You know there are suddenly a lot of "brain" options for robot projects. It's not a new concept. My first robots used an Apple II E for a brain. It was great fun, but not very practical. The computer weighed around 100 lbs (with the monitor) and the poor little bot weighed a few ounces. Imagine setting up a roving bot to monitor burglars with that setup! Now we have the Pi, the Edison, Galileo, pcDuino, and others that I haven't even heard of. My favorite, however, which hasn't been considered is the Asus EEE.
Wait the Asus EEE is a netbook you can't hook that into a robot! Why not? Well, you can but not directly. So that said, how and why? Why not just get one of the other systems and build a computer with parts that are open source, and use an OS that is free. Why not build the systems onto computers that are built to directly control the parts and pieces that are "robofriendly"?
Before USB we could use the parallel and serial ports to control the bits and pieces. Now you have to use a FTDI device (like an Arduino) or cable to give that functionality back to the computer. I'm actually using ROB's brain to write this blog. So why not use a netbook? Here are a few of the reasons that I have thought of, and coincidentally, why mine has worked so well;
1. A netbook already has all of the features needed to run a complicated robot. This one has 3 USB hubs, a keyboard, and a monitor. With the ability to hook up to an external monitor. This one has Bluetooth and WI-fi, but I don't use either. It has a battery backup that, under normal functions, can last for around 8 hours, and through some tinkering I can get that to 12 hours and possibly longer. If I can get the cpu to pass out at times, I think I could get it to last longer. It has sound and Ethernet built in. It has an SD card slot. This thing is loaded. It is even faster than its robocousins, and it has an x86 processor so I can run Windows. There is another part that an EEE has that those others don't, and to me it's a crucial part that I am adding here as an afterthought, a Case!
2. I'm comfortable with the x86 architecture. I've written about the cloud, Android, and Unix style OSes before. I don't like them and won't use them if I can get out of it. Not because I'm a fanboy, as you know, but because I just don't like them. I don't like how they work. Some of you may remember my theory about it being a right brain, left brain, thing. After using Windows 7 and 8, I'm convinced of it. With those OSes the brain got thrown out. Still, I think that an EEE equipped with 8 as an OS might be extremely useful. Especially, if the EEE was a new Transformer mounted on the bot.
3. Cost. Cost is always a component of any build. The EEE is a very inexpensive PC. Especially if you go after the after market, or scratch and dent, models. You can get several models for under $300, with everything included. Even a transformer with the keyboard can be had for less that $300. I would stay away from the E-bay sales models though. Parts are hard to find and all of those that I have seen on E-bay are only good for parts. Really $300 is excellent for a brain. If you get the robocousin equivalents you could be looking at more than that for less bang for your buck anyway.
4. Size. As most of you know, I like big robots. They have their problems, but to do anything useful they have to have the size to do the job. That means that they have to room and pull to accommodate the EEE as a brain, but that also means that the EEE is small enough to stay out of the way. Even a moderate sized bot can use the EEE as a brain without being overwhelmed. Also, while it looks big and menacing, the actual components for an EEE are really no bigger than their stripped down counterparts. Inside the EEE you have 3 or 4 boards (depending on the model), a hard drive a battery pack, and a fan. If you arrange the development board, shields, and other parts to get the equivalent functionality to the EEE, you would actually have to take up more physical space by two or three times. Why? The EEE is 1.6ghz to 2ghz, with 2 cores. The closest equivalent that I have heard of is the pcDuino at 1ghz. So you would have to run 2 to get the equivalent power wise, taking up twice as much space. I know some of you will argue, but I can't spell out the entire reason here just for your benefit.
5. Power is another really big reason. An EEE can be plugged into the panels and recharged. If that makes you nervous you could use an inverter on the panel and then recharge. Like I said before, you can get 8 hours under the normal settings. What if you could put the CPU into a super low power usage state? you could theoretically stretch that into days. Even so, you could put a LiFe battery into the mix as a primary and keep the onboard battery pack as a backup. I may show how to do that at some point, as that is the setup I will have to use for the clambot.
In the end what you use for your robots brain is up to you, and whether your bot even needs higher brain functions. A lot of them don't. Those that need to run a lot of sensors, functions, and need to log data need something a bit more robust, and relying on an arduino with a crapload of shields isn't always feasible. Anyway, for me the EEE has always been a handy, tidy, way to go. It's miles better than an Apple II E!
Monday, September 21, 2015
Friday, September 18, 2015
Lead and LIthium
There are a lot of people, "experts", some might claim, who argue endlessly about the virtues of lead acid batteries, lithium batteries, and everything in between. Ultimately their arguments always come down to cost. I care about what works more than cost and so my arguments are a little different. I care about more bang for my buck and the quality of craftsmanship. I can't afford to keep spending money on something that shouldn't break or wear out after only a week of use. My projects always seem to be broken often enough without crucial parts aiding their state of being broken. I digress, this is supposed to be about batteries. This may be a "growing post", one that I add to as I think of things that I've forgotten.
Lead Acid is the most common battery type. It's over 150 years old. There is a lot of information, misinformation, and myths concerning them. Some of which is still viable, and some not. Here are a few things about lead acid that you should know;
1. LA batteries are heavy.
2. LA Can be lethal. Unsealed LA spew toxic fumes and must not be used in areas where you want to stay alive. I know of people who have used them in houses. My parents had one hooked up to a CB radio when I was a kid. Our house was drafty and the battery (which leaked I might add) was near a window. That didn't keep the gases from building up at times. It's a bad idea, don't do it.
3. There are Sealed Lead Acid batteries. SLA batteries can be used in living quarters. These are the only ones that I recommend for robots if you have to use lead. SLA batteries are not as heavy because they are starved. That means they don't have extra electrolyte adding weight and they won't leak if you crack the case.
4. LA batteries are cheap. If you pick one up and find it to be kind of light, it's really cheap. That is because modern LA batteries use a lead foil or mesh instead of plates. It makes them cheaper to make and ship, but it also makes them last only a few years as opposed to a few decades. Of all of the lead acid batteries I have bought over the years, I only have 2 left and only one is more than 4 years old. Most only last a year or two.
5. LA may have it's shortcomings, but unlike Lithium, LA batteries can be charged while they are being used. That's why ROB uses a LA for his locomotion power.
6. LA batteries make great ballast weights. Instead of using lead weights, or bricks for ballast, use SLA batteries. The space will be comprised of something useful, and you can always use more space to store power.
7.LA batteries should never be rapid charged as it can warp the lead structures and cause the battery to die. So they should be charged slow and steady.
Myth #1 Don't put lead acid on a concrete floor; This is one that used to actually be true. Originally lead acid battery cells were made from glass jars. In cold climates, if you put glass jars on concrete floors and the ground freezes, the jars break. I'm talking about really frozen ground, not just a little cold.
Myth #2 Lead Acid batteries are the only true battery; False. It is a battery only in the sense that it is a battery of cells. These cells are "Super Capacitors". They cannot produce a charge. They merely store it, then release it in a controlled manner.
Lithium Batteries are not all equal and you need to know a lot more about them before you even try to use them. They were invented in the 1950's (I think), but it wasn't until the 1990's that they became viable. Actually, I seem to remember a version of the lithium battery being at least attempted around 1890. If you want to pick nits, some alkaline batteries have had some lithium in them since about that time. The problem is that most of the chemistries suffer from "thermal runaway". They literally melt down. With some of the chemistries that can be lethal, because some of these power houses contain cobalt and some other nasty ingredients. I recommend only using true lithium Iron Phosphate (LiFe), yes there are some fakes out there. True LiFe batteries contain no toxic metals and cannot combust. I prefer Lithium batteries, but they have some shortcomings. All of my information will concern the LiFe chemistry, because it is the safest and a good compromise between "goods" and "bads".
1. Li batteries are light.
2. Li seem to be expensive, but they really aren't. Under good conditions you can harvest several more amps a day with Li when compared to LA, and they last many years longer than LA.
3. You MUST use a BMS (battery management system) with Li. Otherwise you will ruin the battery. Actually, I've never understood why BMS boards aren't used with lead, as well. The life of the battery would be greatly increased, and recharge rates would be improved, because the discharge wouldn't be so great.
4. Li batteries "generate" power. This is confusing to most people. LA batteries only "store" a charge, but Li generate power. This is because for the Li battery to work, lithium moves from one pole to the other. When you "recharge" a lithium battery, you aren't renewing the charge as you would with LA, you are running the battery backwards very quickly to get the lithium back to the negative pole.
5. Li cannot recharge and discharge at the same time. This is because of what I just stated. The lithium cannot move both directions at once. If you try to make it do so, you run the risk of frying the battery. Instead, we use "power paths" to give the appearance of recharging while discharging. I don't understand why doing this works, I just know from experience that it does.
I know this is a lot of boring information, but it's stuff you need to know if you are to understand anything about digital robotics.
Lead Acid is the most common battery type. It's over 150 years old. There is a lot of information, misinformation, and myths concerning them. Some of which is still viable, and some not. Here are a few things about lead acid that you should know;
1. LA batteries are heavy.
2. LA Can be lethal. Unsealed LA spew toxic fumes and must not be used in areas where you want to stay alive. I know of people who have used them in houses. My parents had one hooked up to a CB radio when I was a kid. Our house was drafty and the battery (which leaked I might add) was near a window. That didn't keep the gases from building up at times. It's a bad idea, don't do it.
3. There are Sealed Lead Acid batteries. SLA batteries can be used in living quarters. These are the only ones that I recommend for robots if you have to use lead. SLA batteries are not as heavy because they are starved. That means they don't have extra electrolyte adding weight and they won't leak if you crack the case.
4. LA batteries are cheap. If you pick one up and find it to be kind of light, it's really cheap. That is because modern LA batteries use a lead foil or mesh instead of plates. It makes them cheaper to make and ship, but it also makes them last only a few years as opposed to a few decades. Of all of the lead acid batteries I have bought over the years, I only have 2 left and only one is more than 4 years old. Most only last a year or two.
5. LA may have it's shortcomings, but unlike Lithium, LA batteries can be charged while they are being used. That's why ROB uses a LA for his locomotion power.
6. LA batteries make great ballast weights. Instead of using lead weights, or bricks for ballast, use SLA batteries. The space will be comprised of something useful, and you can always use more space to store power.
7.LA batteries should never be rapid charged as it can warp the lead structures and cause the battery to die. So they should be charged slow and steady.
Myth #1 Don't put lead acid on a concrete floor; This is one that used to actually be true. Originally lead acid battery cells were made from glass jars. In cold climates, if you put glass jars on concrete floors and the ground freezes, the jars break. I'm talking about really frozen ground, not just a little cold.
Myth #2 Lead Acid batteries are the only true battery; False. It is a battery only in the sense that it is a battery of cells. These cells are "Super Capacitors". They cannot produce a charge. They merely store it, then release it in a controlled manner.
Lithium Batteries are not all equal and you need to know a lot more about them before you even try to use them. They were invented in the 1950's (I think), but it wasn't until the 1990's that they became viable. Actually, I seem to remember a version of the lithium battery being at least attempted around 1890. If you want to pick nits, some alkaline batteries have had some lithium in them since about that time. The problem is that most of the chemistries suffer from "thermal runaway". They literally melt down. With some of the chemistries that can be lethal, because some of these power houses contain cobalt and some other nasty ingredients. I recommend only using true lithium Iron Phosphate (LiFe), yes there are some fakes out there. True LiFe batteries contain no toxic metals and cannot combust. I prefer Lithium batteries, but they have some shortcomings. All of my information will concern the LiFe chemistry, because it is the safest and a good compromise between "goods" and "bads".
1. Li batteries are light.
2. Li seem to be expensive, but they really aren't. Under good conditions you can harvest several more amps a day with Li when compared to LA, and they last many years longer than LA.
3. You MUST use a BMS (battery management system) with Li. Otherwise you will ruin the battery. Actually, I've never understood why BMS boards aren't used with lead, as well. The life of the battery would be greatly increased, and recharge rates would be improved, because the discharge wouldn't be so great.
4. Li batteries "generate" power. This is confusing to most people. LA batteries only "store" a charge, but Li generate power. This is because for the Li battery to work, lithium moves from one pole to the other. When you "recharge" a lithium battery, you aren't renewing the charge as you would with LA, you are running the battery backwards very quickly to get the lithium back to the negative pole.
5. Li cannot recharge and discharge at the same time. This is because of what I just stated. The lithium cannot move both directions at once. If you try to make it do so, you run the risk of frying the battery. Instead, we use "power paths" to give the appearance of recharging while discharging. I don't understand why doing this works, I just know from experience that it does.
I know this is a lot of boring information, but it's stuff you need to know if you are to understand anything about digital robotics.
Wednesday, September 16, 2015
One Year Solar Door Update
What you don't see is that this system was also an experiment using a LiFe battery instead of the usual lead acid or more dangerous lithium chemistries. I have been running about 2 amp loads (with peaks of about 4) for the year and it has shown to be a very good system. The only times we have run low on power, are during our 3 day storms.
I think the nicest thing about this setup has been that even though we are only using a 15 amp hour battery, it has done far better than our prior lead acid setups. The LiFe battery can completely recharge in a few hours whereas a lead acid would take days. The only real problem is that we have to have better quality light to open the gates and get the battery charging. Lead acid can charge in lower light or filthy light.Really this proves that lithium does have its place in solar power. It's not the same as a lead acid based system, and it can't compare as such. Both systems are unique and should be treated as such.
Where do we go from here? I may do a breakdown of the system, if there is enough interest, but I think it's ready for expansion. Electricity rates are climbing like crazy now, and the print shop uses A LOT of electricity, so that is a consideration in producing and using as much as we can ourselves. In our case it's not just lights. All of our equipment uses electricity, but we don't have to use the old power hungry equipment that we started out with. This all applies to our robots as well. ROB is proving to be quite the little power hog and he is broken most of the time. We also want our security bots to be solar as well, and that is going to require more capacity than we have at the moment. We'll just have to see where we go from here.Monday, September 14, 2015
No Power! Again
Electrical power seems to be a bit of a pain in the butt. Using it. Generating it. Storing it. I know that everyone has their favorite ways of doing things, and they all claim, near to, or completely 100% up time. To that I say; good for you. I happen to live in what we call "Hell's Armpit". Nothing works here all the time and it seems like things often work in reverse.
I live in a geographic bowl, that was once a super volcano. It has some nasty "habits", that work against almost every form of electrical power production available. I would like to just toss out a solar panel and be done with it. That doesn't work. In the winter when everyone around me are using their heating stoves and driving their vehicles, the pollution gets so thick that the panels stop working. In the summer the same problem exists from increased tourist traffic and smoke from forest fires. I have had to go weeks without any meaningful output. Other times our output is OK, meaning that it will take a day to charge a battery that can charge in a couple hours when the air is clean. So it seems pointless to have huge panels covering everything.
There are alternatives, that could work. Peltier (TOC) units, or whatever you want to call them, could work in the summer, but what about winter? This last winter we were dealing with 56 degrees Fahrenheit, BELOW zero. That isn't very good for electricity generation. I could use a sterling engine, but that again, needs heat in the form of a flame, or in theory, the sun. I don't know about having an open flame on a robot. It just seems like a problem waiting to happen. I could use acid and make a Daniel cell. That might actually work except that to recharge them you have to replace the acid, which I can't get where I live and it would be hazardous to ship. Especially in the quantities needed to run a robot the size of ROB. Wind and hydro are also possibilities, except that ROB doesn't move about that much and waiting for the wind could be problematic.
That leaves hdyro. Pico-hydro actually. If I could figure out how, I've had this idea of a portable hydro generator. I finally found the equations that I need to at least get started, but I'm not sure if it would work. Theoretically, if I could keep it from freezing, I could build a 6' version and it would generate 960 watts at 12 volts. Which is way more than I need. I think I could scale it down and get about 60 watts. That would give me about the same number of amps that ROB produces via solar. Even if it wasn't portable, maybe ROB could just link up to recharge, that would kind of defeat his original purpose though.
I don't know what I will wind up doing, but it's a problem that not too many people bring up.
I live in a geographic bowl, that was once a super volcano. It has some nasty "habits", that work against almost every form of electrical power production available. I would like to just toss out a solar panel and be done with it. That doesn't work. In the winter when everyone around me are using their heating stoves and driving their vehicles, the pollution gets so thick that the panels stop working. In the summer the same problem exists from increased tourist traffic and smoke from forest fires. I have had to go weeks without any meaningful output. Other times our output is OK, meaning that it will take a day to charge a battery that can charge in a couple hours when the air is clean. So it seems pointless to have huge panels covering everything.
There are alternatives, that could work. Peltier (TOC) units, or whatever you want to call them, could work in the summer, but what about winter? This last winter we were dealing with 56 degrees Fahrenheit, BELOW zero. That isn't very good for electricity generation. I could use a sterling engine, but that again, needs heat in the form of a flame, or in theory, the sun. I don't know about having an open flame on a robot. It just seems like a problem waiting to happen. I could use acid and make a Daniel cell. That might actually work except that to recharge them you have to replace the acid, which I can't get where I live and it would be hazardous to ship. Especially in the quantities needed to run a robot the size of ROB. Wind and hydro are also possibilities, except that ROB doesn't move about that much and waiting for the wind could be problematic.
That leaves hdyro. Pico-hydro actually. If I could figure out how, I've had this idea of a portable hydro generator. I finally found the equations that I need to at least get started, but I'm not sure if it would work. Theoretically, if I could keep it from freezing, I could build a 6' version and it would generate 960 watts at 12 volts. Which is way more than I need. I think I could scale it down and get about 60 watts. That would give me about the same number of amps that ROB produces via solar. Even if it wasn't portable, maybe ROB could just link up to recharge, that would kind of defeat his original purpose though.
I don't know what I will wind up doing, but it's a problem that not too many people bring up.
Thursday, September 10, 2015
Another start?
This is yet another attempt at getting things going. I'm still waiting on parts for ROB. I also got a little side tracked trying to clean up his wiring and stuff. As some of you know, I was burgled (again) about a month ago. The other businesses are putting up cameras, but I have to do everything the hard way, and am building what I am calling "SAM"'s. "Surveillance and Monitoring" bots. They are what I call "clam" bots. The idea is that if someone breaks in or whatever, the bot will monitor the situation and record only the important info. You know hot chicks, animals, and of course the bad guys being bad. OK mostly just that last one. They don't move around except their eyes, which move on a stalk. If anyone gets too close, they should "clam" up. If they are picked up, they should send out a massive noise, hopefully loud enough to wake the dead. I need at least 3 SAMs. I would like them to have speech recognition, but we'll see. Anyway that's enough for this update.
Tuesday, March 10, 2015
A Tether for Rob
This is why we prototype. To find the bugs. Rob has a big one. Ok, a few big ones, maybe. He is very violent in his movements. That causes hell with the solar supports. As in, they fell apart. I'll have to redesign that part I think. He also has a problem with too much power. If you can really call that a problem. The car I used is supposed to have a 60ish pound kid in it. I tested it before I started building with about 80 pounds so it can pull a lot. At the moment though it's not getting any traction, so he's too light. I may just have to add ballast to fix that. At least now he can move. The bugs keep it interesting
Monday, February 23, 2015
Robs' New Panel
Normally I advise people to wait until the weather warms to build panels. I was unfortunate enough to be forced into putting Robs' panel together early. The amorphic panel that I was using to keep his battery topped off over the winter failed to do its job at all. $100 POS. Fortunately, I had an experimental panel fail, and I was able to recycle its cells for this one. How's that for thinking ahead. If I didn't have them already strung together I probably wouldn't have been able to put this together. It's just too cold to do any serious work with the cells, the temperature difference would have led to some serious breakage.
Anyway, Rob has power now, I don't know yet if his battery is damaged. I'm in the process of getting some meters in so I can see what's going on, and I still need to get his motors going again. I'm tired of kicking him around the garden.
Subscribe to:
Comments (Atom)