ฝัง
- เผยแพร่เมื่อ 28 ธ.ค. 2023
- #engineering #PCB #entrepreneur #productdesign
Hey Everyone! If you haven't seen episode 1, in this series I'm making a treatment device for TMJ Disorders that will let the user more consistently and effectively apply heat and massage to the jaw muscles - removing muscle knots and spasms that can lead to disc displacement.
But the twist is this isn't just going to build a prototype - I'm really going to try to turn this into a product I can sell. So follow along for not just the engineering and product design journey but also how I end up fundraising, manufacturing, marketing, and delivering.
In this episode, we start with the electronics and PCB and circuit design, with a ton of mistakes made and lessons learned. - วิทยาศาสตร์และเทคโนโลยี
Amazing insight on design and problem solving that come with it. I wish you all the best for treating your TMJ disorder and I hope this final product will be most beneficial for others with same condition as well!
Very cool series idea. The electronics design/programming/etc will be interesting to some, but I imagine your channel will get tons of traction if/when you start uploading buisiness vids (market research, professional prototyping, sourcing, funding, marketing, etc). Thats where things seem to get hard... and expensive. Looking forward to seeing your approach to the actual to-market strategy!
You don’t know how right you are. I’m with starting branding and marketing right now actually in parallel with development. Hopefully I’m able to catch up soon with the videos because I definitely agree that’s the interesting part for most people.
When using the shunt resistor you need an op-amp. That’ll get you over that hump. GL
Yup, I am! Amplifying it to around 2.5V with gain of 50. Nichrome just has super low temp coefficient of resistivity so that’s where I’m still getting the borderline tiny delta V. I’m thinking about subtracting the nominal shunt voltage from the measured and amplifying the difference to improve resolution. But that’s a lot of components so we’ll see.
Watching this I feel your pain. My one bit of crucial advice would be to stop doing endless cycles of crash-and-burn PCB design, assembly and failure.
Build a Frankenstein breadboard first- build every part of the circuit as a module that you can remove, rewire and replace with ease. How do you get to the nodes you need to probe? Break these out so it's really easy to probe and connect to everything. Only then, when all of this is working PERFECTLY would I start laying anything out on a PCB. (except just to confirm it all fits) You might need to make small test boards for the SMD parts, these should be placed on a big breadboard that includes all the hardware. Too many beginners think that making final PCBs is the only way to design things but it's a total waste of time to think this way for a complex project, unless you are very experienced.
Yeah great advice man. I can’t say I haven’t thought of doing that, I guess I’m just a bit of a PCB gambler lmao. There’s honestly so much going on with this project I enjoy the wait times to do parallel work. But you’re 100% right, it’s just my first project where I’m able to make PCBs instead of Frankenstein breadboards so I’m hype about it for now.
@@noamaiz Designing hardware is never a "one-and-done" affair, I like to think of it more like sculpting; A great final design takes a lot of "beating things into place", a lot of testing, and a lot of interaction with the code and hardware, together. The final PCB is a terrible environment for these activities, every design change brings on a ton of shadow work that has nothing to do with getting the project done, the difficulty of making measurements leads to just "assuming it works" rather than taking the time to actually test each function block completely, to really know what is going on. It's so easy to spend hours calculating some theoretical design stuff only to find a subtle real-world problem that you never considered that mucks it all up, the key is to discover these problems up front, at the stage where they have the least impact on your project timeline. I always hear people say "Ohhh breadboarding takes so much time, what a waste when I can just go for the final PCB?"
@@leosbagoftricks3732Well said. I appreciate the advice
So cool!!
Really cool. How hard was running the traces? Were you able to do any simulations beforehand? Also, did you need to buy all the equipment? I've been working on a project like this too, but there's a ton of sitting around and waiting. How long did it all take? Sorry for all the Qs!
Thanks for the kind words brother. Running traces was pretty annoying. Latest version took like a days work since I avoided autorouter. I don’t use any sims for this. All the equipment I’ve kind of just been collecting through my life. The O-scope I got from my ESE department back in college. I have one 3D printer but I use my friend’s bambus cuz they’re way nicer. And yeah man, there’s definitely waiting but honestly there’s always something else to work on with this. I’m still in the process of engineering, but soon I’ll talk a bit more about what’s going on outside of engineering like dealing with FDA, patenting, and fundraising. At this point my videos come pretty much as things happen so stay tuned!
10:20 That is a seriously vintage website design. Making me all nostalgic. I salute your ambition. Be aware what you're working on will almost certainly be classified as a medical device and there's an insane amount of regulations to follow when bringing one to market. Researching those sooner than later could save you a lot of redesign hassles if your device doesn't end up meeting regs.
Haha yeah, the vintage ones always have the best stuff though. This device will be classified as class 1 so no need for FDA approval or trials or anything (like most massage guns). But you’re definitely right, there are a couple major things that go into registration I’ll get into soon.
Awesome, very helpful
Goat
Heat