Wikipedia:Reference desk/Archives/Science/2019 September 24

Welcome to the Wikipedia Science Reference Desk Archives
The page you are currently viewing is a transcluded archive page. While you can leave answers for any questions shown below, please ask new questions on one of the current reference desk pages.

September 24

LED lights in series

can I use two LED lights, each 110V 500W 60Hz, in series, with a 220V 50Hz power supply? thanks! --49.203.218.178 (talk) 17:39, 24 September 2019 (UTC)

Are these powerful searchlights? 500w is big for LEDs. Dbfirs 21:03, 24 September 2019 (UTC)

Yes, the pair would be equivalent to like 6500 watts of incandescent lights, or 65 hundred watt incandescent bulbs. SinisterLefty (talk) 02:18, 25 September 2019 (UTC)

The first component in a series circuit experiences the full input voltage, so if you did this, the first light would get 220V input and go kablooey. LEDs are semiconductor devices that require direct current. The rating of "110V 500W 60Hz" is the rated input for the power supply in the lighting unit that takes AC input and produces the DC input for the actual LEDs. --47.146.63.87 (talk) 21:58, 24 September 2019 (UTC)

Okay yeah, I misinterpreted that. To the poster, don't do it anyway. You would need to be wiring the two light fixtures together in series, and you shouldn't be doing mains wiring unless you're an electrician. --47.146.63.87 (talk) 03:00, 25 September 2019 (UTC)

If these were 500w incandescent floodlights, then putting them in series would be fine because, for a purely resitive load, each item in a series circuit with matching impedance gets exactly half of the voltage. I don't know what 47.146.63.87 was thinking above, but I agree that if the lights contain electronics then putting them in series would not be adviseable and might give unpredictable results. Dbfirs 22:10, 24 September 2019 (UTC)

Voltage will be OK (220 /2 = 110V ). The question is, how the electronics used to turn 60Hz AC into DC will react when fed with 50Hz AC instead? Depending on the design, it could work fine, just don't work without other consequences, or be destroyed. Could be mentioned in the user manual, if you have it

BTW 500 W LED devices do exist https://duckduckgo.com/?q=500W+led&ia=web Gem fr (talk) 00:03, 25 September 2019 (UTC)

They do, but they are extremely bright -- mainly for industrial use. That's why I wondered if the OP's units were halogen. Dbfirs 07:53, 25 September 2019 (UTC)

"Voltage will be OK (220 /2 = 110V )" is only true for resistive loads. Switching LED power supplies are not resistive loads and will react unpredictably. Many of them are open loads during part of the cycle and heavy loads at other parts of the cycle -- and it isn't a sure thing that it will be the same part of the cycle or even that the switching power supplies won't go crazy as they feed back into each other. Maybe it will work. Maybe no current will flow. Maybe the LEDs will catch on fire one at a time. --Guy Macon (talk) 07:57, 25 September 2019 (UTC)

• 500W LED? Check the label. For that power, they're probably fairly sophisticated electronics and may simply run from either 110V or 230V automatically. Andy Dingley (talk) 17:56, 25 September 2019 (UTC)

• Thanks everyone! --117.219.217.63 (talk) 10:26, 26 September 2019 (UTC)

Lead-acid batteries for grid storage

Tesla has started delivering some of their batteries for grid storage and stabilisation. As I understand it, these are Li-Ion-Batteries. I can understand why these are chosen for electric cars, and for small, expensive portable devices like laptops and mobile phones. But is there a good reason for them to be used for stationary applications? Lead-acid batteries are mature and cheap. The can be configured to withstand a large number of cycles, and they can deliver large currents. And they can be recycled nearly perfectly if eventually worn out. Moreover, they are not flammable. Is there a hidden technical reason not to use them? Yes, they need more volume and are heavier, but that should be a very minor consideration for a stationary system. I can understand Tesla going for economy of scale, and pushing their very good Li-Ion-technology into other applications, of course. --Stephan Schulz (talk) 17:45, 24 September 2019 (UTC)

The following reference might be of interest:

• May, Geoffrey J.; Davidson, Alistair; Monahov, Boris (1 February 2018). "Lead batteries for utility energy storage: A review". Journal of Energy Storage. 15: 145–157. doi:10.1016/j.est.2017.11.008. ISSN 2352-152X.

2606:A000:1126:28D:D5F1:51F:BD4D:80FD (talk) 19:15, 24 September 2019 (UTC)

There may be some hesitancy to implement lead-acid batters due to the substantial greenhouse gas emissions associated with their manufacture [1]. I'd also note that the linked article above, while favorably cited in the literature, was literally authored by consultants for the lead industry. Probably a fair assessment of technical superiority, but not expected to discuss downsides. Someguy1221 (talk) 20:35, 24 September 2019 (UTC)

Thanks to both of you. I knew about the first paper, which somewhat triggered my question, but the second one is new to me and very useful. --Stephan Schulz (talk) 13:21, 25 September 2019 (UTC)

There is no good reason except the one you mentioned Tesla going for economy of scale, and pushing their very good Li-Ion-technology into other applications, of course., combined with marketing reason (technically illiterate people, buying fashionable stuff). Gem fr (talk) 00:15, 25 September 2019 (UTC)

The lead would be problematic if it ended up in a landfill, but we already have systems in place for recycling car batteries, at least in the US. Dealing with the acid is also hazardous, but proper safety equipment and precautions can reduce the risk. But for a less environmentally risky method, I like pumping water from a low reservoir to a high one. SinisterLefty (talk) 02:22, 25 September 2019 (UTC)

As I understand it, car batteries have a 99%+ recycling rate in the US (the lead is valuable), so for large scale installations I would expect near perfect recycling. The electrolyte is 30-50% sulfuric acid, which is corrosive, but is already in use in a bit less than 300 million vehicle in the us, or 45 million in Germany without frequent problems, so the risk seems to be very manageable. --Stephan Schulz (talk) 13:21, 25 September 2019 (UTC)

Pumped-storage hydroelectricity. If we ever get room-temperature superconductors that are suitable as cheap powerlines, we might be able to stop caring about the fact that some locations are unsuitable for some storage and generation technologies. We could just locate the power plants and batteries wherever they will be cheapest to operate. Maybe vast pumped storage up in mountains that no one lives in, or swaths of desert covered in massive Solar updraft towers. Someguy1221 (talk) 04:40, 25 September 2019 (UTC)

You always care about location, if only for security reasons. Powerlines are cheap and efficient enough already. Of course if we could have superconductors we would profit, but this would not be a revolution. Anyway electricity storage is just stupid nonsense: just realize that a storage device is 1)a production device 2)that actually do not produce, you need another for that 3)with some regenerative feature (you have to pay for this). An actual production device, that you need anyway, is better. Gem fr (talk) 11:15, 25 September 2019 (UTC)

When you're talking about the possibility of distributing electricity over a thousand miles from the generator, the power line losses become substantial, but you're right that it's possible to do today. Regarding storage, what? Storage can easily be cost efficient over generation when dealing with peak demand that is far beyond average demand. Otherwise you would need a vast excess of generation capacity that will normally not be used. Someguy1221 (talk) 20:00, 25 September 2019 (UTC)

depends on your idea of "substantial". Nowadays high-voltage direct current transmission losses are quoted as less than 3% per 1,000 km, which is pretty good. Now of course, since these can transport 1000 MW or more, 3% of 1000 MW mean 30 MW, worth thousands of $ lost each hour, which as the owner of such thing (if I were one), I would indeed find it substantial... but not enough to give up the even bigger money from my operation. You obviously failed to notice that a storage device IS "generation capacity that will normally not be used"... except it doesn't even produce of its own! So your second point self-destruct. Gem fr (talk) 20:38, 25 September 2019 (UTC)

you cannot do Pumped-storage at home. rising 1m3 of water 1m only store less than 10,000 J, that is, less than 3 Wh; a finger-sized, dollar worth, accumulator do the same. a device the size of house (100 m² x 3m heigh= 300m3, 6m elevation between basement and cellar) would store 5 kWh, less than a dollar worth of electric energy. And I just cannot imagine the price...Gem fr (talk) 10:55, 25 September 2019 (UTC)

"Never spend a dollar accomplishing something when you can spend $50,000 dollars doing the same thing". --The US Federal Government^{[Citation Needed]}

--Guy Macon (talk) 13:02, 25 September 2019 (UTC)

The original Q was about "grid storage and stabilisation", not home use. And to keep the cost down, natural reservoirs should be used. One option that might work is generating electricity as water is poured down into an abandoned mine, then pumping it back up to the surface when excess grid electricity is available. Of course, you'd need to be careful that there wasn't something like arsenic in the mine that would contaminate the water. SinisterLefty (talk) 17:14, 25 September 2019 (UTC)

Pumped storage is widely used, but the number of suitable locations is quite limited - and often competes with other uses like tourism, at least in Europe (which, as a rule, is pretty full of everything). Batteries are comparatively easy to create and install. IIRC, a car battery (not optimised for grid storage, of course) stores about 1kWh, so a million of them would store 1TWh - enough to replace one100 nuclear power plants for a night. And since we have 45 million batteries in cars (which live about 10-15 years), we seem to produce several million of those per year already. --Stephan Schulz (talk) 17:49, 25 September 2019 (UTC)

check your math. A nuclear plant is ~1.5TW, 12TWh in a 8h night, 12 millions 1kWh battery energy-wise. Gem fr (talk) 21:07, 25 September 2019 (UTC)

Well, the commercial nuclear reactors I know are all in the general region of one GW (check e.g. Biblis Nuclear Power Plant, 2.5GW nominal with two reactors and 70% capacity factor), not one TW. There is that factor of 1500 between your estimate and mine. The math seems to be good enough (modulo rounding) on both sides, but your power plants seem to come from Iron Sky. Or I'm totally confused - it does happen ;-). --Stephan Schulz (talk) 07:32, 28 September 2019 (UTC)

N.B.: If you have a large multi-reactor powerplant, you get on the order of one TWh/year - maybe that is the source of confusion? And yes, I probably should have written "100 typical commercial nuclear reactors", since power plants often have more than one reactor. --Stephan Schulz (talk) 07:44, 28 September 2019 (UTC)

My bad, you are right, I jumped from M to T over G and should had taken my own advice to check my math, yours was good. Gem fr (talk) 13:06, 28 September 2019 (UTC)

No worries - it happens to all of us. And it's refreshing to meet someone who can actually admit to a mistake! ;-). --Stephan Schulz (talk) 16:22, 28 September 2019 (UTC)

There has to be a lot more than 45 million car batteries in the world. SinisterLefty (talk) 19:47, 25 September 2019 (UTC)

Yes, sorry. The 45 million is the number in Germany - I should have been clearer. --Stephan Schulz (talk) 20:11, 25 September 2019 (UTC)

Some recent hybrid cars, such as the Mitsubishi Outlander, can already also be used as power banks by V2H. Andy Dingley (talk) 17:55, 25 September 2019 (UTC)

Just do the math. Why people don't do the math, FGS? What is the cost of the battery (~$100), how many cycle you can expect(~350)->cost of a cycle($0.3). You'll find that it cost 5-10x more just to store a kWh in a common, cheap, lead battery than to produce it in a nuclear/gas/coal/whatever power plant (~$0.03-0.05/kWh)Gem fr (talk) 19:58, 25 September 2019 (UTC)

Have you done the math on the capital expenditure required to have production capacity that can exceed a peak demand that might be 150% higher than average demand, but for less than 1% of the total hours in a year? What about excess production capacity to deal with load shifts during both planned and unplanned plant downtime and power line maintenance? Someguy1221 (talk) 20:07, 25 September 2019 (UTC)

I did not, because this is entirely another point: about power management, not energy management. Now, if you switch the point to power instead of energy, then, tapping on cars engines idling at home would be more sensible that just tapping on just their batteries. Gem fr (talk) 21:13, 25 September 2019 (UTC)

350 cycles for a lead-acid car battery seems rather low. Of course, they aren't normally fully discharged, so the question would be what's the maximum number of kWh you can get out of one, using the ideal charge/discharge cycle. SinisterLefty (talk) 20:22, 25 September 2019 (UTC)

(ec)A normal brand name car battery is more like $50 - retail price. And stationary lead-acid batteries, with a reasonable battery management system, are good for a lot more than 350 cycles. On the other hand, looking at Cost of electricity by source, the overall cost of a kWh from nuclear seems to be more on the order of 10ct/kWh. And neither prices for nuclear or fossil sources include environmental risk and damage. --Stephan Schulz (talk) 20:23, 25 September 2019 (UTC)

Yes, and the retail price reflects markups by middle-men. If you bought a million of them directly from the factory, presumably they would cost far less, due in part to economy of scale. SinisterLefty (talk) 20:25, 25 September 2019 (UTC)

then again, just do the math. With more relevant figures if you find mine, that is, Lead–acid_battery figures, irrelevant/false for some reason. Won't change the result: storage is expensive because, again, is is just generation power with added feature (regeneration) you have to pay for, and lacking actual production feature (with no saving on this one, acutally requiring you spen on something else) Gem fr (talk) 20:57, 25 September 2019 (UTC)

Many forms of electricity generation produce too much at times and too little at others. Solar and wind, certainly, but also nuclear, as they can't quickly change the reaction rate. So, on the "too much" side, you end up just wasting that electricity, unless it can be stored and reused when needed. Therefore, it doesn't make sense to add the price of storage to the price of that generation, as the cost of generation is a sunk cost at that time. Instead, the cost of storage should be compared with the cost of generating that electricity a second time. You might further compare with rainfall, which also tends to come in too much and too little cycles. When there's too much, instead of letting it all drain down the river, it makes sense to store it for when there's too little. Hence reservoirs, and, on a small scale, rain barrels. SinisterLefty (talk) 21:35, 25 September 2019 (UTC)

There is a colossal flaw in your reasoning, though. And it really shouldn't surprise, because it's unlikely that engineers designing power grids in the US, Europe and China all made the same mistake. That flaw is assuming excess unused capacity is equivalent between generators and storage devices. It is not. A power plant does not magically become cheaper to operate when the demand suddenly drops. It does not become magically more expensive to operate when the demand suddenly rises. The fuel does not change the rate at which it is consumed to match instantaneous demand. Changing the operational state of a power plant in a cost-effective way can take hours or more. For a power plant to have 2 or 3 times more capacity than it needs to meet unexpected spikes in demand, it has to operate at that level at all times. The cost of fuel is a minority of the operating costs of the plant. So your electricity is now 2 to 3 times more expensive to generate (In fact, power plants generally waste electricity anyway because operating at higher-than-necessary capacity is the norm, especially for plant designs that can't ramp up/down at the speed of ordinary daily cycles anyway). Do you want to be able to handle the possibility of ordinary demand spikes and part of the grid going down? Have fun paying for all those continuously operating plants. One of the design features of a power storage device is that it can store energy sitting still, and release it on demand. It does not need to be continuously in use. It's like saying it's cheaper to take an uber than to buy a car, in terms of marginal costs, but you pay the uber driver >$200 a day to follow you everywhere you go in case you need him. Someguy1221 (talk) 01:35, 26 September 2019 (UTC)

I generally agree, although some types of on-demand power plants do exist, such as coal-fired. Despite being dirty (they actually put far more radiation into the air than nuclear plants, on average) and coal being not particularly cheap any more, it is fairly easy to ramp up and down the electricity production at one to match demand. Natural gas is a somewhat cleaner alternative (but still puts out greenhouse gases). Hydro can be used for this too, but there the environmental damage is in building the dam and flooding the area behind it (with some exceptions where the geography gives you a nice height difference naturally, like at Niagara Falls). Having one of these plants along with the base generation capacity of nuclear and maybe some wind and solar works well. But, if we go with all renewable energy, then some storage mechanism is the best alternative to dealing with a mismatch between supply and demand. Much better than massive overproduction from, say, windmills, to cover spikes. Also, methods of blunting those spikes help, like encouraging chilled water systems that run at night to provide cooling during the day with low electricity usage then. SinisterLefty (talk) 01:50, 26 September 2019 (UTC)

Coal-fired plants are quite slow to adjust - you need to feed in more coal, and that takes a while to actually start burning. Oil-based plants are faster (one reason why HMS Queen Elisabeth and her sisters changed from coal to oil), and gas-plants are faster still. Pumped hydro is very fast, but batteries beat all the generating plants currently in use.--Stephan Schulz (talk) 17:08, 26 September 2019 (UTC)

What's the power source for charging those batteries? ←Baseball Bugs ^{What's up, Doc?} carrots→ 18:40, 26 September 2019 (UTC)

The batteries are usually charged by the power plants at times when the demand is less than the supply (at night on a warm day power is usually super cheap). It can be cheaper than turning the power plants down or off and then back on when demand returns. Someguy1221 (talk) 02:50, 27 September 2019 (UTC)

Zero math in this. Gem fr (talk) 19:02, 26 September 2019 (UTC)

For math follow the links and read the refs. No point in re-inventing the wheel here. SinisterLefty (talk) 19:26, 26 September 2019 (UTC)

Right, I was wrong to associate the question with homes. Doesn't change the fact that storage is stupid, though. Gem fr (talk) 19:42, 25 September 2019 (UTC)

Unsourced and incorrect. Britain has one large storage system - Dinorwig Power Station which is used pricipally to cover short-term changes in supply, often from changes i solar power - see http://www.gridwatch.templar.co.uk/ for examples of this. I don't believe those who built it and run it regard it as stupid. Sir David John Cameron MacKay FRS FInstP FICE, Regius Professor of Engineering in the Department of Engineering at the University of Cambridge and from 2009 to 2014 Chief Scientific Adviser to the UK Department of Energy and Climate Change says in his book Sustainable Energy - Without the Hot Air (amongst much other discussion of the need for storage once we need to rely on renewables completely) "We need to solve two problems - lulls (long periods with small renewable production), and slews (short-term changes in either supply or demand). We've quantified these problems, assuming that Britain had roughly 33 GW of wind power. To cope with lulls, we must effectively store up roughly 1200GWh of energy (20kWh per person). The slew rate we must cope with is 6.5GW per hour (or 0.1 kW per hour per person)." He then goes on to explain how stored power is vital to do this. You might like to read his book.

That a thing would be stupid never prevented it being done. All the more so, that even if a thing is generally stupid, it may make sense in special case. Besides, experimenting is not stupid, so you may have one. That said:

yes, Britain has ONE. Conceived in 1974, commissioned in 1984. for the PURPOSE of managing short-terme change in POWER supply (not energy management: can you understand the difference?), hence the very telling ratio of power (1.8 GW) to energy (9 GWh); the thing can run ~5 hours. It is part of the National Grid Reserve Service to manage POWER, not ENERGY. Changing the purpose of this thing into a energy storage device... well, just do the math with the figure you provided yourself: 1200GWh needed/9 GWh this thing has= 133 similar station (good luck finding 132 place to build them, and convincing people...), except it would be ridiculous for them to be 1.8GW (adding up to 240 GW) just to manage 33 GW not working wind power. Point is : if this thing was an energy storage device, it would be ~7x less powerful, ~0.3GW... and actually close to useless to cope with the real issue, that is, power issue. This illustrate the silliness of the storage scheme.

Now explain what this power station does, that a same power gas-powered powerplant won't? hard, isn't it?

Gem fr (talk) 18:56, 26 September 2019 (UTC)

• Britain has four pumped-storage stations. What was (AFAIK) the world's first (1890) was in England, but was destroyed by a flood in 1952.

Ignoring Ffestiniog (it's old) your point about Dinorwig's high power is a good one, because Dinorwig was built to even out load on the nearby nukes. The more recent Scottish pumped hydros, and even more so for the others planned, are about energy storage, rather than power. They have comparable storage capacity to Dinorwig, but a quarter of the power. These are more about infilling gaps in local wind than boiling kettles during Coronation Street. Andy Dingley (talk) 20:04, 26 September 2019 (UTC)

I stand corrected on the number. four it is, then. Obviously, the more stupid, uncontrolled power source, like wind, you connect to the grid, the more balancing devices you need. I could even support wind power if it were a supporting device of hydro power (that is, used to replenish the hydro energy supply). But using the best (hydro) as a way to support the most idiotic (wind or high latitude solar) just blows my mind. Gem fr (talk) 10:53, 27 September 2019 (UTC)

• But hydro is far from "the best" internationally. It's rarely used in the UK because we have so few locations for it. Where we do have it (Scotland), it's a long way from consumer demand. You can't "install" hydro somewhere where geography doesn't support it. Andy Dingley (talk) 13:10, 27 September 2019 (UTC)

Suitable places are rare indeed, alas. The fact we are looking for these, and are not finding enough, is tribute to its value. Damn Scots, why don't they let us flood their country ;-) ? Gem fr (talk) 15:39, 27 September 2019 (UTC)

"Now explain what this power station does, that a same power gas-powered powerplant won't?". Read the Wikipedia article. It gives the following: 1. Spool up from 0 MW to 1800 MW load can be achieved in approximately 16 seconds. 2. Another important feature of Dinorwig is that it has been designed to assist restarting the National Grid on the occasion of a complete power failure (a black start). 3. It fills an important need in the system by responding to sudden surges in electricity demand because of its rapid ability to deliver power on load spikes. And in your desire to be insulting, you've missed a main point: that one of the most respected scientists in the world states that large quantity energy storage will be necessary in a renewables-only economy. So I was addressing both power management and energy storage, that inexplicably you seem to have missed. I suggest you read Professor McKay's book and do some learning rather than make unsourced statements.--Phil Holmes (talk) 08:36, 27 September 2019 (UTC)

Only your 1. is good, although any hydro do that (no question hydro is the best in so many ways; real hydro, that is: dams that don't need some other plant to actually work), and gas-powered are not that bad (matters of minutes, unless I am wrong) in delivering power from scratch. Your points 2. 3. are just irrelevant: gas-powered do just the same. "large quantity energy storage will be necessary in a renewables-only economy" indeed; which is not argument supporting the former, but rather an argument against the latter, mind you. Gem fr (talk) 10:41, 27 September 2019 (UTC)

Please quote some reliable sources for that opinion statement.--Phil Holmes (talk) 16:31, 27 September 2019 (UTC)

No opinion statements here, just facts you'll easily find (Peaking_power_plant#Types, [2] etc.). Not sure you actually care... Gem fr (talk) 17:42, 27 September 2019 (UTC)

• But what do you mean by "gas-powered"? There are three common types which could be described thus, two are fairly quick to bring up to speed but are inefficient, the one that isn't inefficient isn't quick to start either. CCGT now provides half of UK generation, because even though we've abandoned coal we're still burning fossil gas like crazy. Andy Dingley (talk) 17:23, 28 September 2019 (UTC)

The lede of Base load has a nice summary of key terms and links to our various articles relating to the ability of a powerplant to ramp up or down vs variations in load. DMacks (talk) 02:50, 26 September 2019 (UTC)

Not sure if it's mentioned in any of the above sources, but our article Grid energy storage mentions a lead-acid storage plant in Texas, and NiCd one in Alaska. Most of the new ones do seem to be lithium (from checking the sources) although [3] does mention a planned zinc one in Indonesia. As for lithium, an interesting option is mentioned in our article of a plan to use old electric car batteries in Lünen. If recycling lithium batteries is more limited, I guess this could be an attractive proposition as electric car use increases assuming it works well. I suspect even the best recycling is likely to cost more than reuse anyway so it depends whether the other costs from reuse compared to new batteries designed and intended for the purpose (losses, greater management and risks, lower capacity, more frequent replacements etc) exceeds the recycling cost. And of course with any waste or emissions deciding how you want to price these can get complicated. You could try reusing lead acid car batteries but there are plenty of possible reasons this may not work even you can do it with other batteries like lithium ones. Nil Einne (talk) 08:34, 26 September 2019 (UTC)

Yes, repurposing lithium car batteries makes more sense than buying them new. There are likely to be a lot of them available that will only hold charge at less than 90% of the original capacity, and thus they are replaced in cars, since electric cars already have a rather limited range, so owners sure don't want it reduced further. On the other hand, for grid storage, they can just add as many as they like to increase capacity. And if lithium batteries are not reused, I'm not sure how likely they are to be recycled. I do hope they take proper anti-fire precautions, though, as old lithium batteries occasionally catch fire. SinisterLefty (talk) 11:23, 26 September 2019 (UTC)