable skills will have been lost. One of the fundamental economic problems with nuclear is that the high safety standards and construction costs drive you to build very big reactors - of which you don't need very many, so there isn't much opportunity to learn how to do them better and as such you never get to move down the cost curve as you find out how to actually build it. In the 1970s and 80s the French were building a lot of reactors, so learnt how do do them well and they will have been pretty cheap. Once they stopped building lots of new reactors - because they didn't need the power any more - the wheels started to come off the economic model.

The French started very much with the same plan as the UK talked about when HPC was first planned. Get one built and then build a lot more. For France, the original plan was 20 sites/40 reactors, but the plans have since changed. The UK government did not seem to be planning, instead, the Conservatives just wanted nuclear and the LibDems just wanted their PR referendum.

 8 
 on: January 17, 2019, 08:01:39 PM 
Started by dan_b - Last post by pdf27
Mart, thats a bit rude, but I'll let it pass.

However, In what way did the French, with 70+ reactors, producing 90% of their electricity, "not make it work"? It seems an absurd claim.

You dont have to answer, as this is a dead end, with 2 totally different viewpoints. I won't reply.

Sorry, not rude at all to respond with facts even if you don't like them. You made a claim and I provided information to suggest it wasn't true. If it was a financially viable solution, then one of the countries you referenced as evidence to support your claim, would not be considering the exact opposite regarding its economical viability.

You then made a claim that you hadn't made a claim, and I quoted your 'claim' to show you had.

You've now claimed that it worked because it generated 90% of their leccy - but did it 'work', that's to say, was it done economically? Is the technology sustainable going forward?

EDF facing bankruptcy as decommissioning time for France's ageing nuclear fleet nears

I appreciate that you don't like me pushing back against your nuclear support, but I don't see why that's not allowed. Is it not allowed? Was it not expected when you posted your claim? One thing I have noticed on other forums is that those that promote/defend nuclear get 'upset' when pushback occurs - but if we don't discuss the matter and consider the facts and figures that are out there, then we won't learn anything and the same old claims about nuclear will be posted ad infinitum and may mislead.
I think a lot of the problems the French have had with the EPR is simply one of time - only 2 reactors started work between 1984 and 2007 when they started work at Flamanville, and there was a 5 year gap between when the last reactor being built (Civaux 2) entered commercial operation and the start of the new EPR. That's an awfully long time to keep people on gardening leave, and a hell of a lot of perishable skills will have been lost. One of the fundamental economic problems with nuclear is that the high safety standards and construction costs drive you to build very big reactors - of which you don't need very many, so there isn't much opportunity to learn how to do them better and as such you never get to move down the cost curve as you find out how to actually build it. In the 1970s and 80s the French were building a lot of reactors, so learnt how do do them well and they will have been pretty cheap. Once they stopped building lots of new reactors - because they didn't need the power any more - the wheels started to come off the economic model.

 9 
 on: January 17, 2019, 04:24:03 PM 
Started by M - Last post by RIT
Sorry, I'm not following.

The companies would have to compete for the lowest price or they'd risk losing out to the competition, hence there would be competition.

The issue is that the current CFD process already provides for this. So far no detail has been provided that indicates that a new CFD would cause lower bids to be made.

There is no CfD today for PV and on-shore wind as the government says they don't need support/subsidies. This idea is not to give them subsidies, as the price is expected to be below the average, it's to give them assurance going forward.

For PV and on-shore wind, the real issue seems to be political, not financial - our current government seems to want to make these options impossible regardless of cost. Even if our government changes or at least changes its view the current CFD process provides a valid assurance as the strick price of any CFD can only be at or above the project costs + profit margin. To lower that strike price you increase the number of companies (additional competition) bidding and/or lower the cost base - planning, surcharges (rates), etc. You don't do it by offering additional profits in exceptional circumstances as these cannot be quantified and so cannot be priced.

I think you are missing the point about 'new CfD' providers making a profit, that's not a bad thing, that's great, the more they make, the more RE that will roll out, and if prices rise, the CfD doesn't rise, so even less chance of receiving a subsidy. Remember the 'enemy' here going forward is subsidy payments, not profitable leccy generators.

The current CFD process allows for the provider to make a profit, the issue is that the new structure would also allow a provider to make additional profits if the market was to go badly wrong (inflated prices) without any real downside as they keep the excess. The term 'windfall profits' is often used in such situations. This is where the detail in the report is to vague to know what they mean by their CFD floors and separate reconciliation periods.

Regarding falling prices from a wunder invention, then yes, such as scheme as this would mean subsidies being paid out, but remember, the subsidies come from a pot we all pay into as part of our energy bills. If prices fell then we'd benefit that way, despite paying more subsidies to wind and PV. Also, and I'm not belittling your wunder suggestion, it could happen, but as you say 'in 10yrs time' well that's 2/3rds of the way through the RE CfD's, so it's really the 35yr 100/MWh HPC deal you should be worrying about, not a 35/MWh wind deal.

The example I gave here was to show that both the old and new type of CFD would result in the same level of protection to the provider. So the new type of CFD has no downside, but the opportunity for additional profits if the opposite situation was to happen.

 10 
 on: January 17, 2019, 03:34:24 PM 
Started by M - Last post by M
I saw that part of the text, but am not sure what it really means, in terms of payments and clawbacks as independent six month periods is what caused me to raise the question regarding 'averages'.

Perhaps best to ignore averages, I only mentioned them as averages are a way to estimate how much subsidy a CfD will pay out.

In reality, under CfD's a supplier sells their leccy on the market and get a price. That price is then ignored (to avoid cheating)! The market price for that time slot is then used as the sell price, and if it's less than the CfD strike price, the difference is made up by the subsidy. Looking at HPC with a 100/MWh CfD, and an average 50/MWh sell price, they will on average receive 50/MWh in subsidy. Sometimes they will get much more, sometimes much less, and sometimes they will pay some back (when the price is higher than 100), but on average they will receive approx 50/MWh.

Under this scheme/idea, let's say a wind farm gets a CfD for 35/MWh. Then with prices averaging around 50, they will get no subsidy. In reality, sometimes of the day/year, they will get less than 35 and receive a top up, but then, when they sell for more than 35, they will have to give the extra back to the subsidy pot until funds received have been repaid.

Basically, we are talking about a zero subsidy scheme, but one that offers assurance for the cheaper forms of RE that are currently struggling as they are just about there, but need their hands held.

Pages: [1] 2 3 4 5 ... 10
Powered by MySQL Powered by PHP Powered by SMF 1.1.20 | SMF © 2013, Simple Machines
SMFAds for Free Forums
Simple Audio Video Embedder
Valid XHTML 1.0! Valid CSS!