Pragmatic 2D: The Battery Problem Follow-up 4

3 February, 2014

CURRENT

Follow up (Part D) to The Battery Problem where John addresses some more follow up about Flywheels as a form of energy storage, the challenge of disposing of or recycling dead solar panels in 40 years and realistic expectations of DC efficiency.

Transcript available
This is Pragmatic Follow-Up Part D for Episode 2, The Battery Problem. I'm Ben Alexander and my co-host is John Chidjie. Well, we've had some more feedback regarding elements of the battery problem that we talked about. One of the alternative energy storage mechanisms out there, something I didn't talk about was flywheels. So this particular suggestion came from Matt T via email. And Essentially, he just wanted me to go over flywheels and discuss some of the pros and cons. So, essentially, if you've ever done a flywheel, a flywheel is essentially a spinning disc. So, think of it like a donut shaped, usually the modern ones are a donut shape. The old ones used to be a solid disc that spin on an axis like a spinning top. They spin at extremely high speeds, though. The idea is that when you have an abundance of energy, you store that energy as rotational energy. In other words, you spin up the disc with your excess energy when there is abundance. And then when there's a lack of energy in the system, you can then engage that rotational energy and pull it back out again. And because of the law of conservation of momentum, you can essentially, you know, you can then pull that energy back out of the flywheel. When I first heard about this many years ago on a TV show in Australia, it was called Towards 2000 and eventually Beyond 2000. And what it was, it was a show on the ABC that looked at emerging technology from around the world. And there was a bus in Europe that they were developing that used flywheel technology to store brake energy from the bus. So every time you would brake the bus, all of that, the braking energy was coming from the flywheel. So, the flywheel was not spinning very quickly. As you braked, that energy was used to spin up the flywheel and that resistance from inertia, that slowed the bus down rather than using brakes, which are just carbon on a steel disc and essentially just dissipating heat. So, then of course, when the bus needed to accelerate again, they would recouple to the back wheels and the flywheel would then give it a kick to start the bus going again. So, it sounded like a great idea, but it never really took off and there's a whole bunch of reasons for that. So, essentially, and I guess I mentioned that because that's the first time I heard of it. Truth be told, that's really not one I want to talk about too much because the feedback is as it relates to industrial scale energy storage or personal energy storage like in your garage or in the shed in the backyard for rather than using batteries. So that's the whole the point of this feedback. So the applications for cars is really beyond the scope of this discussion. Still there is a great link in the show notes where in from an article in The Economist from a few years ago oddly enough and it actually talks in pretty easy to understand language about a lot of the technical implications of flywheels in vehicles. So if you're interested in that, then please check out that link. It's quite a good read, actually. So the problem was with early flywheels that they were made heavier. The idea is that energy stored is proportional to their mass. So the heavier it is, the more inertia it has. Therefore, you spin this thing up, it'll store more energy. However, the problem was that at high speeds, the flywheel would then just rip itself apart. And the heavier it was, the more likely it was to rip itself apart, especially using something like steel. However, the thing is that if you double the speed that it spins, it actually quadruples the amount of energy stored. So, the focus changed from being on mass of the flywheel to speed instead. But in order to get higher rotational rates, they had to move away from steel. So, more modern flywheels are made of carbon fiber composite, which is a lot stronger than steel. One of the expenses with flywheels is they have to have a protective case or shell that has to be able to withstand the impact of the shattering of the spinning disc inside. So, if there's a defect, if there's a crack and the flywheel flies apart, then this device has to contain it entirely inside in such a way that you don't get a shard of carbon fiber flying out and then hitting someone who was walking past or punching a hole in the wall across the room. It's not cheap to do that and it's definitely not light. So, these things are not portable. But in these sorts of applications, that's okay, I suppose. So, ultimately, friction is your enemy and like anything stored mechanically. The longer it spins, the more energy you lose to friction. And that could be, first of all, let's talk about gas or air. If you want to think about air being a gas, whatever. What that does is it causes parasitic drag. And essentially, there's actually a good article on Wikipedia about that if you want to learn more about parasitic drag, but it's basically where you have a solid and then you've got a fluid, and air acts as a fluid. And as it's spinning, essentially, you will get a resistance pushing in the opposite direction. And that's where at the point of interface between the gas/liquid and the solid that's spinning. So in order to avoid that problem, flywheels have to be kept in a sealed near vacuum environment, kind of like a way a thermos bottle works. So that's the first thing. Second thing is the bearings. Mechanical bearings are just too lossy. They just, too much friction no matter how much grease you put on them, they will just bleed all that energy away through friction. So the best kind of bearings to use are magnetic bearings. However, the problem is that magnetic bearings like purely permanent magnetic bearings have issues with alignment and stability at high speeds. What they found though is to get higher speeds the most effective way of doing it is to use superconducting magnets. And there's a great article again there's a link in the show notes to it I invite you to to read that if you're interested about some of the reasons why they want to use superconducting magnets in essentially in high-speed large flywheels. So, unfortunately, if anyone knows anything about superconductors, once we're involved with them, we're stuck with cooling them. Because high-temperature superconductors are still a long way from room temperature. I think the best is a negative 98 Celsius, and that is still very, very cold, long way from room temperature and they consider that high temperature. So, you're down to cryogenics of some kind, you're down to liquid nitrogen, usually the one they use because it's the cheapest, but with any kind of liquid at that temperature, whether it's liquid nitrogen, liquid oxygen, well, they wouldn't use liquid oxygen or anything else. The problem is liquid helium maybe. There's an effect called the Leidenfrost effect and it's annoying because Because when a liquid that touches a solid and the solid is of a significantly higher temperature that's above the boiling point of that liquid, essentially that gas turns, that liquid turns into a gas at that contact point. Once it does that, the gas literally insulates the liquid, further liquid surrounding it from directly touching the surface. And that insulating effect essentially prevents further cooling of that object, whatever it is you're cooling. And it's a pain in the neck. There are ways around it, but it adds complexity and it adds cost. It's kind of like a thermos. Yeah. The thing is also with liquid nitrogen, it may well be cheap to make, but you've got to be handled very carefully. And even though you can store it in a thermos and in special storage containers or basically just large thermos flasks, it still will go off. Eventually, it will warm up to a point where it is a gas and that's the end of it. So you've got all this extra handling that needs to happen and all this extra complexity and cost in order to keep the superconductors cool, in order to keep it spinning at high speeds without friction losses. So for these reasons, the only area that flywheels have had any involvement at all is in vehicles. I said I wasn't going to talk about it, but just quickly, flywheels are good for that application because they store energy for short periods of time only between acceleration and braking. So recently they had them in, I think it was Formula 1, and for using for short periods of boost. But the fact is that they just don't scale. They don't really work on a private level because of all the cooling that's required. And if you go with the cheaper methods of the permanent magnets, you simply won't be able to store enough energy in them to make them worthwhile. However, as high temperature semiconductors actually do approach room temperature, and that is only a matter of time, I believe, that they'll come up with one that does. So with better material science, maybe one day flywheels could replace batteries in some ways. However, at this point, I have a long way to go. So that was flywheels. Next one had some feedback from Chris Oten via Clinton Phillips. Chris Oden was a former editor of the Australian Macworld, and he raised the interesting issue of recycling solar panels. The concern is, as I said, if everyone goes solar, and solar panels last for 25, 30, 40 years, if everyone, if there's a mass adoption, and there is at the moment, a mass adoption going on, what happens in 40 years' time when there's a massive supply of essentially dead solar panels? What do you do? Now, I did say that during the episode that it was that they were recyclable, but I really didn't go into any depth. In the slide deck that I prepared for one of my previous jobs and I published along with episode two, the slide deck is called Solar System Design. On slide number 38, which is nearly at the very end, essentially, I published back on this on tech distortion a month or so ago. Basically, I did address the issue of recycling the solar panels, but I didn't go into a huge amount of depth. So there's basically three components that are recyclable. The first one is the glass, and that's pretty obvious. It's very easily separated from the rest of it, melted down and recycled, no problem, highly recyclable. Aluminium frames and support ribs, again, same deal. They're easily separated and recycled, just like an aluminium can. The third one, and more interestingly, the third one is silicon wafers themselves. These are a little bit more difficult, mainly because they're bonded to special backing and support and all that sort of stuff. So there are ways that they can separate them and they can reclaim most of the silicon that was used in them to be reused. And I found a really good essay written by a gentleman by the name of Nick Weadock, or Weadock, called "Recycling Methods for Used Photovoltaic Panels." A link's in the show notes. If you're interested, please check it out. It's quite good. and it concludes the cost of recycling new solar panels is actually pretty good. But, you know, rather perhaps obviously, I think, there simply is not enough infrastructure to support the sort of scale that is going to be needed in coming years. But I look in this to the way that batteries used to be held, used to be dealt with. Because when I was a kid, batteries just went to the dump, right? I mean, no one was really too concerned about heavy metals and toxic chemicals getting into the water table. I mean, I should have been. Nowadays, that's criminal. You can't do that because there's recycling plans in place where you drop the battery off at a special place at the rubbish dump, the rubbish tip, and they will then take the, drain the acid and process that so it becomes inert. And then they'll take whatever lead is left and they'll melt that down for use as anything from lead sinkers, to lead linings, to new plates for new lead acid batteries. So, eventually, it's only a matter of time before they scale it up for exactly the same thing they've already done for car batteries and TV sets and all sorts of other things that they recycle. So, that was recycling of solar panels. The last one I just want to talk about, if you're still hanging in there. A gentleman by the name of Brian Nering via a feedback form and he asked the question, would there really be an improvement in efficiency with purely DC appliances? I made an offhand, oh is it offhand? No, not really. I made the comment that it would be better if we just use DC. So his point is that you have to do switched mode conversion to go from DC to DC voltages. So you you want to change from 48 volts to 120 volts or down to 24 volts or five volts, you have to do some kind of switching of the power on and off, on and off, on and off. And that's where you get losses from. Just as much as you do going from DC to AC, whether it's 120 volts AC, 230 volts AC, whatever, and then back down again in the appliance. So his point was, you have switching losses no matter which way you choose to go. So I thought I'd like to elaborate on that. And there's actually quite a bit to say about it. I found a report from the IEEE, which is the Institute of Electrical Electronic Engineers, which is a global organization. And it suggests that there could be as much as a 25% efficiency improvement moving from AC to DC in a household environment. However, I really doubt that that's actually the right number. I think that's one of those optimistic best possible case numbers. I think 15% is probably more realistic based on my observation, because you have to realize that a lot of houses 20 years ago were done with DC if they had without solar, because inverters were simply so expensive. And multi-point power tracking on solar panel modules simply didn't exist. Now we've got microcontrollers and everything built into these things. The cost of the IGBTs for the switching of the power is so cheap. We can do this conversion to AC. But my point is that it doesn't make sense to do it if you're striving for the maximum amount of efficiency. So, I'd say about 15%. And that 15% equates to less, 15% less batteries, 15% less solar panels, and that could be thousands of dollars of upfront costs. Oh, yeah. You know, it's, yeah, people say 15%. Oh, well, yeah, it's like, it's like a 50% saving or something massive. Well, actually, it is massive. Yeah. 15% is huge. It is. And but you know, I was thinking about people, my lawyer 25% that's like a quarter. So yeah, but I don't believe that number. Find a stockbroker over promise you 15%. It's decent. So it's worth striving for. So let's break that down and look at the individual pieces. There's a few angles there that I did the IEEE report did not cover, which I want to cover. First one of those is AC has a problem. And AC's problem is that its voltage is quoted in root mean square. So the RMS value is the square root of two of its peak value. So a sine wave goes up and down, up and down, up and down in a nice sinusoidal waveform. It's all very pretty. But the point is that the root mean square voltage has essentially got to do with the effective amount of power that is actually transferred. So you actually do your maths based on the RMS value of the voltage, not the peak value. when it comes to insulation, actually the peak value is the value that matters because if insulation is going to break down, it's going to break down at that point where it hits the peak voltage. And remember, it's a positive and negative peak as well. So, you've got a voltage differential with AC that is swinging. If it's 120 volts AC, that's an RMS value. But the peak voltage of that is actually 170 volts and that's plus or minus 170 volts. So you now have to design all of your power cables to handle 170 volts minimum plus a bit of headroom. So they'll always oversize the cable insulation or make it thicker than it needs to be. But the funny thing is, if it was all DC, you wouldn't have to worry about that 120 volts DC is 120 volts DC. So your installation would not need to be as thick and therefore over a lot of cable that will be cheaper. So it's not necessarily an efficiency thing in that case, but it's definitely a cost saving. So AC also has the downside of coupling and we talked about this in the episode. And it couples through magnetic coupling with parallel cables. It's much more evident over a long transmission line. So if you're transmitting power over 10 kilometers, you know, 30 miles, whatever, then obviously the impact of this can be far more noticeable. Around a house, we've got wires going up, down, left, right, all over the place, it's going to be very difficult to quantify, but there will be loss. Honestly, my best estimation, I would say on an average household, you're probably looking at about one to 2% of loss through coupling. Now that's not a huge amount, but it's part of the contribution to that 15%. So going DC essentially eliminates that. So that's a cost saving right there. But the biggest efficiency gain is the avoidance of AC altogether. Because once you go AC, you've got something called total harmonic distortion. And the thing with THD is that when you switch, when you switch the power on and off, you switch it on and off as a series of pulses and the width of the pulses changes. So they start extremely narrow, get slightly wider, slightly wider again, till a big fat wide one, then they get narrower and narrower again. And then when you filter that, you end up with a waveform that looks like a sine wave. That's when you're switching from DC to AC, which is what most of the current systems does like the one in my garage. The problem with that is it's not perfect. I mean, it works and pulse width modulation is a wonderful thing. However, the problem is that no matter how much you filter it, you will always have what they call harmonic currents and voltages. Now these harmonics represent, if you look at them on a spectrum analyzer, you'll have the fundamental frequency of your voltage. In North America, that's 60 Hertz. In Australia, that's 50 Hertz. Europe, I think, is also 50 Hertz mostly. So you'll have your harmonic frequency, but you're gonna have a whole bunch of, sorry, you have your fundamental frequency. You have a bunch of harmonics, and those harmonics will be double, like double, triple, odd and even harmonics and so on and so forth, up and down the spectrum. And the amount of energy in each of those harmonics is loss. So that's a problem. And it's not a problem if you've got DC. I mean, yes, when you do DC to DC switching, yes, you are going to get noise. It's a lot easier to filter out on a DC to DC converter. You'll never get rid of it, but then you'll never get rid of it on AC and you've got a lot of harmonics because you're trying to create a fundamental frequency using a non-ideal source. Whereas a DC, you may well chop it into different pieces and if you've got a boost converter or buck boost converter or whatever else, there's an article on LinkedIn, Wikipedia, please check it out if you wanna go into the details the nitty-gritty of DC to DC converters but the point is that it will always be a smoother output than you will get from an AC DC to AC converter. So in terms of actual efficiency however, the reality is and the numbers vary because every design is different but based on the information that I was gathering I knew it wasn't a heck of a lot but the efficiency of DC to DC over AC to DC is about 8% so that 8% goes in with that 1% to 2% from before and essentially there are a few other losses but they're harder and more specific and I won't go into those but the key ones are those. It depends on the converter's design and depends on the output requirements of the DC to DC converter as well so that 8% could be a lot higher it could be 10% 12% on its own so I hate saying it It all depends, but it does all depend because it's kind of like saying that the efficiency of all DC to DC, all AC to DC converters are the same, but we know that they're not because if you get an Apple five watt iPod charger for your wall plug pack, that is an AC to DC converter that drops it down to five volts, guarantee you that its efficiency will be different from a run-of-the-mill black wall wart or wall plug from no-name brand from somewhere in Taiwan. That is not to say that there aren't ones in Taiwan that are better, I'm just saying they'll be different. So it's hard to give a definitive percentage on that. But I thought about what would be the biggest thing in a household that would benefit from going to DC and there is a perfect example and that is LED light bulbs. So LED light bulbs, LEDs are just diodes, they're LED is light emitting diode and a diode is just a PN junction and when you've got a PN junction as a diode it only allows current in one direction which is forwards. So AC is already a bad choice because what that means is that you're only going to get light 50 times a second it won't be continuous it'll flicker because you're only going to get on that positive part of the sine wave where you've got forward voltage over the PN junctions the only time you're going to get light. One of the ways around that is to have diodes in both directions. So you get a burst of light in the forward and a burst of light in the reverse direction from two different LEDs wired differently. But that's a bit wasteful because I mean you're using twice as many LEDs as you need to. So the solution that most of these LED light bulbs currently use is they have an AC to DC converter in them. You know they take the the AC 120 volt 240 volt whatever it is, they rectify it and they turn that into a DC signal and regulate it down to the required voltage in DC. Because LEDs are essentially, you want a DC constant current going through them so you get constant light output. Now there's other issues of course with LEDs in the higher current you pass through them the more light you get that's great but unfortunately the more current you put through a diode you have to be able to cool it otherwise the 0.7 volt drop across the diode you get from many pn junction that dissipation will essentially melt your diode. That's why they've got big heat sinks on these things. I have to get the heat out of the pn junction or it will destroy itself. So even with all of that led light bulbs are still more efficient, much more efficient than a CFL compact fluorescent or a tungsten filament for sure or definitely or definitely a halogen. So what would happen if you had an led light bulb that had no ac it was just a dc voltage straight out of the wall. Obviously, that would be more efficient straight out of the blocks. There'd be no conversion required, no AC to DC loss. That could be a 15% saving right there because there's no AC to DC conversion. And the funny thing is when LED lights first came out, they all were DC. So, this whole AC thing has been a direction they've gone simply because DC is just not available in houses generally. But if you're building a solar house 20 years ago and you were having light bulbs, you did it off DC. There were other low voltage light bulbs, of course, LEDs weren't quite that point yet, but you can still get DC LED light bulbs that are very efficient. So let's say the last piece of this is the lower voltage. So what I would see is in the future, if you had DC in your house, you'd have two voltages. The lighting circuits essentially hidden. It's just a bunch of sockets in the roof. You don't plug something into it. into it. So you're not going to have high current demand because your highest current demand LED light that I've seen is about 13 watts really and that's very bright. You're not going to need that much more than that. So the obvious problem is if I keep putting a lower and lower voltage because P equals VI, the problem is going to be that my current requirements will go up. So if I have a lower voltage like say let's say my lighting circuit's 12 volts and that's fine because low voltage is better for the LED lights is less that they need to regulate away and it's all good. So what you'll do then is you'll run the cabling around the house just like you would for the AC light bulbs, but you'll have about 10 amps roughly. Most electrical cables are sized for 10 amps. So the copper cross-sectional area, the copper can handle 10 amps. At that point, let's say you've got eight watt LED light bulbs. And I'm saying eight watt because I'm just plucking that number out of my head because that's the last size LED light bulb I bought. But I've got a couple of nines and a couple of sevens, let's say eight as an average, that'll give you 120 watts per circuit and that will be about at eight watt led bulbs that's about 15 lights per circuit that's pretty good that's pretty comparable to what you'd expect now 240 volt bulbs with exactly the same copper in the roof so there's no cost difference there at all. Now because you've got larger appliances that require more current you would not have them at 12 volts you would have them at some higher voltage probably 48 volts or 24 volts either or you're going to have more current requirements so you may spend a little bit more on copper up front potentially maybe they'll standardize on 96 volts or you know 120 volts I don't know but if it's DC then obviously the higher the voltage because P equals VI the higher the voltage the less the current and you're trying to reduce the amount of current because you don't want to have really thick copper cables because you go up in current, copper is expensive, you don't want to spend that money on your cables. And that's all I have to say about that. So thank you very much for the feedback, everybody.
Duration 26 minutes and 22 seconds Direct Download

Show Notes

Related Links:


Premium supporters have access to high-quality, early released episodes with a full back-catalogues of previous episodes
SUPPORT PRAGMATIC PATREON APPLE PODCASTS PAYPAL ME
STREAMING VALUE SUPPORT FOUNTAIN PODVERSE BREEZ PODFRIEND
CONTACT FEEDBACK REDDIT FEDIVERSE TWITTER FACEBOOK
LISTEN RSS PODFRIEND APPLE PODCASTS SPOTIFY GOOGLE PODCASTS INSTAGRAM STITCHER IHEART RADIO TUNEIN RADIO CASTBOX FM OVERCAST POCKETCASTS CASTRO GAANA JIOSAAVN AMAZON

People


Ben Alexander

Ben Alexander

Ben created and runs Constellation.fm and Fiat Lux

John Chidgey

John Chidgey

John is an Electrical, Instrumentation and Control Systems Engineer, software developer, podcaster, vocal actor and runs TechDistortion and the Engineered Network. John is a Chartered Professional Engineer in both Electrical Engineering and Information, Telecommunications and Electronics Engineering (ITEE) and a semi-regular conference speaker.

John has produced and appeared on many podcasts including Pragmatic and Causality and is available for hire for Vocal Acting or advertising. He has experience and interest in HMI Design, Alarm Management, Cyber-security and Root Cause Analysis.

Described as the David Attenborough of disasters, and a Dreamy Narrator with Great Pipes by the Podfather Adam Curry.

You can find him on the Fediverse and on Twitter.