This was supposed to be the first part of an electronics build series - it still is, but it's different from what I intended, because it's really about stopping, rather than starting...
*Afterthoughts & Addenda* *FAQ* is below the fold in this pinned comment... *Your Critique* - I am of course very interested to hear your civil and constructive critique on all of this - after all, I'm not even really sure I have considered all of the options. Please be aware that if you post a comment containing a link (for example a product you think I might use for this), you should allow a little time for the link to make it through the comment moderation process (links are automatically quarantined on this channel due to the high level of scam/spams around) *FAQ:* *Why didn't you just use a ready-made keyboard?* - the keyboard needs to be wired USB and less than 200mm wide. There aren't a whole lot of ready made keyboards that fit this two simple criteria. *Why not design and make a custom PCB?* - great idea, but I would have to develop the skills necessary to do that, which is quite a journey for a one-off prototype - not that I would really mind having those skills in my arsenal for other projects* *Why not just do it [some other way]?* - you may be missing the point - this was supposed to be an exploration of a really basic, no-programming, low-skill-required potential method of making a custom keyboard. It failed, but *this* was something I wanted to explore. Doing it a different way would not be *this.*
Shrimp, you are a big poopy head. Oh... " _civil and constructive_ "... Ok. I'm an Electronics Engineer / "Embedded Systems Engineer" and over the last 30 odd years I've had your little happy fun time not quite successful project experience a great many times. All I can say is, sometimes it's a fustercluck. Don't be deterred, you uncovered numerous pitfalls and if you have another crack at it you'll know plenty of things to watch out for.
Could you save yourself the last difficult/fiddly task if you used Veroboard with tracks one horizontal and one vertical. Would the key pins be long enough to go through two layers?
Here's just a few bits of advice; take it for what its worth: 1, You might want to have scanned the original keyboard sheets. In Photoshop (or gimp or whatever) you can just color bucket the individual lines. Makes everything very, very much simpler. You could even just take a picture of the original keyboard and have 3 layers. top sheet, bottom sheet and scale the keyboard picture to the right size. I'm guessing it would take half an hour to map the whole keyboard. 2, Try making your own PCB, and by making your own, I mean sending the files to a PCB maker. I've used GEDA successfully a few times and its surprisingly easy. Then all you need is the matrix from step 1. You can put the key switches wherever you want and it will auto-trace all the paths. (GEDA is FOSS). 3, Y no flux? Flux gud. Seriously, liberally using flux makes soldering so much easier, and that rosin core stuff just doesn't work well. 4, unless you bought very expensive switches, test them all first D: But really, I think people are intimidated by having their own boards produced. Don't be. Start with something simple like a USB power supply with a reference design and work your way up in size and complexity over a couple of projects. I've made things like what you did, jerry-rigging solder breadboards (the most complex thing I did was a well depth sensor machine that worked off pressure sensors) and its just awful compared to having boards made. The "I did that" feeling is a lot stronger when you receive a professionally done, professionally cut board than when you solder together a rat's nest (at least for me). IF... IFFFF you insist on using a board like that, I've done it two ways successfully: one, use enameled magnet wire kinda like you did, but scrape the enamel off areas that need it with a razor blade. The other way is just use wire from a cat-5 cable and strip it at the locations you need to (the plastic will slide for at least 30cm or so). As I was writing this while watching the video, bend the switch pins, not only to hold them in place, but bend them towards the center of the switch so you can tell which switch they belong to. B̶e̶w̶a̶r̶e̶ ̶o̶f̶ ̶f̶e̶a̶t̶u̶r̶e̶ ̶c̶r̶e̶e̶p̶.̶ ̶F̶e̶a̶t̶u̶r̶e̶ ̶c̶r̶e̶e̶p̶ ̶i̶s̶ ̶t̶h̶e̶ ̶p̶r̶o̶j̶e̶c̶t̶ ̶k̶i̶l̶l̶e̶r̶.̶ -edit Oh, also, don't be afraid to add more features. YOUR PROJECT ALWAYS NEEDS MORE FEATURES.
@@AtomicShrimp no problem. Condensing a lifetime of experience down into a few points means leaving almost everything out. I'm not good at that and I tend to leave a lot of TL;DRs.
I absolutely and wholeheartedly second having boards made. There's some growing pains for sure, but it's so amazing to assemble a design that does exactly what you want rather than fight with a protoboard... Especially for something this conplex
@@DingleFlop It seems like the adhesion of the copper on the kit boards is a fraction of what it is on manufactured boards too. I never had a problem with pads coming off or anything. I think a lot of kit boards sit around for years or decades and really arent any good anymore by the time you get them, if they ever were good.
In addition to your usual (successful) videos, I always appreciate your "fail" videos Mike. Most youtubers just bin the footage... you are proud to show it without shame. The video is just shy of 40 minutes long so it is clearly not clickbait. Some people just have the short clips when things go wrong or not to plan, but you have kept your thought process in etc. and the result is some great discussion in the comments... as we have all done projects - be it electrical, DIY or crafts etc - which didn't go as planned. A few people are sharing their experiences of when they did similar in the past or giving suggestions. A great community vibe. Happy New Year!!!
Mike, this is an excellent example of “just one more hurdle” leading you down a whole road of Wrong. It’s a great lesson: there are things we do ourselves, and things that don’t pan out for us when we try. Don’t roll your own ad-hoc time/date library, don’t roll your own keyboard with individual microswitches and a breadboard.
I don't think there's anything wrong with building your own keyboard from individual switches; there is no suitable ready-made keyboard, so I have little choice. The error is choosing specifics of approach (the harvested controller) that seem simple, but are not, and fighting setbacks with solutions that will cause further setbacks.
Definitely feels like a better way would just to be grabbing a microcontroller, programming it to act as a keyboard, and wiring up your own keyboard matrix to it. There's a couple open sourced keyboard designs based on the Raspbery Pi Pico and RP2040 chips like the PyKey and Env-KB, I've used them to make macro keypads but I haven't done a board larger than 16 keys.
I agree. I guess it's just knowledge of the cheap microcontroller dev board market that would've sent me straight to one of the stm32 "pill" boards, which are easily built upon in qmk. Once the matrix was wired up, it's very simple to get your own variant of the firmware going. I know that because I did my own as a fun project with my own pcb and 3d printed case. Would've even been about the same price as that cheap keyboard the controller board came from 😅 Totally not saying it had to be done my way, just that there's so many ways to do these things 😊
one thing all those tutorials have in common is using if conditions to check in a loop whether a key is pressed, do you have any resources where I can simultaneously press multiple keys at once?
My recommendations when it comes to soldering: 1)355 is too cold even for leaded solder add 10 degrees on to your temp (this will increase the speed your joint forms without melting your switch) 2)Flux! Hearing you say the line "there wont be much smoke as all the flux is gone" while attempting to join enamelled wire onto a pad was an absolute facepalm moment.
Yeah when he said that I had to laugh. If you're not using flux then you're not soldering either. Maybe if both conductors are tinned. Then maybe. Even then no flux isn't ideal. But you stand a chance of sticking stuff together.
When your job is being a pleasant and entertaining bloke on the internet, even a failure is a success! Happy New Year, Mr. Shrimp, thanks for all the pleasant entertainment and best wishes for you and the fam in 2022.
I'm not a hardware person, but when you got to that enameled wire, "Is he trying to drive himself mad?" is all I could say. Looking forward to a less maddening version. ☺
Thank you for releasing such an in-depth non-starter. Too much of the DIY video space is filled with what appear to be quick and easy projects that don't go over the reality of how many attempts and failures will actually happen in real custom projects.
I kinda did the exact opposite with my keyboard. I bought a cheap board with switches and a layout I liked, and replaced the controller board with a microcontroller running QMK. I finished the thing, but it also wasn't smooth sailing. For one thing, I had to modify the matrix slightly to cut down the number of pins, which meant cutting some traces and a few bodge wires. Additionally I wanted N key rollover, so I had to cut all the traces for the columns and replace the traces with diodes. This of course added quite a lot of stuff to the backside of the PCB (and the teensyUSB also took up more space than the original controller), so I had to shave down some plastic from the inside of the case. I got lucky in that I was still able to close the case, considering I of course hadn't measured anything before starting the project.
It wasn't the possible contact confusion that gave me pause for thought, but I was concerned about the durability of the design, given that it was meant to be a portable device. Would those tiny and wire-glued connections to the PCB have held up over time? Would the enameled wire have shorted somewhere after a while? I've been known to build things that tested OK at a bench, and then failed swiftly in service, so I may be over-sensitive! Anyway, RU-vid is full of projects that gloss over the difficulties, and many would not have uploaded a Setback Video, so thanks for the insights.
Hi Shrimp, as an aspiring maker myself, I find your enthusiasm for "garden-shed" tinkering inspirational and I loved this video. Doesn't matter that it wasn't successful as it was an interesting journey and I learned one or two things along the way
I remember a particular keyboard I had back in the late 80's or early 90's. It didn't have a bad feel to it which was surprising when I finally opened it up and saw what was inside. It worked fine but then over time some of the keys became less responsive or didn't work at all. Since I had nothing to lose I opened it up. What I found was far more low-tech than I ever imagined. The PCB was printed the same as the PCB in a typical TV remote. On the working end of the key shaft was a pad of spongy, grey foam that had a square of shiny, silver Mylar looking material on the bottom. That was all there was to it. Pressing the key and compressing the pad shorted the contacts. That aspect I found interesting because the pads were in constant contact with the contacts but only shorted when additional pressure was applied. Then, of course, as dirt and hair and other things fell inside they would get between the contacts and the pad but it was a simple wipe-out with a cloth and it was back to working fine again. Today I use a nice mechanical action gaming key-board and even though I can see it's full of dirt and I've worn many of the letters off it still keeps going. Now, my mouse, on the other hand . . .
Truth was, making a keyboard with any kind of actual discrete switch per key all those years ago would include a noticeable cost per switch/key - having even 80 keys at $0.10 before you got anywhere else would mean their budget keyboard was a $8 item even before anything else was added, let alone assembled, so there were several innovative choices often involving a lot of bits of moulded rubber/plastic/sponge/stamped metal/whatever that could be churned out - and that's better than another decade or two before when all those 'cheap' switches would start you at more like $40.
We learn by experience, and sometime by the experience of others. You've done us a service by actually posting this video so that we might learn. Thank you very much!
You lost me at raspberry pie! Truly my opinion is of absolutely no relevance. I’ve no clue what you were talking about but watched to the end! I’d bow to your decision, if you can’t make it work well! Your excitement, interest and annoyance at the keyboard was incredible. Thank you. Happy New Year to you.
Despite having dabbled in building my own not-so-good-keyboard, this was quite interesting! That might be, because your approach was the direct inverse to mine: I had a broken laptop and decided to hook up a teensy to the 25 pin ribbon cable of its keyboard. The programming ate my weekend for breakfast, but overall I'm quite happy. Seeing your struggles, it seems I have made the right decision.
@@theodiscusgaming3909 Depending on the kind of ribbon cable you can also just cut/pull the strand cables it made out of apart and thenn it's simple wire to wire soldering and no issue at all. Just make sure to not forget the heat shrink. I have done this dozens if not hundreds of times already.
I have done things like that before,.. Some things that really help me in these situations: Firstly, For the switches, you may bend the pins in the direction of the respective switch. This tells you what switch the pin goes to as well as keeping them from falling out. Secondly, Use colored marker on the pins to denote which trace is which. This will help you keep track of things as you solder the wires on and it will go much faster. Thirdly, Switch to PVC jacketed #30 wire. The PVC can be easily pulled apart where you want to expose the wire and it is usually pre-tinned. The magnet wire is a good idea, but it can be problematic trying to burn off the enamel efficiently. Also, you have 10 colors of jacket to choose from which will also help keep your matrix oriented and coordinated with the established marker colors. Maybe cut out the section of the perf board that has the KBD connection so you don't have to resolder the connections on the keyboard module? Heat gun may help you pull the switches off the old perf quickly so you may be able to reuse them. Use additional flux that can be applied via syringe for faster and better solder bond. Wrap the wire around the pin a couple of times to help keep it in place. Your endeavor has a few setbacks, but all is not lost! Edit: Make sure all the connections on the keyboard module are functioning then pot the module in epoxy for durability?
Been there. Done that. Several times. Totally different projects, totally different process in train. Just part of the game if you want to be a "maker". The ones that worked well far outnumbered the ones that did not. :)
@@perdiddlepanskew8018 I didn't make it up. The theory is if you're not failing a significant amount of time then you're leaving potential unrealized. Because you're just not finding what your true capability is.
if you do a solder job like this again. I find it helpful to use a no wash flux pen. Even though the solder has flux in it, I always find it helpful to add extra flux.
"Every bursted bubble has a glory! Each abysmal failure makes a point! Every glowing path that goes astray, Shows you how to find a better way." I enjoyed the video! Also, I appreciate that you shared with us your failed attempt. As the quoted song says, " From the ashes of disaster grow the roses of success!"
I don't know why but I found this video and particularly the ending really quite humorous. I can only imagine your frustration when you realised what had happened and that you had to come clean to 823,000 people. Brilliant!
Normally I would be highly critical of the stupidity involved, but in this case, I fully appreciate the effort you put in and also the ability to admit it was an exercise in futility and give up. You have come full circle from smart to stupid and back to smart again. Looking forward to the next episodes! (I watched this 40m one in about 20 minutes)
I did this reverse engineering of a keyboard once. I simply scanned the plastic sheets and colour coded each trace in image processing software. Much quicker and you don't have to keep following the traces by hand meticulously.
I wish I had had teachers like you when I was at school for the subjects I wasn't naturally interested in ❤ I know next to nothing about techie stuff but I love watching your videos of this type of thing, I'm always inspired by people who are methodical and inventive. You make a very good point about knowing when to stop a project, that anticipated buzz of success against all the odds often (for me anyway) overrides my ability to take a step back and assess the situation. I'm looking forward to seeing how this project of yours unfolds.
I am so thankful that you have posted this. My projects never go to plan and I can never cover all my bases with certainty. It is an important lesson, knowing when to quit. I hope you figure out a way to my it work eventually, you certainly have the patience for it!
Thanks for sharing. What differentiates the good from the bad here is that you edited the video to be very clear in the first few minutes that the project turned not out as it should have. When I wanted to start my career "in computing" about 20 years ago, I talked to some people trying to convince me to start with electronics. Which seemed fair, so I did that and things got very confusing. I had higher expectations than most back then, but frankly everyone was very bad at project management, meeting customer requirements and generally sub-par teaching students. Which of course, could not be... so I was at fault. May be that is the reason why some even struggle today in IT environments. The project seemed like a good idea. Effective cost cutting. It started to get confusing when you had 3 to 4 PCBs on the desk, which had so much potential and you wanted to make best use of all of that. And more. Smaller, higher integrated components would have been more expensive (or not making use of the components at hand) but the right way to go before one gets lost in all intricate details and what could have been. Repair Cafés are also a really nice idea, when you keep the focus on things that can be repaired. Thanks. This was rather painful to watch despite best efforts, but it should be shared with everyone underestimating how projects can fail and where to stop to avoid the sunken cost fallacy (while trying to cut costs). 4 PCBs and a rubber dome keyboard. I would have stopped there. Not because of laziness, but because of fear what the project may become. "Laser engraving the keys, charging other devices, doing some computing on the device itself." That made me feel very excited. All the cool technology we could have at our hands (or in a maker space). But then slowly realizing that it has to stop and you have to set priorities.
This video is a good example of a design process in general. I think there's a common misconception that highly skilled programmers, engineers, whatever, don't make many mistakes, when the reality is they just know how to identify and address their mistakes better than a novice. Failure is the best teacher, so they say.
I did the same thing around 2006, to build myself an emulation controller for an emulation machine made out of an old cash register computer into an old carry case; when I was trapped on a ship 100's of miles out to sea(USNAVY). I did the same thing, by recording the matrix, but I just drew on the clear plastic with a permanent marker I wrote the name of the key down, and on the PCB (0123 for rows, ABCD for columns etc). Really cool vid sir!
Thanks for sharing. This is a really common problem at last when it comes to programming - it’s always a hard decision to say STOP when the project goes sideways and on the other hand you invested so much time in it :)
Ok. So. I didn't sub to this channel for RaspberryPi content... But I got exactly what I needed right now. Love this video. Thank you for showcasing components that I didn't know existed. This video is essential for my own Pi laptop build.
Ah Shrimp, although this vid was the biggest pile o gobbledegook to the lay person, I watched every second of it with the gusto of a heart surgeon watching a heart bypass, and I found it utterly fascinating though why I will never know. Happy New year Mike x
I've next to no technical aptitude for building stuff in this fashion, but I think the most admirable points of the video are two things: First, admitting the failure. That's one of the hardest things to do - saying, "well, that didn't go the way it was supposed to," - particularly when doing so exposes you to ridicule (and you had no need to actually admit you made a mistake). Second, realizing the bad path you were on and not falling into the "sunk cost" hole. Always a good lesson. Be cognizant of the effort you're investing in a project, and be clear-headed about what furthering that project will take at any given point. Good show.
I'll never knock someone for trying something. You never know until you go. He could have pressed on and finished the project. But he knew it wouldn't have been what he wanted. That would have been a greater failure. Better to take a little temporary l than a big L that hangs around.
Good call man- I have LEARNED to do that a lot more in my line of work. I get to a point where I realize it is going to take longer to fix something than to just start over on the path I know I would prefer to take.
I'll take this as a reminder to mark your board when it's big and same-y and confusing. I probably wouldn't take this route and would be more likely to scavenge a keyboard from one of those super cheap ipad knockoffs you find in discount stores. They usually have a keyboard built into the case.
I am very new to handy work, especially of this variety, so I appreciate the way you explain as you go. I anticipate enjoying the next video about this project's process of progress!
I honestly appreciate it when companies give the free gift no matter what kind of review you leave. I suppose they get a boost whether you leave a good or bad review, like disliking on RU-vid. But I've left a not good review on a camera I got and they didn't care and sent me the accessories I asked for as my "free gift" regardless.
In Slovak we say _na chybách sa človek učí,_ meaning essentially that we learn by making mistakes (including the mistakes made by others). So this was definitely a very educational video. Thanks for it.
Thank you for uploading this, I plan on a similar project sometime in the future and seeing the problems you faced during this will be very helpful for me to avoid them.
I was surprised to learn that the heat of a soldering iron could remove the insulation from copper wire - it never has in my time. So, my shock was limited when it didn't work :-) Bravo for posting this !
It depends on the wire. There's wire made for motors and transformers that requires excessive heat to remove the insulation, and then there's the (far more expensive) wire with self-fluxing polyurethane coating, for example Verowire.
I actually found that attempting to tin directly (solder on the edge of the iron, rubbed gently on the wire) is enough in some cases. You can tell when it's tinned because the solder will "wet" the wire, rather than pool around it while stuck to the iron.
I have no understanding at all of what your doing but I loved listening to you doing it! I normally watch you budget food videos but I was hooked watching this!
What I'm going to take from this is: before starting a project, you should consider all the required steps and then assume it'll have unforeseen issues and then decide if you're interested. And that it's nice when a project is such that half-finished would still be worth the effort
I'm not much of a fan of country music in general but sometimes the classic songs have some good lessons, and one that's always stuck with me is these iconic lines from "The Gambler" by Kenny Rogers: You've got to know when to hold em, Know when to fold em, Know when to walk away, And know when to run. It applies to a lot of things in life... including complicated keyboard builds that are driving you mad.
I have a jukebox project sitting in my garage. t runs off a Raspberry Pi. Since I wired the buttons to a USB joystick controller, the number of keys that it uses is limited. If I need a full keyboard, I can plug one in. I've learned more from failure than success. People are less likely to analyse success than failure and I am no different. I am glad that you took the time to analyse yours. It now would me think twice before starting something like this. Hopefully, I can pick up the jkebox project next summer when I can work on it in a warm garage with an open door for airflow. I hope that this setback does not dissuade you going forward, Have a good new year
The Raspberry Pi is a surprisingly functional beast and the ready made keyboard model made me sit and take notice. Anyone can deal with straightforward issues. I spent ages trying to get an old Mac Book to run a Linux OS only to put it on to the ‘back burner’ for a time when my knowledge is sufficient. This has been a truly instructive video. Thanks for your honesty.
When soldering up, use plenty of flux, solder and wash the tabs with solder then take off with desolder braid to leave nicely tinned. This makes it easier to then solder with tidy solder contacts. You should invest in one of those solder tip cleaning wells that contains a metal copper wool like the metal pan scourers. They help clean off the tip of your iron better than a wet sponge. You benefit from an iron that maintains its tip temperature and not being cooled and contaminated by what is in the water you wet the sponge with.
i think these videos are really important and ultimately kind of end up more meaningful than if everything went well? idk, thank u for uploading this regardless was a nice way to spend some time 💛
Hahaha, yeah... as someone that has done a very similar mini keyboard project, on an even smaller scale. it is a bloody nightmare. Respect for stopping midway. Wish i had, as now i have a very lacklustre mini keyboard that i just dont really like using lol
For anyone wondering, those beautiful blue keycaps in the background are MT3 Dasher! I personally use the same profile Susuwatari set. Absolutely delicious to type on
I don't know how many times I've walked away from a project I started because I realized too late I screwed up...bad. It's a learning experience, for sure.
I had fun watching. Anyone who has ever tried to DIY anything failed at certain steps. Researching every detail for days isn't pleasant neither, and DIY should be pleasant. It's not like we're preparing a project with mass production in mind. And I too had my share of grief with those switches. I ended up actually removing 2 of the 4 pins for easier identification and short prevention (in a side by side pattern). Great project and I'm looking forward for more adventures!
This sort of projects are of course cost+time worthless, but, on the other hand, it actually teaches an individual the lots of engineering and manufacturing compromises a comercial finished product goes thru (even a cheap one); it's a valuable lesson for everyone interested in, and you can always have feedback from other users to re-do many steps in a more time-efficient manner; and it's also a good thing to re-use electronics and spare parts in your new projects (I have had an apple wireless keboard that had coffee spilled in, so I could salvage the logic board and also will try to re-implement a keyboard with it)
Hello! Electrical engineer here. These sorts of problems happen all the time, once you start tessellating a certain feature over and over, it gets really easy to mix up. If I can make a recommendation for the future, stripboard is much easier to prototype with when you have long busses like this, and the pads tend to be much more robust. I like your style with this though, and am 100% here for a followup. Cheers!
I suuuper enjoy when people post their failures, rather than their successes. The reality is that all of us will see many more failures in our lives than successes, so as you point out here, knowing when to stop, how to learn lessons from it, etc. can sometimes be far more important and interesting than "hey I did a thing but you don't know the millions of failures it took to get here!"
Your patience and endurance combined with keeping a good mood are admirable. I came to your channel because of your fantastic scam fun but since i'm also someone that soldered this kind of stuff (e.g. a EEG brainwave scanner +- eye-led glasses feedback loop to generate a meditative state machine), it's of great comfort to see this video . I'm well familiar with the Sunk-Cost problem :-)
Well done on the video. Far too many, here's one I did earlier types. Quite rare to see one where the mistakes and pitfalls are honestly and genuinely pointed out, and where people show how they got it wrong. Learn by your mistakes, or even better, learn from someone else who has made them mistakes and is honest and genuine enough to show you how NOT to do it. 👍👍👍👍👍👍👍👍👍👍👍👍
Been there. There is something wonderful about pressing on through difficulties in a project. That being said, it's a valuable skill to know when dogged persistence is actually harming the project. Kudos!
Fascinating video, thank you for posting this, what a great life lesson. I found myself so invested in this keyboard that it was hard to watch you abandoning it, even though I was only a spectator! I know nothing about building computers but I love watching whatever you do and listen to your reasonings and explanations 😊
I was working on a project lately which needed keyboard with LEDs on each key, and first i thought that i just could do it by hand for a prototype (it also used keys and leds that are connected to SPI IO expander to save on MCU pins), but then i realized that each button/led potentially could have it's own pullup resistor, load resistor (i haven't decided at that phase which IO expander chip to use - the one that had true tri state ports with pullups or which have it bare bones) etc., and at that point i suddenly realized that this is where i should use PCB manufacturing service. Now, granted, i've already worked with this service once, so i'm kinda familiar with PCB ordering process and their online layout editor, but it's still relatively easy to pick up from scratch. They also usually streamline whole process for customer, so even while i've spent couple of days routing all boards that i needed for the project - in time that i could already done the prototype by hand on a protoboard, i was really happy that i did it via PCB fabrication - it was far less tedious to assemble the project once boards arrived, without any kind of rat's nest of a wire mess (at least on the board side) and without risk of killing the joy of working on a project by having a loose wire somewhere under that mess and hours of debugging because of that. And you may even incorporate your approach of reusing cheap keyboard controller PCB on a board of your own design. tldr - you probably should try PCB fabrication service, it's cool. Some services even allow small batches (~5 pcs) if you don't plan to have lots of them.
I absolutely encourage you to get more and more into DIY electronics. I’ve enjoyed your videos for a long time but I’m a hardware guy only. I can copy and paste a bit of code if somebody’s problem gets wiped but I’m scared to death the whole time I’m doing it. However I have no problem repairing and calibrating very high end metrology great equipment. That’s my whole business these days. Started my own lab five or six years ago completely by accident because I got shocked really bad and realized it was time I learn some things about electronics if I’m going to keep building vintage reproduction guitar amps. And restoring CNC machines and stuff like that. I was shocked so bad it knocked me up against my wife’s car and left a huge dent in the door. 😂 I had no idea capacitors could even store a charge! And when you’re working on a massive tow behind welder that has huge capacitors that look like gallon paint jugs 😂 that’s a bad combo Started looking for a cheap multimeter and dirt cheap adjustable power supply and the next thing I knew a few months later I had mountains of professional grade test equipment that I paid almost nothing for a government auctions, college auctions, and even on eBay. Once I cracked open my first piece of Tess gear and saw the beautiful engineering that went into everything… I was hooked here I am six years later with my own metrology business and a lab full of NIST calibrated equipment. And I do work for the best clients in the world who are super interesting. Mostly like government and research labs and stuff like that because I happen to be in a city where it’s kind of like the birth of science in America I guess that’s a good way to put it. Are used to avoid and hate electronics. And now I absolutely love what I do. But I couldn’t write a single line of code if I had to. I mean technically I can still write machine code andG code from my machinist and toolmaker days but just barely. You are much brighter than I am and you would absolutely be a natural in the world of electronics. Just know that when you’re working with your hands and crafting something that it’s never a linear process. Ain’t no paint by numbers when it comes to working with your hands no matter if that’s doing brain surgery or laying brick. It’s always different than you think it’s going to be and know That sometimes (pretty much all the time) you have to change and evolve in real time as the projects and variables fail/change/or whatever. Adam savage talks a lot about that on his RU-vid channel. Because if you can’t change with your projects and be flexible… Then you will either never finish anything or you’ll fight it the entire way and it will be a miserable process. Sure you might not be 100% happy with it when it’s finished but you wouldn’t have been even if everything went your way. So you might as well finish it. That is how you build up experience quickly. And you build up positive experiences instead of having drawers full of a failed unfinished projects that call out to you like ghosts in the night. I think you will make a wonderful electronics Tinkerer and troubleshooter! If you’re like me and you want to jump in on the deep end …just start buying really expensive golden age test equipment from brands like HP and tektronix from mid80s and 90s… Stuff on eBay that sold as “untested”… i’ll be able to pick up crazy expensive equipment for pennies on the dollar. And sure you might not be able to fix everyone of them but you will learn so much that way. And you are learning from the true masters of systems design and process engineering. Not that some of the products like the iPhone and stuff made today aren’t masterfully and wonderfully designed an engineer… But when you crack open a flagship piece of test equipment from Hewlett-Packard from the mid80s or 90s… it’s like staring up at the ceiling in the Sistine chapel. Just make sure you’re doing it safely and you’re not going to kill yourself. First thing to do is outfit your bench with nothing but GFCI outlets. If you don’t want to wire them in the wall then you can just buy adapters that you plug into the wall and then plug directly into those. Next Make sure you have an isolation transformer, Watch a few videos on them and make sure that you understand the difference between an isolation transformer like used in hospitals and it industry and a true “isolated“ isolation transformer. And third a decent current limiting device like a dim bulb tester. Once you crack into a high end professional piece of test equipment… You’ll be hooked and you’ll never look back.
I love this idea of a microcomputer, you've already got me on the hook with "keyboard" in the title :) can't wait to see the rest of your series. On the subject of keyboards, one of the previous videos with SA Dasher made me very proud that another person shares my same thought process, "if you use a tool a lot, make it a good tool". I have some nice pens and pencils on that same premise too.
@@mattymerr701 Hah, speaking from experience I imagine. I prefer ballpoints because I'm not writing anything important, and replaceable cartridges feel less obscure or complex than fountain pens. Maybe I'll fall down that rabbit hole someday though!
@@revolver265 Ah shit, you got me :P If you ever want to test the waters and want a better writing experience, you can't go wrong with a rollerball like a pilot G2. They don't require pressure to write, same as a fountain pen. But ball points are absolutely a very functional pen. Most people have no reason to look into other kinds of pen and more power to them, means they don't end up spending money collecting hehe If you ever do want to try out fountain pens though, a great starter is the lamy safari. They take cartridges and also have a converter to fill from ink bottles.
So while I do have thoughts about easier ways to do this , things you could have done differently thats just me being picky. I love that you uploaded a 40 minute video of a fail. There is stacks of useful information in the video but the most useful thing here is understanding that sometimes when you make a thing it wont work. Ideas are worth exploring even if they come to nothing if only because it helps to learn of a way not to do something. I look foward to seeing the progress of this project!
This is a really great video for any future DIY projects we may do as tinkerers. Thank you for posting this, I will definitely remember this when I get back to some personal projects and pursue finishing them or trying for something different.
The keyboard uses the same construction method as the Amstrad PPC512 and PPC640 portable XT computer. My Amstrad PPC512 is still working perfectly today after over thirty years of service, with a few mods including a Gotek and a portable 8-inch Color LCD RGB2HDMI Display. Thanks for the video, I was impressed by your willingness to move on. Now looking forward to the next episode in the series. Happy New Year.
I don't have any experience making a keyboard, so I can't really offer any advice. Just a question though. Maybe I missed it in the video, but why make your own keyboard from basically scratch? Why not just mod a keyboard into the case it self? I suppose the answer is probably because you wanted to try, which fair enough.
I took apart a cheap keyboard for a DIY project as well, but simply tinned and soldered leads onto all the pads, found a website that shows all key presses, and tapped the wires together while writing down what each one did. Took a while, but the goal is to make an Apollo mission style control panel for games like KSP. Using friction fit jumpers means that key bindings are easily redone. I've put thought into a similar project to yours of a wrist-mounted computer, and honestly the best solution was to pick up the smallest keyboard I could manage and instead burn myself out on the 3D printing and programming of the thing
Great video! Failures in a project are often more interesting learning experiences than if everything goes smoothly. As someone that has a small pile of projects that eventually hit the point of too much effort to fix my mistakes, it's refreshing to see a diy/tinkering video that shows what happens when things kinda go sideways. I've made my own attempts at trying to bodge together random electronic bits to do things they weren't originally designed to do, only to realize I've managed to wire something backwards which then lets the electronic fairies escape in a puff of spark and smoke.
You can buy really cheap used mechanical keyboards. You can get "broken" ones for super cheap and they're extremely easy to fix. If it's a broken switch you can just replace it, if it's the PCB you can just find a different combination of pins to bridge so you get the proper key press. Sometimes it's just a cold solder joint so you can just reflow it. Most importantly you get both the PCB, the controller and the switches plus keycaps if you decide to go that route. If you go for a TKL layout you'll get compact enough form factor. I recently got the HyperX Alloy Origins Core for about 15 dollars and all I needed to do was install one jumper wire to fix a key that no longer worked because the pad on the PCB was bad. I'm using it daily now and it's just amazing. I also got a full sized keyboard for around 5 dollars that just needed to have some of the joints reflowed.
I've tried to solder to a very similar keyboard controller to make a multimedia hot key panel. It's very difficult to get a good connection, but that's also the fun. Learning what's not worth doing
The thing with lots of your videos is the incidental info which is dead useful and/or new to me - wire glue for instance I'd never have thought such a thing existed. But what grabbed my attention was the usb fan/hose combo. I can imagine loads of applications for this moving hot air in the greenhouse.
I gather the lesson learned here is: Use quality components and parts; take breaks when you get confused; and I’ve heard different solder melts at different temperatures so maybe look into that?
The way we used to do these sorts of projects in the past was to get wire-wrap terminals and then color-code the ends with a sharpie. It does mean you are dealing with a much fatter assembly, but there is no soldering involved, either.
I'm very glad that you decided to share your rational for how you did each step of this process. I'm very interested in IT myself and work in a sort of IT adjacent field (US military, basically just a supervisor), but I always find myself intimidated at these sort of larger projects. I always wanna do things my own way, but find that that often leads to intimidation at the massive amount of work I've got ahead of me 😅Love your videos!
Thank you for this video!!!! I tried a similar project with the same keyboard (branded as Inland, but everything looks the same). I was building a controller for arcade games. I got it about half working, then a trace got pulled off the controller. I set the project down and eventually went back to it later with control boards that are designed for that use. Learning to stop a project and rethink it is a valuable skill to have!
