Electronics projects. One thing about building electronics as a hobby is that people tend to get excited about it because it produces these cool electronic gadget toys. Cool, yes indeed it may be, but coolness aside, there are many practical factors that need to be considered. Over the course of my recent endeavors working with electronics hobby projects, I have found it necessary to summarize some important lessons that I learned, which are certain to be valuable for anyone else to know in advance should they consider picking up such a hobby.
First Question: What is the purpose of your electronics hobby?
First of all, answer this question. What is the purpose you are doing this electronics project for? Many simple electronics projects don’t strictly need to be done with a physical electronics project hardware element. Like any hobby project that requires buying raw materials, the associated costs can quickly spiral out of control if you just want to buy anything that’s fun to play with. But, furthermore, you may find out that electronics is not as fun as you thought it would be if you just buy and build anything you can think of with electronics.
So, it is crucial that before beginning, you present to yourself and answer the question up front: Why can’t this be done purely with software alone? Why is a electronics hardware element actually required?
Often times, as I’ve found out, the answer to this can often times be phrased in a quite clear and concise manner. The reason why you must have an electronics hardware gadget is because you must have an additional means to interact with the physical world that existing computer equipment on the commercial mass market does not provide. For example, the following points are some of my legitimate justifications for myself to get involved in electronics hobby projects:
-
You have an existing computer that works just fine on its own, except that you want to make some minor improvements or repairs to it. The commercial market fails to offer a product that suits your needs because (1) they only offer totally newly manufactured computers for sale, none of which suit your tastes for computer hardware, (2) although they do offer repair services, they are too expensive for you to personally afford, (3) for your own convenience, you want an “extra” working computer at no additional cost, or (4) you need to perform an advanced computer repair which is not commercially available in your country since it requires manufacturing-style skills that are only prevalent in Asia.
-
You have a “legacy computer” with specialized computer I/O interfaces that you must have access to, and no current hardware is in commercial production that supplies those interfaces. Therefore, you must be able to perform your own advanced electronics maintenance and repair on the equipment you own in order to keep it running to satsify your needs.
-
You already have all the electronics hardware parts that you need to build a 3D printer device that you want from old 2D scanners and 2D printers, but the commercial market does not provide you with market options to directly recycle that old equipment into your specifically desired new products. You may also want to make tweaks to your 3D printer so that it can also double as a “robotic pick and place” machine for surface-mount components.
-
You need additional sensors to gather data and make it available on your computer network, and your sensor needs are so specialized and specific that there is not suitable commercial device that satisfies your needs.
-
You need to make improvements to existing commercial sensor designs, i.e. 3D scanners, but since they are proprietary, you must more wholly build your own device instead.
-
For the sake of your specific computer uses, you want to better control the overcomplexity of computers compared to the commercial market’s approach.
-
You have worked with commercial computer hardware maintenance and repair in the past, and you were upset with the lackluster effort put into modularity, repairability, and the ability to repurpose the hardware to future uses. Therefore, you need something more custom than what the commercial market has to offer.
-
You are anticipating wanting to perform specialized and advanced operations on computer hardware, and you therefore need to have Open Hardware to work with. In the event it doesn’t exist on the commercial market, you need to build your own.
Nowhere included in here is the idea of getting involved in electronics hobby projects just for the sake of having fun and playing with electronics toys. Of course not. “Hardware is hard, software is easy.” The day-to-day reality of working with electronics projects is far from fun when you consider the equipment costs involved. The fixed startup costs for electronics projects are quite high compared to the fixed startup costs for osftware projects. You need to invest in both tools and materials before you can even get to your day one electronics prototyping breadboard where you simply start out testing circuits, creating test circuits, and almost just generally playing around seemingly without working toward a real project goal.
But, here’s the key. If you do have a clear answer to the first question and can clearly define a purpose as to why you are getting involved in electronics hobby projects, you’ll be able to tackle the associated costs much more diligently. And, ultimately, minimize them as much as possible. Buying individual electronics components not only has increased costs in terms of per component costs compared to their cost when embedded in a commercial mass market product, but it also has additional complexities and responsibilities when it comes to organizing hundreds of different types parts that you’ll need to have on hand, each with very specific requirements. If you know what goal you are working toward, at the end of the year, you’ll be able to tally up all of the time and resources you’ve spent on the “play” aspects of electronics hobby projects and make it count toward a specific final product that you’ll be able to use and see the value in for a long time to come.
Let’s take another example to run this thought process through its paces.
-
Temptation: Hey, it’d be nice to create a cool electronics gadget that plays different sounds when you push different buttons.
-
Enlightenment: Well, if all I want to do is create some toy that makes sounds when I myself push buttons, I can do that in software. Just create a GUI with push buttons and use the operating system routines or a library to play sounds. If I need it hand-held, I can run the software on a smartphone. After I’m done playing with it, the cleanup is easy with software if it turns out to be an idea that doesn’t really scale beyond my own entertainment. But, were I do to it with hardware, I could just as easily end up with some hardware toy that just collects dust once I am no longer interested in playing with it. And if I kept that habit up, I’d end up with a much bigger problem at the end than the alternative would present me.
-
Problem: I need to use a Serial In, Parallel Out (SIPO) shift register inside a large and complex electronics project, how am I ever going to get it to work correctly? When the large and complex circuit as a whole isn’t working correctly, how am I going to know what’s going wrong?
-
Solution: This is where the value of play can pay dividends, if done as a sub-component of a larger project at steak. Rather than focusing so heavily on building only the final project in an instantly finished form, you can break down the development process into sub-steps. Hey, why don’t I try creating a simple debug board with tactile switches so that I can control the shift register directly with my own fingers? Not only will I be able to verify that the shift register is working, but when I wire it up into a more complicated circuit, I’ll be able to reuse by device to test that the larger circuit is also working correctly by probing on my debug device onto the more distance wire connections.
The laborious journey through learning electronics
And once, and only if, do you have the key premise question answered, then you can proceed to setup elwctronics prototyping. It will take you longer because “no one” has done this kind of stuff in like 50 years, so chances are you will not have a working memory, heck you may have never had any experience with electronics in your past life. And neither will any of your teachers, parents, family, or friends know how it works either.
So, you will have to embark on a journey that seems unnecessarily complicated. Surely there are professionals who are skilled in all of this? Well, there were professionals that were skilled in all of this, of course they all moved to Software Land since that’s where all the paying jobs are. I mean, yeah they have the skills, but they know they won’t get paid to use them, so why bother put them up on a pedestal and advertise and market that they have those skills?
Nevertheless, although good help will be hard to find, you will find enough hints of a trail of breadcrumbs by web searching via the Internet. Slowly you will learn little by little in a random and chaotic order guided by carefully learning and taking notes on whatever random information comes up to you on the Internet. At this point in the process, you must be careful to document your research and thoughtful about your learnings and findings. There will be many times when you are told true facts about practical circuit design but you won’t be able to comprehend the big picture. Then, only after deeper thought will the theory behind the facts become obvious. You can look at some of my own blog notes on motors, motor protective circuits, inductors, and capacitors for reference, for example.
Hardware is hard, software is easy
Construction complexity and cost equations will loom imminently before you. It’s true that surface-mount devices are the commercial mass market industry standard in electronics, but they are quite challenging for novices to work with, and even in general, they are markedly more difficult and expensive to prototype with, even though the raw cost of these devices are cheaper than their through hole equivalents. Through hole devices, by virtue of being larger at 1/10 of an inch pin spacing, are easier to work with in general, and especially easier for novices to prototype with, but because the market is so small for them, there is a lesser diversity of through hole devices available than surface mount, not to mention they’re more expensive. The increased complexity of searching for through hole devices and the increased cost alone, in addition to the increased cost of buying in small volumes, is enough to scare all but the most essential electronics hobby project ideas away.
But, here’s a tip: as I’ve noted previously, there are adapters you can use to mount surface-mount components to through-hole perfboards, but the most common type require surface-mount soldering, and although that’s something you can do with your home kitchen’s oven, the fact that these are separate discussion points is enough to scare novices away from using surface-mount components in their very first major electronics project.
And, that’s not all. Sure, you may have guided your strategy in light of these realizations. Due to the challenge of finding suitable parts, I must make that be one of the first steps in my design process. First I will just vaugely sketch out and write out brainstorm ideas of the functional requirements, then with my vague plan of what discrete and integrated components are required, I will search for those and build a bill of materials, maximizing the number of through hole components I can get and buying adapters for the parts I cannot get in through-hole form factor. Once I have my bill of materials, I can determine how much board area I need by drawing out the design with accurate part footprints in some electronics design automation (EDA) software like KiCAD. Then I’ll be able to buy a perfboard of the required side.
So, finally, you’re at the final step of the process. Your designs are all drafted out checked over three times, you have a clear bill of materials, and you’re ready to place your purchase. But then, reality strikes. Hardware is hard, software is easy. Just as you are placing your order, you notice that some of your previously in-stock parts are not out of stock and on backorder!
And that’s still just scratching the surface of all that you can do with electronics and what you actually wanted to do. You probably actually wanted to reuse some existing electronics parts you’ve had, but to your disappointment, they were all surface-mount and therefore above your novice electronics skill level. Or maybe they were missing some modular connectors your design called for and you didn’t want to start clipping wires so early. Or, maybe you did actually find some parts that you wanted, but not enough, not enough of the same type to use together in one single logical project. There are just so many more variables that although you can account for them when working with hardware, many more of them are beyond your control and you are powerless to do anything about compared to software.
Any bill of materials will be harder than any software dependency problem. No matter how unnecessarily difficult you think the software dependencies problem is, I can assure you, the bill of materials problem with electronics hardware will always be harder.
Why? Because with an electronics bill of materials, not only do you have to check for compatibility of the components per se and whether they are still supported by the manufacturer, but you must also check for the quantity in stock, manufacturer lead time, and the quantity that you are ordering in comparison. There will be shortages some of the time, and you will be forced to wait for new factory manufacturing and additional freight-class shipping delays to complete.
Some of my big lessons I have learned working with electronics
The challenge will always continue when working with electronics projects, far more than is the case with software projects. Over the course of working with electronics projects, I’ve learned a few notable lessons that are worth summarizing here.
3D scanners and printers: Stepper motors DC motors? Stepper motors are quite popular here, but surely they are not necessary are they? Old 2D printers and scanners previously used stepper motors, but the newer ones have since done away with them for cheaper, faster, and more powerful DC motors. And they do produce pretty accurate results, don’t they?
Not so fast. For one particular project, I was printing measurement grids with one of those modern 2D printers with a DC motor, and I noticed the lines were all messed up. Looking at it carefully, surely it was caused due to issues with the position tracking of the DC motor! Awww, dang, indeed DC motors are not sufficiently dimensionally accurate for even 2D printing and scanning, so how do you ever expect those kinds of motors to perform when you need dimensionally accurate 3D printed electronics cases?
Second lesson learned: Designing a page counting device for photographic paper scanning without skipping. It sounds like a simple target application, but what kind of actual sensors would you use to solve this particular problem? Weighing is one idea that comes to mind, but that is subject to noise, not to mention I originally wasn’t even sure if there was a sufficiently accurate weight sensor to perform my job. So for a moment, I entertained the idea of height measurement, but even that can be uncertain with papers as they tend to have curl that causes them to stick up unless forced down flat with sufficient weight. Finally, after turning around in many circles, I was able to find a suitable weight sensor that could give me the accuracy I needed, and it allowed mounting inside a custom enclosure so that I could build a page-sized weighing platform. Previously another problme was that the high precision scales only had a small weighing platform and were therefore unsuitable for weighing papers.
However, designing the platform was yet another problem to be solved. What material do I build it out of? How do I mount it to the sensor? One of the main things to beware of when designing weighing platforms is deflection: if the platform itself bends when under load, the force causing the platform to bend is absorbed into potential energy in the platform itself, like stretching or compressing a spring. This also means that energy is not transferred through to the other side of the platform, so this causes your weight measurement to sag in comparison to the actual weight. If you want an ideal weighing platform, you want one that remains relatively rigid and stiff when the load is placed on top of it. This means you can’t use a thin pressed cardboard platform, but a corrugated cardboard platform may be adequate.
Likewise, thin aluminum and copper sheets may be too flexible, but is there a good metal you can use for the platform? Steel may work well. Brass is also very likely to work well… it is the stiff metal used in pin headers that are designed not to bend very much, unlike the copper wires you attach.
Conclusion
As you can see, electronics hobby projects are not for the faint of heart. The journey will continue for a long time, and the challenges are many.