Tuesday, June 09, 2020
Why "renewables" aren't the solution to our climate problem
One of the glaring flaws (far more than a mere foible) of "renewables" (wind and PV) is that they are unreliable. SO unreliable, as a matter of fact, that they force the adoption of much dirtier fossil-fired generators to accommodate their output swings.Naive greenies think that "RE" can just be thrown onto the grid, but in fact an RE-heavy grid requires different generating technologies than one with little or none. You can generally follow the normal load curve using a CCGT plant, which can be up to 64% efficient (LHV). Following the bumpiness of "renewables" mostly requires simple-cycle gas turbines (the CCGT steam systems don't like rapid power variations); the best open-cycle I've read about gets only 46% efficiency, and I recall that the single-shaft industrial models often get something like 38%. IOW, you're burning a lot more fuel for the same electric output. This puts you way behind emissions-wise.
Let's use a real-world example: the Mitsubishi-Hitachi M501JAC gas turbine, which is available in both simple-cycle and combined-cycle versions. This allows a head-to-head comparison. The single-unit combined-cycle version of the M501JAC is rated at 614 MW and 64.0% LHV efficiency. It doesn't even HAVE a specified turndown ratio, minimum rated output, rated ramp rate or startup time. One can conclude from this that it really isn't suitable for trying to follow the ups and downs of "renewables", though it can probably handle normal load curves because other steam-turbine plants have been doing it for the last century.
The simple-cycle heat rate of this unit is 7775 kJ/kWh (LHV). Since a kilowatt-hour is 3600 kJ, we just divide that by 7775 to get 0.463, or 46.3%. The rated output is 425 MW and the rated ramp rate is 42 MW/minute, or about 10% per minute; it can be turned down to 50% of rated output, so it can go from minimum to full output in 5 minutes. This can track things like surges and sags from passing clouds and weather fronts pretty well. Its startup time is specified as 30 minutes.
What you pay for this flexibility is efficiency. Going from 64.0% down to 46.3% means burning 38% more fuel. Put another way, you need to get 27.6% of your juice from emissions-free sources just to break even on the increased emissions from going from combined-cycle to simple-cycle... and that assumes that you maintain the 46.3% efficiency at lower output power, which you won't. GE makes this data very hard to find, but the efficiency of the LMS100 gas turbine drops from 44.3% at rated power down to under 40% at half rated power (the minimum). This means even MORE fuel required.
Typical capacity factors for wind are 30-40%; PV is much lower. If you're getting 30% of your juice from "renewables", and you're burning at least 38% more fuel per kWh to get the rest, you're saving less than 3.3% from the CCGT emissions figure. At low enough capacity factors, you can actually burn more fuel with the addition of "renewables" than what you could do with all-fossil.
Is it worth spending so much money for such paltry gains? Even if your wallet can stand it, can the planet?
Now, don't let it be said that there aren't ways around this. With enough excess RE capacity you can just brute-force the issue by dumping excess power to resistance heaters in a CCGT's gas turbines, substituting electricity for fossil fuel and managing the rapid power swings on the demand side. But this is going to hit the economics, and nobody even seems to be thinking that far out of the box.
Labels: renewable energy
¶ 6/09/2020 07:01:00 PM
Comments:
<< Home
Thanks for putting some numbers on this, they are hard to find.
Surprised that the part load efficiency is still 40%, one would expect it to drop like a stone below 70 or 80% - being a turbine one would expect a typical turbine efficiency curve drop off below the best efficiency point.
Here in Holland they are thinking about a large increase in solar PV. This is not a sunny country, we average 10% capacity factor for PV systems. Rooftop systems can be even worse, with typical values around the 8% mark for west or east facing systems, often at a bad azimuth angle too. There's even a guy nearby that has a steep north facing roof and put solar panels on there. Probably gets 5% or less capacity factor.
How bad would the extra fuel burn be in these cases? As per your numbers it is likely that there will be a substantial increase in emissions for every PV system installed. That's a sledgehammer to the head for the solar PV case.
Surprised that the part load efficiency is still 40%, one would expect it to drop like a stone below 70 or 80% - being a turbine one would expect a typical turbine efficiency curve drop off below the best efficiency point.
Here in Holland they are thinking about a large increase in solar PV. This is not a sunny country, we average 10% capacity factor for PV systems. Rooftop systems can be even worse, with typical values around the 8% mark for west or east facing systems, often at a bad azimuth angle too. There's even a guy nearby that has a steep north facing roof and put solar panels on there. Probably gets 5% or less capacity factor.
How bad would the extra fuel burn be in these cases? As per your numbers it is likely that there will be a substantial increase in emissions for every PV system installed. That's a sledgehammer to the head for the solar PV case.
Regarding the use of LHV figures - don't CCGTs condense the combustion exhaust moisture H2O by heat exchange to incoming Brayton cycle air or condensate return (economizer)? One would think that is an obvious improvement in efficiency especially for natural gas firing. Then we'd be looking at HHV values.
Even for a single cycle Brayton, condensing with the incoming air should still be possible right? If not then that will further aggravate the PV fuel efficiency situation, if CCGTs can economize and single cycles can't.
Even for a single cycle Brayton, condensing with the incoming air should still be possible right? If not then that will further aggravate the PV fuel efficiency situation, if CCGTs can economize and single cycles can't.
"Regarding the use of LHV figures - don't CCGTs condense the combustion exhaust moisture H2O by heat exchange to incoming Brayton cycle air or condensate return (economizer)?"
No, CCGTs do not condense combustion moisture. That would cause condensation of water and acids and corrosion in the exhaust stack. The custom with steam cycles is to tap steam to heat feedwater, and my understanding is that's what's done below about 150°C and also for the balance above that. You can find data on steam-cycle plants to see fairly precise mass/heat flows and work out the details yourself.
No, CCGTs do not condense combustion moisture. That would cause condensation of water and acids and corrosion in the exhaust stack. The custom with steam cycles is to tap steam to heat feedwater, and my understanding is that's what's done below about 150°C and also for the balance above that. You can find data on steam-cycle plants to see fairly precise mass/heat flows and work out the details yourself.
Corrosion can't be the reason it's not done - condensing boilers are standard here. Obviously they do not use carbon steel for the exhaust, typically stainless or aluminized. Exhaust is just a bunch of sheet metal really. Exhaust is just mild acid really, nothing special that would require superalloy or anything. Compared to the conditions at the leading blades of the gas turbine it looks like a pleasant bathtub. (though there have been a few cases where cost was saved on plumbing and the exhaust condensate from the heating boiler was dumped on roof drains, causing corrosion of the gutters).
I suspect the reason it isn't done is just economics. Gas is cheap so adding the extra HX and paying a few dollars more for stainless or aluminizing would be considered a bad thing for the bean counters that want to shave off every buck on a project. Same with a coal plant, and there's tons more sulphur and other nasties in there that might require even more expensive alloys.
I suspect the reason it isn't done is just economics. Gas is cheap so adding the extra HX and paying a few dollars more for stainless or aluminizing would be considered a bad thing for the bean counters that want to shave off every buck on a project. Same with a coal plant, and there's tons more sulphur and other nasties in there that might require even more expensive alloys.
"Corrosion can't be the reason it's not done - condensing boilers are standard here."
Well, there's the corrosion issue and then there's plumbing cost. A steam plant already has a bunch of feedwater heaters which tap steam off the turbines and use it to pre-heat the boiler water; that water can be VERY hot before it gets to the boiler, it just won't be vapor. Recovering low-temperature heat from the exhaust gas would require a plumbing run to the boiler and back, and displace low-pressure steam into the feedwater heaters. This would increase the mass flow into the condenser and the waste heat there, making the effort somewhat of a wash. A steam plant usually has an "economizer" also, but I've not seen the specs for heat recovery from that. Most of the turbine heat input comes from the boiler and reheaters.
Then there's also the issue of the exhaust plume. You want it to be buoyant, so it doesn't come back to ground near the plant. Nor do you want fallout to do that either. Stuff starts to condense at about 150°C, which is probably why that's the temperature at which LHV is measured.
Well, there's the corrosion issue and then there's plumbing cost. A steam plant already has a bunch of feedwater heaters which tap steam off the turbines and use it to pre-heat the boiler water; that water can be VERY hot before it gets to the boiler, it just won't be vapor. Recovering low-temperature heat from the exhaust gas would require a plumbing run to the boiler and back, and displace low-pressure steam into the feedwater heaters. This would increase the mass flow into the condenser and the waste heat there, making the effort somewhat of a wash. A steam plant usually has an "economizer" also, but I've not seen the specs for heat recovery from that. Most of the turbine heat input comes from the boiler and reheaters.
Then there's also the issue of the exhaust plume. You want it to be buoyant, so it doesn't come back to ground near the plant. Nor do you want fallout to do that either. Stuff starts to condense at about 150°C, which is probably why that's the temperature at which LHV is measured.
Obviously, the cycle regeneration steam bleed rate would be reduced commensurate with the added heat recovery from the economizer (which is what this is if applied to Rankine). There would be no increase in mass flow rate to the condenser, just less steam bleed. Should be fine as long as no excessive moisture forms in the turbine - certainly no problem for a modern superheat cycle with reheat, which almost has the opposite problem (too dry a steam reject to the condenser which is slightly wasteful).
Air preheat for the Brayton is still on the table if a steam cycle economizer is inefficient. I guess same argument here - too much plumbing, HX cost, when fuel is cheap so why bother... weird aspect of our government taxing labor so working to improve efficiency is not economically viable. Better to be inefficient and not work, because working is bad. Our government says so... by taxing us the harder we work. what kind of message is that?
As to the exhaust plume - would be interesting to know what kind of mechanical, fan power would be needed for exhaust venting. Can't imagine it will be a lot, what with the enormous chimney height for natural draft. If every 1 kW of fan power to exhaust gets you another 10 kW of power output on the grid because of cycle efficiency improvement then obviously looking pretty attractive.
Air preheat for the Brayton is still on the table if a steam cycle economizer is inefficient. I guess same argument here - too much plumbing, HX cost, when fuel is cheap so why bother... weird aspect of our government taxing labor so working to improve efficiency is not economically viable. Better to be inefficient and not work, because working is bad. Our government says so... by taxing us the harder we work. what kind of message is that?
As to the exhaust plume - would be interesting to know what kind of mechanical, fan power would be needed for exhaust venting. Can't imagine it will be a lot, what with the enormous chimney height for natural draft. If every 1 kW of fan power to exhaust gets you another 10 kW of power output on the grid because of cycle efficiency improvement then obviously looking pretty attractive.
I posted this on BNC which bears out your point I think:
A simple case in point.
My own country of Holland. We have a need for about 120 billion kWh/year, over 13 GWe average draw.
We can probably accomodate some 10 GWp of solar PV with some serious investments in grid upgrades. We have a lot of gas turbines which can accomodate big swings though some plants are must-run, cogen and steel related so that's a problem but let's ignore that to be kind to PV.
We get about 10% capacity factor fo PV. The best systems, ideally angled towards the south, and regularly cleaned get about 11%, but many systems are east or west facing and not so often cleaned so would get 7-9% capacity factor. Let's use 10% as a typical fleet average.
Our 10 GWp of solar PV only generates 1 GWe average on the year. So it provides under 1/13th the electrical energy or some 7% of total electrical demand.
The problem, as you might have noticed, is that we haven't addressed 93% of the problem, which is going to have to be mostly single cycle gas turbines throttled very inefficiently. Almost all coal stations will have to be shut down, which is a good thing on the environmental side, though not much help on the CO2 emissions side, since throttled peaker gas turbines emit almost as much CO2 per kWh as a coal plant.
You will have noticed that the vast majority of this "solar" grid is powered by natural gas, used very inefficiently in single cycle gas turbines.
Now, as a matter of fact, the CO2 emissions would be less if we powered most of the demand using efficient dual cycle gas-steam turbines for mostly baseload and load following service. Ironically, this would have a LOWER CO2 emission than the solar-gas grid.
The main point being that low capacity factor, intermittent, non-dispatchable solar energy sources only serve to lock us into using fossil fuels (very inefficiently) for as long as the solar energy sources persist. Which, I'm told, is forever since it is renewable.
Ouch. Reality hurts!
Post a Comment
A simple case in point.
My own country of Holland. We have a need for about 120 billion kWh/year, over 13 GWe average draw.
We can probably accomodate some 10 GWp of solar PV with some serious investments in grid upgrades. We have a lot of gas turbines which can accomodate big swings though some plants are must-run, cogen and steel related so that's a problem but let's ignore that to be kind to PV.
We get about 10% capacity factor fo PV. The best systems, ideally angled towards the south, and regularly cleaned get about 11%, but many systems are east or west facing and not so often cleaned so would get 7-9% capacity factor. Let's use 10% as a typical fleet average.
Our 10 GWp of solar PV only generates 1 GWe average on the year. So it provides under 1/13th the electrical energy or some 7% of total electrical demand.
The problem, as you might have noticed, is that we haven't addressed 93% of the problem, which is going to have to be mostly single cycle gas turbines throttled very inefficiently. Almost all coal stations will have to be shut down, which is a good thing on the environmental side, though not much help on the CO2 emissions side, since throttled peaker gas turbines emit almost as much CO2 per kWh as a coal plant.
You will have noticed that the vast majority of this "solar" grid is powered by natural gas, used very inefficiently in single cycle gas turbines.
Now, as a matter of fact, the CO2 emissions would be less if we powered most of the demand using efficient dual cycle gas-steam turbines for mostly baseload and load following service. Ironically, this would have a LOWER CO2 emission than the solar-gas grid.
The main point being that low capacity factor, intermittent, non-dispatchable solar energy sources only serve to lock us into using fossil fuels (very inefficiently) for as long as the solar energy sources persist. Which, I'm told, is forever since it is renewable.
Ouch. Reality hurts!
<< Home
