I am just starting to learn electronics, reading and experimenting, at
47, wife and kids, going to school for a hobby right now is not an option. I read in this group and in MANY other places I have read about,
1N4001s, 5555s, 8051s, on and on. In the reading I have done so far it has mentioned a few IC's what I'm wondering is how do you know which IC to use? Is there someplace on the web or in a book that breaks these it groups. It seems SO overwhelming.
Thanks in advance to anyone for here help in this.
Purchase one of those "100-in-One" kits from RadioShack (in-store or from radioshack.com). They come in different sizes, and the bigger kits contain more components and allow for more experimentation and learning. These kits all contain a book that explains what each component does, how they work (at a basic level), and lead you through making different circuits step-by-step. I have even seen them for sale at hobby and craft stores.
The down side to these kits -- they do not cover microcontrollers and microprocessors. But I would advise learning the basics before jumping into a more advanced subject such as a microcontroller.
Things break down into various categories; like microcontrollers, memory, small logic blocks, CPLDs and FPGAs, ADCs and DACs, opamps, voltage regulators, voltage references, oscillators, etc.
When you consider a specific application, you take the important specifications (which you may have a strong hand in deciding) and prioritize them and use your design knowledge to develop some reasonable method to achieve a solution and then break this down into the functional areas needed. Experience will guide you, regarding what is available to help as a matter of standard parts (a hobbyist does NOT often have the luxury of getting customized ICs made.)
Without experience as a guide, you may know what you want to try but not really know what approach to take or what is available to help out. So one good way is to talk here about what you want to try and as much of the specification as you can reasonably describe. I'm sure you'll get some good guidance.
Regarding books, my own experience is rather modest and unique (I used some old electronics books designed to teach WW II repair crews, various hobbyist books from the '50's and '60's, read magazines and their articles when I could follow them, and tried out a few things.) I'm not sure it would be the better path for someone starting out now. So I'll let others much better informed about this recommend some ideas.
It's probably better to start out with some small things you are personally interested in, and then broadening out from there. For example, I was interested (as a kid) in Jacob's ladders and simple AM radio receivers. So that's what I built for myself. But to do so, I had to learn many other things. And that's how learning often starts...
You might start by describing a little of what you've done or tried to do, what has worked for you, what hasn't; what interests you generally have; what immediate projects interest you, for now; what theory you believe you understand; etc.
Also, you may want some tools to use. What kind of budget are you willing to consider?
Finally, have you read through other similar threads on this subject in sci.electronics.basics and sci.electronics.design, using google groups? This kind of question is an oft-asked one and there are many good responses. Is there something about those that misses the mark for you? If so, what?
In other words, help folks to understand who you are, what you know, where your interests lie, and how much time and money you can consider investing.
On 25 Feb 2005 15:01:39 -0800 in sci.electronics.basics, "rollajarhead" wrote msg :
Go get a copy of the old Texas Instruments standard, "The TTL Data Book." If you get into the field, you will use it the rest of your life. In the meantime, you can sometimes find datasheets via a google search, downloadable in .pdf format. The datasheets will make a good accompaniment to reading of basic digital circuits. The prefaces in the "Data Book" are good to read, as they give some description of the operation of the IC's listed inside. The first and most important gate to learn: the NAND gate. Packs a lot of concept in one neat package.
You could also try some solderless, re-usable breadboards. These have arrays of holes connected together so that you can plug in parts like ICs, small & medium transistors, resistors etc. Also buy some insulated, solid core, tinned hookup wire, sidecutters and a craft knife (for removing the insulation). You'll be able to build and modify circuits quickly. For anyone starting out, I'd also recommend a multimeter. A decent one to get you started costs £8 in UK (Wilkinson's); Maplin, & probably Radio Shack (if you're in USA) have better offers.
I second that. Proponents of those breadboards trumpet how handy it is not to have to solder and desolder constantly everytime you want to make a change or correct an error, but that's a fallacious argument: after all (as mentioned), you wind up doing the same job twice anyway. Besides, those little wires have an annoying tendancy to pop out at the most inconvenient times, plus there's a greater potential for noise problems with those friction connections. Besides, really, how much trouble is it to desolder a few connections (I recommend solderwick vs. suction devices...admittedly, it's been a few years since I used a solder sucker, but when I did, I found the constant clogging annoying). Use a good iron and you won't find it troublesome. There are a dizzying number available, but I recommend a Weller WTCPT as a reliable workhorse. A huge selection of tips are available (tip it comes with is reasonable for many uses) and it's about $100; you can get by on less, but once you get into construction, you'll find cheaper irons aggravating to use (poor temp control, tip selection limited, etc.), although careful selection of one of Weller's economy irons can get you going for, say, $50.
Finally, a book you gotta have (it's not just a good idea, it's the law!): "The Art of Electronics" by Horowitz and Hill. Older editions are cheaper and still usable, newer editions have stuff you may not be getting into for a while, if ever; also consider an ARRL handbook (older editons also usable, newer ones available in CD, hardcover, and paperback), and, for that matter, ARRLs other publications (they lean away from hardcore theory and toward the practical).
I hope that the OP gets some benefit from your advice. And if he really wants a prototype board he should get one, but maybe he would like to consider this first:
I wanted such a prototype board when I was younger, and finally got hold of one, a big 4 section model. I built a special box for it, with power supplies inside and a front panel with instruments, pots and switches.
But I soon realized that I had to build everything twice, first on the protoboard, and then I had to move the working circuit over to pcb material.
My third or fourth project on that protoboard still occupies it, and it hasn't been used in at least 25 years.
It was much easier to build and experiment directly on pcb, so I could just put the pcb in a box and use it afterwards.
Get some pieces of pcb material, or copper laminate as we call it here. When you start a new project you take a suitable piece of it, and clean it. Rub it with an abrasive kitchen sponge, the green layer. That is the quickest and easiest method I know of to get down to clean copper. It is a lot easier to solder if the copper surface is clean.
Cut a number of straight grooves through the copper, with a sharp tool, so you get a 4 times 5 pads board, or the number of pads you think you will need. Solder the components between the copper pads.
Don't try to cram everything into a very small area. It will become a mess when/if you have to change the circuit. Leave space for changes from the beginning, so you can experiment and change the circuit as much as you like.
If you want to know more about building circuits in a fast and lazy way, look up these keywords on google:
manhattan dead bug style solder
and you will find 500 links to articles about fast construction methods without etching and without first building prototypes. You build the final circuit on a pcb directly. You can mix surface mount components with through-hole ones with a little ingenuity. Bend the wires of old components and mount them without making holes. (Making holes is a lot of work and you need good machinery. By avoiding making holes we avoid a lot of extra work.)
It doesn't matter if circuit boards are pretty or ugly. They will be contained inside a box and the user will only see the controls and connectors on the outside. That's where you should spend more time and care to get a nice-looking device. Spray paint and rub-on letters and symbols makes the outside more professional looking.
I often put small circuits in a tobacco tin, or any kind of metal box, with V-formed slits in the sides for cables into and out from the circuit, so the lid fixes the cables when shut.
By building circuits on small separate circuit boards you can use circuits later which you have built maybe just as an experiment. Just save all projects, whether they are completed or not. You can continue later, or put together a few experiments to something useful. Or change an old experiment a little to get a circuit you need. You cannot do that with circuits you have built on a protoboard, because all circuits you build there must be disassembled, destroyed, to make space for new circuits.
Well, there are other reasons not to use solderless breadboards. The National Semiconductor electronics guru Bob Pease doesn't like them because of the capacitance between different rows, which causes all sorts of problems, slowing down signals, interfering with measured capacitance values, causing oscillation, etc. They also tend to corrode with time, and thus the contact of parts to the little rows of metal get more resistive over time. Finally, the little springs inside that clamp on the wires wear out with use.
That being said, it's also about 100 times as fast to build and rip apart a project on one of these. You don't have to do too much planning, because the boards are set up so you just jumper wires around to make connections. Using these is made easier with 3M preformed wire jumpers, which come in various lengths.
The better ones I've seen have conductive backing plates, which, when connected to ground, allow a groundplane of sorts. I'm not sure if this is good or not. I have seen no problem building microcontroller circuits that switch signals at 20MHz on these. Any DC circuit, or audio circuits, are probably going to be OK as well, unless you end up with oscillations from your amplifiers.
Regarding the problem of moving the project to it's final soldered form, there are premade PCBs that can be obtained that have the same layout as these breadboards.
You can also get them at radioshack. By using these, you can just directly move the circuit over after prototyping, and have some assurance that at least the connections are in the right place.
You are talking about stripboard, also called veroboard. You can get this premade from the guys above (same link), with holes drilled in it. There is also a program floating around called "stripboard magic" that will plan the layout for you, given a schematic. It's fairly primitive, but works ok.
However, I haven't had good results with stripboard. I prefer just using the fiberglass board with predrilled holes but no solder pads. I find that the solder pads just get in the way, cause bridges when you don't want them, and fail to do so when you do. Perhaps it's pilot error.
The problem with this is trying to decipher what the HELL you did 6 months later, when a component gives it up, or you want to make a small change. It's generally used for things like RF, where you are afraid of stray capacitance. However, the inductance of the leads seems like it would be worse. Whatever, I don't know enough to judge.
I like that book, but I don't think it's a beginner's book. It's stated goal is to enable the design of scientific instruments by non-electrical engineers. It was written to be used as a course textbook at MIT.
I actually prefer "Practical Electronics for Inventors" by Sherz. It's more up to date, has much of the same information, and it's a bit cheaper. On the other hand, AoE goes into far more depth on subjects like precision design and high speed circuits. I have both, and I recommend the former for an introduction, the second for when you have a bit more experience.
There is also the Forrest Mims Circuit Scrapbook series, which is designed as a hands on approach to learning circuits.
Your library branch probably has all of these, and a few others.
No, they are so expensive and are not suitable for smd components.
I'm talking about copper laminated fiberglass, which is what is available nowadays, single or doublesided. But the older type of laminate is actually easier to work with, easier to cut into pieces. The fiberglass is a nasty material, but very strong and straight. The older pcb materials, usually brown or yellow, were often slightly bent, but a lot easier to cut. Buy a life supply of the older type at a surplus store for a few dollars if you get the chance.
If I want a ground plane I use doublesided laminate and use the other side as ground plane. Or build up the circuit with manhattan and dead bug methods, and use the single copper layer, undivided, as ground plane. I use the ground plane for all ground connections. All circuits are above the laminate then, on small platforms of small pieces of pcb material, glued to the laminate under it (manhattan style). Or use meg-ohm resistors to connect components to the ground plane mechanically, but not electrically. Turn the IC chips upside down, so they look like dead bugs lying on their backs with the legs in the air. (the dead bug style)
10 years ago I bought a collection of smd components from another guy, for 110 dollars. It was an amazing experience to hold thousands of transistors in my hand, it looked like black and silver volcanic sand.
Since then I am using mainly smd components but I often need to use a few old type components too. They can usually be fitted somehow without drilling holes.
On Mon, 28 Feb 2005 12:05:49 -0800 in sci.electronics.basics, Robert Monsen wrote msg :
Breadboards are very useful in a training environment, where small projects are built and discarded on a regular basis. I haven't used a breadboard since college. For Senior Projects, I used copper solder prototype boards, similar to what you describe below:
== Regarding the problem of moving the project to it's final soldered form, there are premade PCBs that can be obtained that have the same layout as these breadboards.
You can also get them at radioshack. By using these, you can just directly move the circuit over after prototyping, and have some assurance that at least the connections are in the right place. ==