Robi_g

So it's been a little over a month, and lots of progress has been made. 1) Received dongle and keypad PCB, the keypad works, the LEDs work, the uC and FPGA are both recognised and programmable, but in a rookie error I had the connectors mirrored, so have had to modify the keypad board (waiting on that to be shipped). 2) Written and mostly simulated the VHDL for the FPGA. 3) Built the CNC, with some extra bits to stiffen it up, and some limit switches so it's actually usable. 4) Cut out a draft of the keypad backplane (see pics) the rough edges are my fault, not the router's. So hardware-wise, things seem to be rounding off nicely. Won't be long until it's time for the software.

I've since figured that buying a cheap Chinese CNC router and cutting my own plate would only be about £30 more than getting it cut in the UK. They're only meant for wood and plastic, but if it only ever makes 2 things and then completely breaks I'll be quids-in! Either that or I can make all the acrylic stuff I want, with the thickness caveat geo3 mentioned. Will report back with the good or bad news

I've found a website that gives rough templates for keypads, and then I've added my own bits to it. So now have a design. I'm finding that for laser cutting of cheap steel in UK is ~£80! Will look abroad tomorrow and see what I can find.

Hi All, Some background for those interested: Following on from a previous thread (forum link), I've been through a PCB revision, thrown out the NXP stuff and gone over to Microchip, and now have a second revision of my custom keypad project. This one consists of a dongle that does all the USB, and several keypads of different sizes. The keypads will all have CPLDs on them to talk to the dongle and handle the keypress events on the keypad. This allows me to keep costs down by only having a small 4-layer PCB, with all the keypads being cheap 2-layer ones. i.e. I'll plug the dongle into a full-sized keyboard and it will behave as a keyboard, or I could plug it into a smaller keypad and it will act as that. The dongle will have different profiles and such on some EEPROM on the dongle. The actual question: The first keypad I'm doing is a 4x5 switch layout, like a number pad, but with 20 keys instead of 17/18. I'm getting a friend with a 3D printer to do a proper encolsure, but need a plate to mount the keyswitches on. See attached photo for a rough idea of this. I know there's pre-made plates for standard keyboard layouts: https://mechboards.co.uk/product/universal-alu-plate/ but they don't give thicknesses, drawings, or other useful info. In order of preference, does anyone know: If anyone makes exactly what I'm after (4x5 cherry keys with some border around the edge so I can mount it to something). Drawings, dimensions, or any info I could use to speed up designing a plate myself. Specifically the thickness of plate required, as I already have the switch dimensions since having done the PCBs. A product like this that I could buy and then measure myself. It would also be helpful if anyone knows of any places in the UK that do cheap sheet metal machining if I can't find a COTS part. Thanks in advance. When all this is done I'll post code and build log stuff on the forum if anyone else is interested. The final version will have macro functionality and other fun bits.

You've given me a lot to think about. Thanks so much for the advice.

- We're in design and manufacture of low-volume equipment, some PCB design, some embedded software, the odd bit of analogue or FPGA stuff. - A really large project might be 10GB, with most being ~5GB. We'll use the NAS more as a nightly backup rather than working off of it, and as a place to keep all our admin synchronised. Security and long-term stability trumps anything else for our use case, because we're not reliant on the network to get work done. - The server 'room' will indeed just be a table, or maybe a cupboard. You're not far off with the place being run out of a garage - We absolutely require external access to the network as I frequently work remotely. This is the main reason for the pfsense router over some other less complicated thing, because I heard about its good security and logging capabilties. But again, speed isn't a priority here. - The reason for planning on going for some dedicated machines from the outset is because we don't know where we're going to be in a year or two, and don't want to be caught short with some limiting network. I.e. we'd rather spend some good time now while we're still ok with just cloud storage and aren't under any pressure from that. - Some background on us - we're savvy with electronics (having both got degrees in it and a few years in industry) and consumer PC hardware, but neither of us have yet to do any serious networking. I've run some basic LAMP webservers and have played around with some consumer NAS stuff but that's about it. So we're lacking in experience, but have the means to learn reasonably quickly.

I built my own reflow oven, and thought that my experience might help others looking to do the same. Attached is a brief build log and design explanation, along with a link to the code running on the oven. Pricing is also included, I always find it annoying when other build guides don't give a guide to the price of things. It should give people an idea of what it's going to cost before they get started, and shows the mistakes I made that cost me extra. Reflow Oven Build.pdf

Me and a partner have recently started a business. We're currently using Google drive to share and sync files but we don't want to rely on this long-term (and that's no fun anyway - we're both electronic engineers). As a basic setup I was thinking a pfsense router, unmanaged switch, and a FreeNAS NAS. All gigabit for now. I'm planning on running the router and NAS off of some Dell servers, that'll probably come to £300 total without drives. I've looked at Dell Optiplexes and other cheap desktops, but by the time you pay for better network cards, and consider the lack of reduandant PSUs / remote management / ECC memory it's a similar value. So what I'm asking is are there any things to look out for with Dell R710 / R510 / R310 with regards to compatabilty or other 'gotchas'? I got burned a few months ago with an HP DL180 G6 that was going to be a NAS for personal use, but it turns out that the fans stick at 100% if you install a non-HP PCIe card. Also, are there any good resources for networking hardware that people can recommend? Thanks in advance

Looks perfect, thanks :)

I'm getting along with a project to make a 'programmer's keypad', where keypresses type out common chunks of code in python and C (e.g. there's a key to type out a switch statement in C, or a standard file header (author, date created, etc...). This project is from the ground up, including PCB and software (I might put together a build log at some point), and part of what needs doing is finding some keycaps that have the commands printed on them. I was initially thinking stickers, but I want to make a proper job of this and reckoned that since mechanical keyboards have become so popular in the last 5-6 years that there'd be a place that will put custom decals / text on a set of kecaps. The board only has 20 'caps, but I don't mind paying for a MOQ of a full set (104 / 87 / 61) as it'll leave spare sets for friends and / or let me try out a couple of styles. Can anyone recommend a place(s) that does this sort of thing? (preferably based in UK)

Just build a new Ryzen system, and the plan was to have Windows 10 on an M2 SSD, then (as half-novelty / half-actually-useful) Windows 7, XP, and Ubuntu on an old 500GB hard drive for when I wanted to run old / otherwise incompatible with W10 software. So Windows 10's running just fine, but I've given up with everything else (mostly, I'm ashamed to say, through lack of trying). The XP Journey BSOD for ACPI support, but you can skip that. Another BSOD due to windows XP being too old to include AHCI drivers, and the motherboard being too new to support ATA mode. Read about loading drivers with a floppy drive (but only a floppy drive, not a USB floppy drive), and installing via USB with Windows Server 2003, but couldn't be bothered. I'm never going to actually use XP anyway. 0/1 The Windows 7 Journey Ryzen USB drivers aren't compatible with Windows 7 because Windows 7's so old (I'd forgotten that Windows 8 was released 6 years ago!). I do have a PS/2 keyboard to get things going, but it's 150 miles away (my words 2 months ago of "when am I ever going to need this again?" rang loud and clear in my head) Read an Anandtech article on unattended booting, but again, I couldn't be bothered. I have windows 7 on a laptop so I'm only losing some performance by not having it on the new desktop. And anyway, I'll just wait until I'm home at Christmas and grab the keyboard then. 0/2 The Ubuntu (15.04) Journey A load of error messages came up, I didn't read them. Changed some bios security things because that's always the problem with these kind of things. Nope. Same errors came up. I've been working on this for too long. Screw this, who needs Linux anyway?.... 0/3 Guess it's going to be an old Dell Optiplex that becomes the 'it won't run in W10' PC instead

That's why I said shottkey diode across the transistor, where its lower breakdown voltage would allow current through it when reversed biased hard enough. I'm aware that It's called an h-bridge, and I also know that there's 'more to it than that' (I have a Masters in Electronic Engineering and work as an electronic engineer). I was trying to reduce the amount of fluff around my answers for the OP, to save them going down the rabbit hole of learning everything about everything. It would be a bit of a baptism of fire to ask how PWM works and end up with some huge glut of jargon. I'm not saying you can't do things your way, quite the opposite, I completely agree with you on that. But I'm also saying that you can stick a diode with low breakdown voltage across the transistor to protect it. The method you're pushing is probably better, as you're not injecting a load of current into the ground plane (although that's what capacitors across the supply are for I guess), but it's wrong to say that the other method just flat out won't work (unless you use a regular diode, then it won't work for the reason you stated). Again, I have seen working circuits that use either method. Also, diodes in regulators is a different use case. You can use diodes to create voltage references too, but that's got nothing to do with motors.

I'd save the money on the AIO cooler and spend that on the GPU instead

Wow, such a journey, glad it worked out for you in the end.

You can do either, I've seen circuits with both. Across the transistor allows you to run the motor either direction (with some extra transistors etc...) if you just stuck it across the motor you could only go in one direction. In the end, all you're looking to do is protect the rest of the circuit from the motor.