4 ms·

Doing PCBs for 12th year now without a complete engineering education. Just as the lack of formal CS background imprints on your lifelong experience with compu

Doing PCBs for 12th year now without a complete engineering education. Just as the lack of formal CS background imprints on your lifelong experience with computer languages, the same way is the lack of EE background. From time to time, a recent grad hire comes by, and fixes my work, and shows me where I screwed up with a few calculations on a pocket calculator.

cinntaile 5y ago
It's likely that it's very specific knowledge that you lack no? If you wantee to address that it should be possible with self-study?
- crispyambulance 5y ago
  It is theoretically possible with self-study, but the problem is it's still very easy to make practical mistakes even if you "understand" all the theory. Eventually you run into a problem and have to re-spin the board with no margin of extra time (assuming this is for work). In that scenario, having access to an experienced expert is immensely helpful.
- baybal2 5y ago
  I pretty much continue self-study every time I stand corrected. Things learned from junior hires: a lot in RF domain, opamps, 74XXX trickery, signal integrity measurements, fiddling with high freq PCIE, DDR, PAM4 lanes in the sciency way, and whole new world of advanced power electronics, and designing it according to standards, and codes...
- NikolaeVarius 5y ago
  Its very hard to replicate 4+ years of "required" study because you don't know what you don't know
  - cinntaile 5y ago
    It's not like EE curricula are a well guarded secret though, not in todays day and age at least. So that shouldn't be impossible to figure out.
  - gorjusborg 5y ago
    On the other hand, not knowing something that doesn't matter to your design shouldn't stop it from getting done.
fxtentacle 5y ago
Same here. I have no formal education but I recently finished my first 12 layer USB3.1 SuperSpeed + FPGA board. Every once in a while, I have to book a freelancer to explain niche knowledge to me - for example how to calculate the impedance for the USB3 specs - but thanks to good software, it was a lot easier to learn than I thought. By now, I have my own tiny KiCad fork to make things easier to read and look like I expect it to. https://gitlab.com/fxtentacle/kicad/-/commits/vector_font https://gitlab.com/fxtentacle/kicad/-/commits/vector_font
- gmurphy 5y ago
  Any tips on how to find/book those freelancers? I've been looking for similar help
  - fxtentacle 5y ago
    I asked a lecturer at our local university to introduce me to some of his PhD students. That said, many hobbyists also just ask on the KiCad forums. And I know there are PCB review tasks on upwork.com
- le-mark 5y ago
  This sounds exactly like a pcb project I have in mind, plus a bunch of ddr3. How hard is it starting from almost zero? I imagine diy isn’t realistic, how much does design of such a board cost?
tremon 5y ago
We didn't even cover PCB design in my EE years, the course load was pretty much evenly split between circuit theory, VLSI (processor) design, power distribution systems, and embedded systems programming. Plus the necessary theoretical background in physics, mathematics, control theory, and management.

cinntaile 5y ago

It's likely that it's very specific knowledge that you lack no? If you wantee to address that it should be possible with self-study?

crispyambulance 5y ago
It is theoretically possible with self-study, but the problem is it's still very easy to make practical mistakes even if you "understand" all the theory. Eventually you run into a problem and have to re-spin the board with no margin of extra time (assuming this is for work). In that scenario, having access to an experienced expert is immensely helpful.
baybal2 5y ago
I pretty much continue self-study every time I stand corrected. Things learned from junior hires: a lot in RF domain, opamps, 74XXX trickery, signal integrity measurements, fiddling with high freq PCIE, DDR, PAM4 lanes in the sciency way, and whole new world of advanced power electronics, and designing it according to standards, and codes...
NikolaeVarius 5y ago
Its very hard to replicate 4+ years of "required" study because you don't know what you don't know
- cinntaile 5y ago
  It's not like EE curricula are a well guarded secret though, not in todays day and age at least. So that shouldn't be impossible to figure out.
- gorjusborg 5y ago
  On the other hand, not knowing something that doesn't matter to your design shouldn't stop it from getting done.

fxtentacle 5y ago

Same here. I have no formal education but I recently finished my first 12 layer USB3.1 SuperSpeed + FPGA board. Every once in a while, I have to book a freelancer to explain niche knowledge to me - for example how to calculate the impedance for the USB3 specs - but thanks to good software, it was a lot easier to learn than I thought. By now, I have my own tiny KiCad fork to make things easier to read and look like I expect it to. https://gitlab.com/fxtentacle/kicad/-/commits/vector_font https://gitlab.com/fxtentacle/kicad/-/commits/vector_font

gmurphy 5y ago
Any tips on how to find/book those freelancers? I've been looking for similar help
- fxtentacle 5y ago
  I asked a lecturer at our local university to introduce me to some of his PhD students. That said, many hobbyists also just ask on the KiCad forums. And I know there are PCB review tasks on upwork.com
le-mark 5y ago
This sounds exactly like a pcb project I have in mind, plus a bunch of ddr3. How hard is it starting from almost zero? I imagine diy isn’t realistic, how much does design of such a board cost?

tremon 5y ago

We didn't even cover PCB design in my EE years, the course load was pretty much evenly split between circuit theory, VLSI (processor) design, power distribution systems, and embedded systems programming. Plus the necessary theoretical background in physics, mathematics, control theory, and management.