 Hello and good afternoon, good evening or good morning, depending on where you're joining us from. Welcome to Engineer for Change or E4C for short. Today, we're pleased to bring you this month's installment of E4C's Off-Grid Energy Webinar Series, focusing on Design of Off-Grid Systems Part 2, System Design. My name is Mariela Machado and I am Program Manager here at Engineer for Change. I'll be the moderator for today's webinar. The webinar you're participating in today will be archived on our webinars page and our YouTube channel. Both of those URLs are listed on this slide that you're seeing right now on the screen. Information on upcoming webinars is available on our webinars page. E4C members will receive invitations to upcoming webinars directly. If you have any questions, comments and recommendations for future topics and speakers, please contact the E4C webinar series team at webinars at engineerforchange.org, as you see on the slide. If you're following us on Twitter today, please join the conversation with our hashtag at E4C webinars. Before we move on to our presenter, I would like to tell you a bit about engineering for change. E4C is a knowledge organization and global community of more than one million engineers, designers, development practitioners and social scientists who are leveraging technology to solve quality of life challenges faced by undisturbed communities. Some of those challenges include access to clean water and sanitation, sustainable energy, improved agriculture, lack of connectivity and more. We invite you to become a member. E4C membership is free and provides access to news and thought leaders, insights on hundreds of essential technologies in our Solutions Library, professional development resources and current opportunities such as jobs, funding calls, fellowships and more. E4C members also enjoy a unique user experience based on their side, behavior and engagement. Essentially, the more that you interact with the E4C site, the better we will be able to serve your resources aligned to your interests. For more information, please visit our website www.engineerforchange.org to learn more and sign up and become a member. It's free and it will take you around two minutes, so be sure to do that. Today's webinar is the final in the Off-Grid Energy webinar series as you can see on this slide. Additional topics covered in this series are drawn from the book, Battery from the Mentals of Off-Grid Electrification, authored by our presenter, Dr. Henry Bowie. The past webinars in this series are listed on this slide and can be found on our website under professional development and webinars. So be sure to check that out if you missed the past webinars. For reference, you can find examples of off-grid energy products like the Mobisul Solar Home System, as you see on this slide, in the E4C Solutions Library. There you can learn more about technical performance, compliance with standards, academic research and user provision models of the systems. All the information is sourced by E4C research fellows and reviewed by our community of experts. It's available for free to E4C members, so be sure again to sign up. You can find the Solutions Library on their Learn and Solutions Library on our website. A few housekeeping items before we get started. Let's practice using the WebEx platform by telling us where you are in the world. In the chat window which is located at the bottom right of your screen, please type your location. If the chat is not open on your screen, try clicking the chat icon at the bottom of the screen in the middle of the slide. You can use this window to share remarks during the webinar. And if you have any technical questions, just send a private chat to Engineering for Change admin. So let's take a moment now and do that. Type your location on the chat window right now. So I'm seeing Chennai, Colorado, Dallas, Denver, Arizona, Milwaukee, Palestine, Italy, Spain, the Emirates, Czech Republic, South Africa, Alabama, wow. From all over the world, welcome. We're thrilled to have you here. So now you know where the chat window is. So be sure to type any remarks, comments that you may have during the webinar in this window. During the webinar, please use the Q&A window which is located right below to type in your questions for the presenter. Again, if you don't see it, click the Q&A icon at the bottom of the screen in the middle of the slide. Very important to note is that we will stop 15 minutes or 10 minutes before the hour. So 11.45 a.m. Eastern, but we'll make sure to set aside enough time for the Q&A. If you're listening to the audio broadcast and you encounter any trouble, try hitting stop and then start. You may also want to try opening WebEx up in a different browser. If you keep having issues, please again contact through the chat window or Engineering for Change admin. E4C webinars qualify engineers for one professional development hour. To request your PDH, please follow the instructions on the top of E4C professional development page after the presentation. As seen on the screen. Great. So that's all I had. So let's take a moment now to tell you a bit about our presenter before we get started. Dr. Henry Louis is an associate professor and Francis Wood Endowment Research Chair in the Department of Electrical and Computer Engineering at Seattle University. His research areas include electricity access in developing communities, renewable energy and appropriate technology. He's the president and co-founder of Kilowatt for Humanity, a non-profit organization providing electricity access and business opportunities in Sub-Saharan Africa. Dr. Louis served as a Fulbright scholar to Copperville University in Kidway, Zambia. He's recognized as a distinguished lecturer of the IEEE and is an associate editor of the Journal Energy for Sustainable Development. He's the author of the book, Upgrade Electrical Systems in Developing Countries, published by Springer Nature. Welcome Henry. Thank you for joining us and over to you. Yes. There you go. All right. Thank you. Well, it's an absolute pleasure to be back again for our sixth and final webinar on off-grid energy. In today's webinar, we're going to be talking about how we design off-grid electrical systems. So if you haven't been a part of our other webinars, again, they're archived and don't feel like you won't be able to follow what we're talking about today. Although we are going to be drawing from some of those concepts, I think there's enough new content here that you'll be able to follow along and learn something. So today's webinar, like the others, is based off the book Off-grid Electrical Systems in Developing Countries. The book contains more detail than I can get into today or any of our previous webinars. It contains, for example, problems, homework problems, all of that additional content. You can get it on Amazon through Springer, who is the publisher. And you'll find a bunch of other material on my website, drhenryluby.com, a slash book. So today's webinar, we're really going to be focusing on the design and implementation of off-grid systems. And we're really going to focus more on mini-grid systems and in particular the design of the energy source and energy storage aspects of that. So the design of the distribution system we're going to leave for the book. So today what we're going to do is I'm going to step you through the typical life cycle of an off-grid project. Then we're going to go over typical methods that you might use to design off-grid systems. And we'll highlight some of the component properties that you should be aware of when you select component. So what I've shown here is a typical life cycle of an off-grid system. Now different organizations are going to have slightly different life cycles depending upon their objectives and goals. But I think this is a fair representation of the steps that one goes through when they deploy an off-grid system like a mini-grid. We begin with prospecting and screening. And here we are just looking for communities that might be in need and might be a good fit for our off-grid system. We will window that list down to a few communities and we'll go visit them and that's the site assessment step. Based on the data that we've gathered from that site assessment we'll make some sort of decision on which community or communities to install the system in. And then we move on to the technical and commercial design which happens more or less simultaneously in an iterative fashion. And that's going to be really the emphasis of today's webinar. We then move into the pre-implementation stage followed by actually installing the off-grid system. And then we move into the ongoing operation and then finally we might decide to expand our system or retire it. So we're going to step through each one of these in greater detail. So a valid question is which community should I look to implement the off-grid system? In many regions there could be hundreds or thousands or even more communities that really need access to electricity. You could literally throw a dart at a map in any community that that dart hit could use an intervention. So which one should you choose? And we also note that the success of that off-grid system no matter how you define it really depends upon the location. So sometimes your off-grid system will thrive in one location but it won't thrive if you had installed it in another. So the task at hand then is to quickly and inexpensively screen a whole host of potential communities. And fortunately there are some online tools that exist today and there are more that are coming out each year that let you do some screening. So this is the electrification pathways tool that is a neat tool. It lets you visualize the communities and screen them based upon like distance from the grid of the availability of different energy resources and other demographics. And many of these tools will also give you a basic calculation of maybe the cost of serving energy to that community with various technologies. So it lets you give an idea of where you might be targeting these communities and based upon other factors like maybe the location of your organization's office or maybe contacts that you have in the country or even language consideration, you'll reduce the list even further. The next step is site assessment. And so you can't rely on national level demographic data to really select a community and feel sure about it. You actually need to go and visit it. And so you'll send a small team to visit the site and what you're really trying to do is get a sense of that community firsthand. You'll probably meet with some officials, maybe some of the traditional leadership as well as the government leaders. You want to also make sure that the data you saw in your screening tool is in fact true and believable. So you need to set your own eyes on it. And a typical activity that also happens during the site assessment is doing surveys and focus groups to get a sense for the community to get a sense of the electricity demand and so forth. You might recall from our last webinar that we discussed using surveys to estimate energy production and how that is sort of a challenging thing to do. You'll also begin collecting some information on the potential energy resources there that you might power your off-grid system with. So you might look for hydro resources. You might start taking wind measurements. And then you might be looking for locations for those assets. So you're kind of collecting information that will let you begin a preliminary design. After you've gathered data from the communities, you need to make a decision. And that decision can be which community you're going to work with or maybe a prioritization of those communities. There's several different approaches that you might use to decide which community to have your off-grid system in. Some of the more common ones I have listed here, but basically you want to take a holistic accounting of that community. You shouldn't just lean on one aspect like their ability to pay for electricity. Now depending on your organization, you might have other objectives like looking for health clinics or schools that you could also provide electricity to if that's the mission of your organization. So again, it's really critical that you do this data gathering, that you have a good sense of what you're going to be getting into. In addition, you need to make sure that the community is of need in need of this intervention and that they would welcome it. So some of the characteristics that you might be looking for are shown here. And I'm sure the list goes on and on, but you're looking for things like competition. Will the grid get extended to this community soon or will there be another off-grid system installed? You want to look for that demand for electricity. Make sure that it's going to be used by the people and that there's an ability and willingness to pay if that's what your organization's goal is. There needs to be energy resources. And then other things like the population you want to have. Usually it's preferred for a denser population so that distribution costs are lower. And there's a few other things that are maybe more political in nature that you need to keep your eye out. So after you've selected the community or communities that you're going to install your off-grid system in, you begin to design. So you've already invested now time and money in these communities even though you haven't laid a single conductor or installed a single solar panel. Because you've visited them, you've collected data and so forth. So what you're going to do is you're going to do the technical and commercial design. And these are usually going to be in an iterative fashion and they are interrelated with each other as we'll see a little bit later today. And so what you'll do is you'll come up with whatever targets are meaningful for you and your organization. This could be the capital expense of the design, the operating expense. It could be the access tier that you're able to provide. You might have certain targets that you want to do in terms of energy reliability and quality and so forth. Or you might have other metrics like the average revenue per user. But you will iterate on your design until you come up with one that meets your targets. And if you are able to come up with a design that meets your targets for that community, then you would consider the next community on your list and sort of continue down until you find one that does work. Then we move into the pre-implementation phase. So here you might be doing tasks like seeking permitting. You might be identifying vendors that will actually install the grid or provide you with the components and you'll work through the contracting. This is really important especially if you are a smaller organization that does work abroad to have strong local partners. I think that really you should be working with local suppliers, local installers whenever possible. It supports capacity building. It supports the local economy. And it prevents sort of reverse outsourcing of bringing, for example, you know, American volunteers to a developing country to put people out of work there. So really relying local talent if at all possible. Procurement, you need to have a long lead time on, especially if you're going to be importing components. This can take months. There's a lot of uncertainty. You might have to pay tariffs. So some organizations like my own kilowatts for humanity, we don't accept donated equipment for several reasons. But one, it's more work than it's worth just trying to get it in country. And other things you might do is you might be meeting with a community to identify customers, to sensitize them to the system that's going to be installed. So they have an awareness, they have buy-in, and they really understand what will be happening. Then you move into the actual implementation. So here you're actually constructing the system. You're installing the solar panels perhaps, the distribution system, you're wiring homes. And this can actually go relatively quickly in some cases just a few days. And then really you want to make sure that you verify and commission the system. So you want to make sure that it performs properly, that if you hired a vendor to install something that they did it to your spec, and that you don't pay them until every last issue is resolved. And then the fun part. So after it's installed, the system is up and running. Here you're serving users. You're making sure that maintenance is being done and equipment is being repaired and replaced. And then also importantly, you should be collecting data. You should be collecting technical data about how the system is performing, about the electricity that people are using, about their ability to pay, and about how that grid is really impacting the community. All of these sets of data are really important, and they're really important to share with others. One of my visions for this whole space is that the practitioners, be it for-profit or non-profit or governmental, really share their experiences and share their data. That's how we'll all become better and deploy systems that are more capable of doing good work. So you'll be collecting data, you'll be analyzing it, and that will help you be more effective in your next project. And then at some point you might decide that the demand is sufficient to expand your grid. And this doesn't necessarily mean adding customers. It could be improving the quality of service to the customers or the users that you have by improving the electricity access here. So maybe you increase the availability of the energy. Now in some cases you'll actually have to retire the grid. Maybe the targets that you thought the grid would meet aren't meeting it. Maybe it's not having the impact that you thought, and there's another community that would be better served by the equipment. Additionally, maybe you have the grid finally makes it to that community. So retirement doesn't have to necessarily be a bad thing. When you retire a grid, you really should restore the land and the environment to its original state, to how you found it if not better. The equipment shouldn't just be abandoned there, especially lead acid batteries. They should be disposed of in a proper and responsible way. And then there might be community relations that need to be repaired. It really depends on how you entered the community and what they understand the project, what it's going to be about. So that's the life cycle of the mini-grid. And I'm going to add one final note here. And that's when you do site assessments, when you are involved in that community, one role my organization always follows is to never promise anything. You can get communities really excited about an off-grid system, but really you're not ready to commit when you've visited a community for the first time. And so you can create a lot of problems down the road if people come and they expect you to give them free electricity and everyone gets the TV. And that's sort of how the project will be perceived to be like. So just make sure that the expectations are set early and repeat that message of what the system would be should it be installed, but don't guarantee it unless you're able to follow up on that guarantee. So the remainder of the webinar is going to focus on the technical design of off-grid systems. So that's step number four. I think one of the most important things to keep in mind is that there's this cost versus reliability curve. And as engineers, we usually want things to be highly reliable, but we should never forget that there's a cost associated with that. So our goal then is to strike a reasonable balance between the cost of the system and its reliability. So it might cost, for example, twice as much to improve the reliability from 97% to 99%. And that might seem like good from an engineering standpoint, but you have to remember you could have served two communities for that same price. Would you rather have two communities with 97% reliability served or a single community with 99%? Most organizations, or I should say at least many organizations, target about a 95% reliability or availability of the system. So they understand that maybe sometimes during the rainy season the system will simply not have enough energy to supply their users, but the trade-off there is that the system is less expensive and therefore they can install more of those systems. So this is really important, that curve is really important to understand. So one of the first steps that you do in designing a system is to understand which energy conversion technologies are available or would make sense. And from a previous webinar we understood that really we're going to use solar, wind, gensets or micro hydro. Those are really the only games in town for off-grid electrification. And you're going to look at the availability of those resources, the underlying resources, the cost, their lifespan, maintenance requirements, and several other factors that I have listed here. And what your goal is is to understand which of these are viable and which should you not consider further. Now the typical design process looks like this. So the main inputs are going to be the demand of electricity, so hopefully something like an hourly profile of your best estimate of what the consumption will be, and an estimate of the energy resource. And you take those two inputs, and again those are just the main inputs or several others perhaps, and you will apply one of two types of design approaches, either numerical or intuitive. And then your output are going to be the ratings of the major components like the solar panels, the batteries, and you'll leave the design of wires and fuses and control systems to a later time when you finalize the design of the macro component. So the two design approaches can be described as being either intuitive or numerical. And in an intuitive approach what you're doing is you're following a recipe more or less. So you're consulting an established standard or guideline, or maybe your organization has a rule of thumb that they follow, and you take your inputs and you make some estimates on some parameters, and you end up with the sizes of your component. So it's an open loop process. You don't gain insight into how the system is going to perform in terms of reliability. You simply adjust some parameters and you end up with design. The numerical approach is more high tech. Here it's a computer aided design approach, and it's usually done through a simulation that underlines that computer aided design. So we're going to talk first about numerical design. So here what you do is you usually purchase a program. I think there might be some free ones that are out there, but a lot of the ones are for purchase. And so you have a designer, there's a human in the loop here, and they're going to describe the technical environment and the economic environment in which that off-grid system is going to live. So they're going to describe things like the amount of sunlight or the flow rate of the hydro resource, the different costs of the fuel, and they're going to specify a design, an architecture of the size of batteries. And they're going to rely on the computer program then to simulate that design under those external conditions that the designer specified. And it's often like an hourly simulation over the course of one year or it could be a 20-year simulation. And the output is going to be how that system performed in a technical sense and usually also an economic sense, and it can provide some summary statistics as well. So the important thing to note here is that these programs do not design the system for you. You need to have that human in the loop to propose a design to describe the operating environment, and the design program just tells you how a particular design would perform. So then the human might look at those results and decide that they're acceptable or not acceptable, according to their organization's criteria. And if they're not acceptable, they'll propose a different design, or in fact they might automate it where they're actually simulating hundreds or even thousands of designs to tease out that cost versus reliability curve before ultimately selecting the design that they're going to use. So numerical design is really useful if you have a complex system, a hybrid system, for example, with wind and solar. But the thing to note about it is that the input data requirements are high. You have to have a really good idea of the load profile of how that might change throughout the day, how that might change throughout the year. You need to have an understanding of the wind resource, for example. And so the results are really only as good as the input information. And so just because you're using a computer doesn't mean the results are reliable. So in other words, if you feed it garbage, you're going to get garbage out. If you're making wild guesses on the load profile, you're making wild guesses on the sunlight or the behavior of the wind, your results really aren't going to be reliable. So my organization, we use a numerical approach as a good starting point. So we know that there's going to be some uncertainty in how it will actually perform, but it's a convenient way for coming up with that first design. So we use a program called Homer, and it lets you specify the system architecture easily. It lets you input the different resources that you might be powering your mini-grid with, and it provides the results of the simulation in a nice form. And then it also lets you simulate lots of different configurations, possible configurations, and visualize the results and select the best one. Now, full disclosure, Homer does support my nonprofit organization with in-kind support, so I just want to make that clear. But still, it's a good example of a numerical design approach. Now, intuitive design is, again, when you follow a standard, a guideline, or really a recipe for designing your off-grid system. And there are several standards that you could consult, and what we're going to do is we're going to walk through an intuitive design example for a hypothetical community. And I'm not proposing that the intuitive design recipe that I'm going to present is optimized in any way. I simply want to show you the steps that you might follow and the thought process, because it really illustrates what you have to consider in doing your off-grid system design. So please don't think just because you saw this on a webinar that this is the gold standard for design, using it for illustrative purposes. I'm also only going to look at the design of solar array and a battery. There's other components like charge controllers, inverters that you need to specify and design. Those are covered in the book, but I'm not going to cover them today. So we'll do a hypothetical design for a hypothetical community called Mawasi. And this community is far from the electric grid. It's going to be too expensive to bring the power lines to it. Mawasi is hard to get to for several months during the year, and there's no wind, micro-hydro, or biomass resources available. So really we're looking at solar as our main energy source. A gen set could possibly be used, but we're worried that we're going to have trouble refueling it during the rainy season. So we've decided to use solar based upon that, as well as the fact that it's in a place with lots of sunlight. And so our design horizon is going to be five years. And so we're going to look at the system not only at installation, but five years down the road and make sure that our system is adequate to serve the community as it grows over five years. So we'll assume that we did a survey and we came up with an estimated load profile, and again we expect there to be a lot of uncertainty around that. One of the things that we've estimated is that there's going to be a 5% growth per year. So what you see on this table then is the values that we expect the load to be like when we launch or when we install the grid, and then after five years. So again, these are just estimates, and so we're going to have to account for some of that uncertainty in our design because we know that these most likely aren't going to be the exact values that we encounter. So we start with our load estimation. Then we have to come up with our architecture. And we know that the houses and businesses that we're going to serve, they want AC connection. And so we're going to have to supply AC from our PV system, which means that we need an inverter there. We also know that we're going to need a battery because people are going to want to consume power in the evening when the solar array is not producing. And then we know that our PV array is going to be large enough that we need to have a charge controller with that maximum PowerPoint tracking capability. So this is the overall architecture of our system, and that's a good starting point then for our design. The first thing that we're going to look at is the battery bank. And using this intuitive method for simplicity, we actually can design the battery bank entirely independent of the energy source. So we basically operate under the assumption, at least in terms of design, that the battery bank is supplying the load entirely and that we're not accounting for the production of the energy system when we design the battery. So the main factors that we're interested in then is the DC bus voltage, what voltage the battery bank be, the average daily load that the battery might have to supply, the discharge current, and the required reliability. So we'll address those in the upcoming slide. So the first thing you look at is the selection of the battery voltage. This is also the DC bus voltage. And we're going to use a typical rule of thumb that says the more energy the battery is expected to supply, the higher voltage you want it to be. And so this is just a rule of thumb guideline. And we know that after five years, Milwaukee will be consuming about 10 kilowatt hours a day. And so we're going to select a 48 volt system for the DC side of our bus. We next need to consider the average daily load. So this would be the load that the battery would have to supply in the absence of any energy production over the average 24-hour period. So we simply take the average daily load, divide it by the battery voltage, and then divide it by the inverter efficiency. Because remember, we're on the DC side of the inverter. The load has been estimated on the AC side. So we have to account for the efficiency of the inverter. And so we'll estimate that efficiency to be 85%. And so our final value then for the average load that the battery has to supply each day is 246.25 amp hours. Again, this is at 48 volts. So that's the energy the battery needs for a single day. We need to account for days in which maybe the load is higher or maybe the energy production system is going to be down for perhaps several days. So we describe the reliability that we want the battery to be able to provide as the days of autonomy. So quite simply, the days of autonomy is the number of days that the battery bank can supply the average daily load before being depleted, assuming that no recharging happens. So imagine that the PV panels were stolen or the charge controller is damaged and we're not able to replenish those batteries. So how long do we want those batteries to last? That's the number of days of autonomy. And so typically we'll pick a value perhaps between 2 on the low end and 12 if we want the system to be really, really reliable. And we'll note that there's sort of a nonlinear relationship here between autonomy and reliability. Now the other thing to note is that we want our system to be able to provide the days of autonomy even after the battery is quite old. So remember, we're looking at a five year horizon. So if we want to provide two days of autonomy, we want our battery even after the wear and tear that's undergone for four years and 11 months and 29 days, we still want it to be able to provide two days of autonomy. And we know that batteries degrade over time and with usage. And typically after at their end of life, their capacity is maybe 80% of what it could have been when it was brand new. So we want to adjust our required capacity by that end of life rating. So for Milwaukee, then what we're going to do is we're going to pick a two days of autonomy and an end of life rating of 80%. We're going to assume that after five years, the battery capacity is no longer what it was when it was brand new. It's simply 80% of that. And so we get a value of 615.63 amp hours as a requirement. Now you might recall from our webinar on battery fundamentals that you can't describe the capacity of a battery with a single number. The battery varies based upon the amount of discharge current. And the higher the discharge current, the less capacity, the less charge you're able to get out of that battery. So that means we have to pick a current at which we specify the battery capacity for. And there's two reasonable approaches. We can perhaps select the average current supply or the peak. The peak current will yield a more conservative value. So that's what we're going to select. So to calculate the peak current that the battery will be discharged at, we need to look at the peak value that the inverter needs to supply, which will also be related to the peak load and account for its efficient by the nominal battery voltage. For velocity, then our peak load after five years, we expected to be 1.94 kilowatts. We know that there's some uncertainty, so we'll add a design margin of about 20%. And so the peak inverter efficiency is 2.33 kilowatts converting that to current on the DC side accounting for the efficiency. The inverter means that three will be expected at its peak to supply 58 amps. So therefore our battery capacity then needs to be specified at 58 amps. And we can do a little bit of math to figure out what the corresponding hour rate and C-rate are. So we could continue to just use 615.63 amp hours as the capacity and move on to designing the battery bank connections itself. But there's a few other things that we probably should consider, and I'll go over those briefly next. One of the things to consider is that at the end of the two days of autonomy, so that worst case scenario where the PV panels aren't producing anything for two days, at the end of those two days, do we want the battery to be at a 0% state of charge, or do we want to have something left in that battery, some charge left? And the reason why you don't want the battery to be at 0% state of charge even after that worst case scenario happens is that the battery might be damaged from that deep discharge. We know that deeply discharging a lead acid battery shortens its life. So typically we're going to leave anywhere from 20% to even 10% in that battery, not for reliability, but to improve its health and longevity. So then we can actually adjust the capacity that we calculated by the maximum depth of discharge. And so if we're going to let our battery discharge even during that worst case scenario to 80%, meaning 20% of the battery capacity left, we will need the battery to be 769.49 amp hours. And the final thing that we need to look at is that daily depth of discharge. We know that the life of a battery depends upon how discharged on a regular basis. So what we want to do is we want to look at the battery, the particular battery that we've selected perhaps, and see how deep it can be discharged and still last our five years that we're targeting. So for five years, which is 1,825 cycles, the battery cannot be discharged more than 40% each day on average. We can do a calculation based on our average load and our battery capacity and see that we're only discharging at 32% on average a day. So therefore we expect our battery to last at least the five years. So that's a good check and we don't need to adjust our design further based upon that. The last thing we do is we apply a design margin and here we're trying to account for any errors in our load, the day-to-day variability of loads of the extreme days where our load might be extremely high. We're also accounting for effects of temperature and other losses. So we simply scale the battery by a design margin. And so in Milwaukee we'll just take a design margin of 7.5%. So we end up with a final battery bank target capacity of 827.2 on amp hours when discharged at 58 amps. So importantly, we do the math here and we come up with an hour rate of 14.25 hours. Again, the C-rate and hour rate are covered in our webinar on battery fundamentals. So then we need to pick the battery itself. Let's assume that we're considering a 6 volt battery. In real life you'd be looking at a whole range of batteries. But let's say we've decided we're going to use a particular battery whose voltage is 6. We'll need at least eight of these batteries to be placed in series in order to achieve our 48 volt battery or DC bus target. So we need at least eight of these batteries. But then we need to look at the capacity of the battery and make sure that it satisfies our design requirement. So like I said before, there's not a single value that describes the capacity of the battery. Rather, you'll encounter a table like the one shown. And it describes how the capacity varies at different hour rates. So we know that we're going to be targeting that 14.25 hour rate, which unfortunately the manufacturer doesn't provide. They provide a 10 hour rate and a 20 hour rate. So we're going to have to approximate what it would be at 14.25. And out of convenience, we're going to pick the nice round value of 200 amp hours, noting that it's somewhere between 190 and 220, which corresponds to the 10 and the 20 hour rate. So the number of strings of batteries then that we need is simply our total capacity requirement divided by the individual battery capacity. And we have to round that up to an integer value. So we need five strings then. So our battery design then looks something like this. We need five strings of eight batteries. Each of them are going to supply about 11 amps, 11 or 12 amps during the peak load. And this is our first design then of the battery bank. The battery bank design that is shown is probably conservative. I'm sure we could tweak it. And in general, we'd want to use larger batteries than the 200 amp hour just so that we have fewer batteries there. We usually try to minimize the number of strings that we have for safety reasons. But nonetheless, this is our first approach, our first design. We then move on to designing the energy source. And the overall thought behind designing the energy source is that it should be large enough to supply the average daily load after we account for all losses in the system. And we will design the energy source based upon the month with the lowest resource availability. So this typically is either the winter or the rainy season if we're going to use PV or maybe the dry season if we're going to use micro hydro. If we expect our load to be seasonal, then we will pick the month that has the highest load and the lowest resource and design around that worst case scenario. So because we're using a PV array, we know that we can tilt it at different latitudes. So we would consult a solar database and come up with the average daily installations. We talked about this in the last webinar. And we picked the tilt whose lowest value is the highest because that's the worst case scenario for that tilt. And we see tilting it at latitude gives us 5.08 kilowatt hours per meter squared per day in the worst month. So that's what we're going to select. And so we can actually do a calculation and convert the insulation that we expect and the energy requirement come up with the size of the PV array. In this case we need 2.33 kilowatts to supply that 10.05 kilowatt hours of daily energy. Now it's important to note that we are ignoring losses here. So we need to actually scale this up. And the book goes into this in more detail, but we want to account for losses like those shown on the table on the right. We want to account for the temperature related losses and then add a design margin. So all of these in sum then bring our required PV capacity up to almost 5 kilowatts to supply our system. We then have to design our PV array itself. So here you're likely going to select the largest panels that you can find and that's typically around 350 watts. So we get a sense of the number of panels we need just by taking the required capacity and by the capacity of an individual PV module. And so we know we need at least 14 of these. Now in designing the PV array we need to consider things like the charge controller constraints. And I'm not going to go into that in detail, but the book does. But it will often limit the number of strings that you have or the number of PV panels that you can place in series. And so we'll assume that in this case each charge controller can have no more than 10 PV modules. The open circuit voltage is such that we can't have more than five strings per controller or five modules per string per controller. So then this would be our first design. There's of course several other designs that meet that condition, but here we have the right number of modules and it actually requires two charge controllers. So I walked you through an intuitive design and again I'm not proposing that this design is optimized in any way quite far from it. In fact this just gives the designer an idea, a starting point. And what they will do after this is come up with a cost estimate for this design and then decide if they're selecting the right size batteries. Will it be cheaper if they use some that have a larger capacity. They might tweak the days of autonomy and other parameters ultimately ending up with the final design. So that's how we do intuitive design. And before I leave, I just want to mention that when you do work in these at-risk communities, there's quite the potential to do more harm than good. And just as one example, I used to love taking pictures of children and showing the pictures to them on my cell phone. It was a really touching moment to do that. But I always thought it was such a fleeting experience and I wanted to give them sort of a keepsake. So I went and got a Bluetooth connected pocket printer and I could print images on demand. And so the next time I went to this one of these communities, I printed a small picture of a group of children, a bunch of boys by this lake. And after printing it out, we decided we'd give it to one of the boys who had a particularly bright smile. So I handed the picture to that boy and all the rest of them started beating him up because they were quite jealous. And of course, we stopped it and I pulled the picture away. But that's just an example of how good intentions can go wrong. And I actually keep that picture. I'm looking at it right now on my desk as just a reminder of that very fact. So what I started to do after that then is I only did family portraits like the one that you see here. And I'd always give the picture to the mother of the household. So the other moral of the story here is to learn from our mistakes. So with that then I will close my last webinar here in this series. I'm happy to answer questions that you might have. So thank you so much for your attention and I hope you found this whole webinar series valuable. And please feel free to reach out to me offline as well. Great. Thank you so much, Henry. As usual, incredible webinar. We have many questions as usual. So I will try to ask a few of them and see where we get. The first one we have here and we're now open for Q&A. So if you have any questions you type, but I have over 20, so let's try to go through them. Could you please give us an example, and this is for Henry, of design programs that allow deterministic simulations of the system? Yeah, I think, again, you know, like for what it's worth, they provide in-kind support to my organization. But even before they did that, I used Homer. It's, I think you can have a free trial of it and see if it meets your organization's needs. So it's just Homer Energy, if you do a search for that. There's actually more activity in this space. If you look at computer-aided design, I think Odyssey Energy, I think that's their name. They have some computer tools that help. RETScreen also does off-grid system design. So there's a few options that are out there and you'll have to find one that matches, you know, what you want as well as what you're willing to pay. Those are just a few, but there are several, and more are coming out each day or each month, really. And there's a follow-up question to this, I guess. Is there another computer simulation program out there that is also very good besides Homer? And I guess you answered that question. Yeah, well, I will say that some universities have come up with their own design problems, or excuse me, design programs. And those might not be as well supported or have as many features, but there are some universities that do have design programs. So if you do enough internet searching, you'll find them. I also have a couple of them listed in my book as well. Great. This is also a question related to Homer. Why not let Homer to select battery sizes based on hourly load instead of sizing before simulation? Could you repeat that one? Why not let Homer to select battery sizes based on hourly load instead of sizing before simulation? Yeah, so unfortunately, you know, you have to let Homer know what size batteries you wanted to simulate. And so you can just pick a solution space or a bunch of different sizes of batteries and then let Homer select from that. You absolutely can do that. But you need to have some sort of starting point to really make the simulation a little more effective in finding what works for you. And you don't need to go through. What I went through today at the latter half of my webinar was not suggesting that you follow this intuitive design and then you use computer aided simulation. This would be if you decided not to use a computer aided simulation, if you wanted to just do it, come up with a design sort of on your own. If you were to use a computer aided simulation, you need to have just a ballpark estimate. So if I had a ballpark estimate, the different sizes of a system, I would start by taking my average daily load and figure out what the PV array would be, assuming maybe four hours of full sun. So if I have a 12 kilowatt hour per day load, I would say, well, I probably need at least three kilowatts of PV. I just take the 12 and divide it by four. And that would be like my starting point. And it's not going to be the optimal by any sense, but it gives me, you know, an order of magnitude. And then I would estimate the load based upon, or excuse me, the battery size based upon the voltage level and an idea of the days of autonomy, maybe two or three. And so that would give me a starting point, which I would feed into Homer. And then I would try a bunch of other sizes around that. So Homer has really good documentation. And if you want to really know how, and lots of support. So, you know, if you want to use Homer, they can walk you through that whole process. Okay, great. We have another question here at DC versus AC minigrid. What is the most energy efficient, which is the most cost effective? Yeah, great question. So it depends on the nature of your load. If your load is only going to be DC, then I always recommend, you know, if your load is going to be DC. And if you're using solar panels, then it would make sense for everything to be DC. Some of the caveats to that though are if you, you know, if you're going to have some AC load, even if that AC load ultimately, so it's a computer, for example, ultimately is going to convert it to DC. You're going to have an easier time finding load that is compatible and expecting, you know, 50 Hertz AC at 220 volts or whatever the national standard is. There's going to be more appliances out there that are designed for AC. And so it's going to be easier to get them into the rural communities. You're going to have the economies of scale that kicks in with mass manufacturing. So that that would be a consideration. The distance that you're going to have to transmit the power also is a consideration. It's easier to get high voltage AC than high voltage DC. And generally there's not a clear cut answer. It depends on your application. In the case that you're just doing LED lighting and you're using solar, then I think you make a lot of sense. But as you move towards higher loads, then I end up a variety of loads. I think AC makes more sense. Okay. Excellent. So the next question, what is the best way to decide the coincidence factor to avoid, avoid oversized in the system and your rule of thumb? Well, unfortunately there's not a lot, and we talked about this briefly at the end of the last webinar, but there's not a lot of data there to, and maybe there is, but I haven't seen it, to support a good rule of thumb for coincidence factor. I mean, obviously what you want to do is you want to have an understanding of the users. And if they're all, say, households that are going to use electricity for lighting in the evening and televisions, then you can expect a lot, a high coincidence in your system. So high coincidence of your load. Maybe not, certainly not 1.0, but it might be like 0.8 or something. But that also depends on the number of users. So the more users you have, the lower the coincidence will be just due to random variation. So unfortunately, again, there's not a clear cut answer. And although if you look at the last webinar, I present a graph, and that graph is based on about like 200 households. So if your mini-grid has fewer than 200, you know, you're going to have lower coincidence than what we picked in that example. So I hope that answers your question. It's tricky, and it is something that you need to be aware of. Okay, great. The next question, would you give us rules of thumb again on integrating different distributed generations, namely when biomass power generation and a battery pack all together into a single mini-grid system? So I will repeat, will you give us a rule of thumb on integrating different distributed generations, namely when power and battery and a battery pack all together into a single mini-grid system? Yeah, so my basic rule of thumb is to keep it simple. So if you can get by with just solar, then absolutely just use solar. You shouldn't be doing wind or making hybrid systems just because you fancy them and they sound exciting. So if you have a good solar resource, usually it's best just stick to that solar resource. However, if you are in a situation where maybe there's a lot of sun and it's also quite windy at night, then potentially installing wind turbines would reduce the required size of your batteries. So you have to really know your resource profile if you want to do that. The other thing to consider would be the reliability requirements. And here I think Gensets, which could be powered by biomass, I suppose, can make a lot of sense. You have high reliability requirements or perhaps there's a few weeks a year where you expect the load to be really high or the resource, the sun, to be really weak because of rainy season. And having a Genset that you only run a few hours during those extreme scenarios can actually save you a lot of money because you won't need to have a large battery bank that largely is unused most of the year. So having a backup Genset makes a lot of sense. They're expensive to run, but they're cheap to purchase. So as long as you don't buy, as long as you don't run them that often, they can complement wind or solar quite well. So I hope that answers your question. Again, I would strive for simplicity in everything you do. The more complicated you make the system, the harder it's going to be to maintain. The easier it is to make a mistake on either the design or operational. Great. I will follow up on reliability. So we have one question related to this. It was mentioned that systems are designed targeting around 95% reliability. How is reliability calculated or measured in terms of the energy demand that is covered or hours that the system is available? Yeah. So what we're talking about there is what we might call the loss of power supply reliability. So it's simply like the number of hours or the percent of hours that you have the system available. So 95%, that's one out of 20 hours that you aren't available. And that might seem like a lot, but I think in practice those hours tend to be clumped up together. So maybe it's a few hours overnight during the rainy season here and there. And again, 95%, that is not going to work for every organization. I know we design around 95%, but we really have much higher reliability than that. And a lot of that just goes into our engineering bias to oversize things and to make conservative estimates. So that being said, you shouldn't pick and design around 100% or 99.99%. You're going to end up with a massive system. And you should also make sure though that people understand the reliability. Fortunately, or perhaps unfortunately, the requirement to have their electricity 24-7, 365 days a year isn't met by the national grid either. I mean, their outage rates can be quite high. And 95% might actually be comparable with that, with the national grid. Great. Okay. Dr. Henry Louie, I really want to thank you. This has been an amazing webinar series. We're out of time, unfortunately. But I will end up with one last question and then remind everyone that all the webinars are on our website. Engineering for Change admin just posted the link, so be sure to check that out. They have additional questions regarding online courses on the design. Any recommendation that you have on online courses on the design of a mini-grid integrating different distributed systems like Coursera or Platform? Any that you would recommend? You know, I won't be able to recommend a single online course. I know that you're seeing more and more of these courses being offered. I'm not aware of an online course, a true online course that goes into the design of off-grid systems in the developing community or the developing world context, which I think is important. I'm sure you can find some that cover mini-grid design, but that's going to be a very different context. That's probably going to be for military bases or college campuses or situations where you have the need for extremely high reliability and you're going to have the latest technology that you're implementing. What we're doing is something far different. We're usually using already commercially available systems. Our reliability requirement isn't as high. There's no grid to connect to. So unfortunately, no, I won't be able to recommend any other online source. If you do want more information though, my website, DrHenryLouis.com, I'll be posting slides and a bunch of other material over the next few months that are related to this book and the course that I teach. And I do know that a few other universities have adopted this book and are using them. So maybe if you're a student, you can go bug a faculty member and have them contact me and we can get them set up. Great. Thank you so much. It was an absolute pleasure to have this series with you. This has been super useful and we have had so many attendants and I'm sure there will be many more questions. So I leave here on this last slide his contact. Again, if you go to our website, you will see the whole series, so be sure to check that out. And we thank you so much for attending and have a great day. Thank you, everyone. Thank you.