Road to Sage Understanding

[usfbullz]

Hello there and welcome to my blog. I intend to document my journey in understanding electronics better through doing projects and mastering basics.

Why am I doing this?
Despite doing a whole degree in EEE, I have essentially forgotten everything already and its only been 6 months since graduation! It doesn't help that I'm not working in a techical role at the moment due to severe imposter syndrome.

I also experienced extreme burn out senior year and barely scrapped what is considered a good grade here in the UK. Yet despite all the setbacks, I still have this massive nagging voice at the back of my head that wants to spend more time understanding and one day working as a engineer, especially in startups doing dope things using tech. Been really interested in wearable health devices (did I mention I'm a gym bro?), audio electronics... especially headphones (I like music), clean tech (especially renewables and batteries), IoT and robotics (its awesome watching them mimic humans and do tasks).

I always used to be on forums as a kid growing up as a gamer and I really felt at home on them, so I figured doing the same for electronics would be a good start to getting better. I hope to hold myself accountable by putting myself out there and forcing myself to get better because I have a tendency to second dobut myself when things get tough and talk myself out of doing this.

About me?
To be quite candid, I'm not the sharpest tool in the shed when it comes to electronics, and that's not just for the sake of putting myself down. I processed things much slower compared to everyone around me during college days, struggled with basic labs as a freshman to the point where my lab partner would do the hands on work whereas I just did the math/calculations/write up and was very slow at "building a picture" in my head of understanding. And I'm OK with that, it just means I have to find other ways to accomodate for it. The only thing keep me going is my sustained interest in PCB design and embedded systems, as in actually working on them and no longer just the idea of doing it.

The way I like to understand electronics is visually. Like, say I'm look at the schematic for a circuit. I like to visually imagine every component and part in the circuit functioning in my head as one system, if I can't imagine it, I struggle.

I haven't been doing electronics since I was 6 nor programming. Whilst I loved video games despite having a lunchbox laptop and tried to program when I was 14, I found it difficult so quit. I didn't even know electronics as a discipline or hobby existed until I was in my final years of high school aged 17 after realising I didn't like Computer Science as much as I thought I did, the only thing I enjoyed was learning about how the CPU works, Boolean logic, gates/karnaugh maps and binary (floating point, two's complement). I also really enjoyed Math and Physics a lot so it seemed like a no brainer to go for EEE... except college was nothing like I expected especially when a global pandemic occurs.

In job interviews for role as Electronic Engineer, I was told I'm more suitable to sales than engineering (which I won't deny because of my personality). There is always technical sales but I like routine and travelling for lesiure, not for work. Although, I always appreciate picking up some commercial knowledge since I do have a business savy side to me as well.

Right now, without giving too much away, I'm interning in a commercial role related to tech at the moment.

Where are the current gaps in my knowledge
Time to embarass myself...

• Circuit Analysis:
  Surprisingly, I still mess up on determining whether certain parts of circuits are in parallel or not (honestly feels like a guessing game sometimes).
  
• Basic Components:
  I believe I have a good understanding of the purpose of resistors, capacitors and diodes in circuits. I sometimes struggle to visualise the capacitor discharging whilst other components are functioning but the one component I still need to visualise better is inductors.
• Op-Amps:
  Even after all this time, I still have a very hard time visualising how the electroncs flow in a op-amp circuit. It actually annoys me a lot despite the amount of videos I've watched on the topic. I've even used Falstad to help me but no luck.
• PCB Design and Testing:
  One of my favourite classes as a undergrad. I'm very slow at designing PCBs and also really struggled with layout especially for mixed signal design. I found multi-layer boarding to be alright, just a bit trippy sometimes with all the different factors you have to consider when it comes to parasitic capacitance and inductance and preventing crosstalk etc.
  
  I was also not very good at debugging PCBs with oscilloscopes. I remember spending hours sat a bench trying to get power through to certain parts of subsystems on a PCB and hooking it up to oscilliscope and trying to get signals out of them but it just never worked.
  
  
• Semiconductors:
  I have a basic understanding of how a semiconductor works. Like Op-Amps, I have a very hard time visualising how the electroncs are following in a circuit that has BJT Transistors. MOSFETs are a bit easier to understand since I just imagine them as a voltage controlled switch. It's just that with the BJT, when the electroncs follow through the base... where the heck are they going? The collector? The emitter? Both? Why? Don't even get me started on CMOS/PMOS...
• Graphs:
  All kinds of graphs. Bode Plots, Frequency Plots. Heck, I didn't truly understand Decibels and its purpose till recently. I just used it if I was instructed to by the professor
  
• Electromagnetism:
  I almost failed this class... hard. I thought I failed but I didn't. I didn't find the math hard, it was the actual concepts I found hard to wrap my head around! I'd like to go back over them and develop a more intimate understanding of them. Also magnetic circuits.. trippy..
• Control Theory:
  Undoubtedly one of my favourite classes along with PCB design as a undergrad. I really enjoyed the math behind it. I did it as both a sophmore and senior. In my senior year, it was taught terribly and I don't remember a single thing from the nearly 700 page slides the teacher had which included key concepts like State Space, PID control... And don't even get me started on those Bode graphs...
  
• Signal Processing:
  Convolution, FIR/IIR Filters, Fourier Transforms, DFT/FFT.
  
• Microcontrollers:
  Interfacing, I2C, SPI, UART, USART, CAN. I know what they are but have never used them.
  

Observations
I've noticed throughout my time as a student, that I never really asked questions or tried to probe deeper to understand something. I would just say Ok and move on. That has severely stunted the inquisitiveness needed for me to ask good questions.

One thing I'm going to try doing differently is actively building a picture in my head as I really need to be able to imagine things to get it I find.

I'm taking a much more visual approach to my learning where I'll be doing projects and making active use of simulations like LTSpice for example to understand things.

What am I working on now?
I'm currently working on a Flight Control System project from scratch done by a engineer called Phil's Lab (https://github.com/pms67/HadesFCS). This is primarily designed for fixed-wing, autonomous drones to enable them to perform tasks autonomously, such as take-off, landing, waypoint navigation, payload delivery, etc.

This includes the hardware, firmware (using FreeRTOS), Kalman filtering for state estimation, control algorithms, base station, communication protocol, flight simulator, signal processing etc.

I've been working on this for an hour or two each evening for the past week now and its actually been very enjoyable which I'm surprised by, because back then I dreaded every second spent of my degree. I think since I'm no longer under any pressure to pass, I can just take my sweet time.

This project will definitely be a challenge though because it involves programming, which I was incredibly bad at during my undergrad. I was somewhat OK at Python, but other languages... not so much. I especially struggled with System Verilog on FPGA (which I didn't even get to use, all virtual  ), C was a lot more forgiving and MATLAB... I still get flashbacks working on image processing tasks.

I'm nervous putting this up but I'd like to get better and actually build out ideas I have in my head one day instead of it just being a what if or self dobuting myself.

Anyways, that's a long enough introduction. If you decide to stay, I hope you enjoy your stay and witnessing my journey to get better. I imagine I'll definitely be editing this first post as time goes on to make it more clear and succinct since I tend to ramble.


If this sort of blog isn't allowed, admins/mods feel free to remove and apologies for the inconvience caused to you all.

[usfbullz]

Have finished the first subsystem for the flight controller. Note that I haven't labelled the components yet, because it slipped my mind. Will do later.
EDIT: Upon further inspection, I realise I still have some important connections missing. Will update ASAP.


I should also note its my first time using KiCad for a project having mostly used Altium, I found it easier to pick up compared to Altium, but perhaps I'm speaking too soon as I haven't actually done any PCB layout using KiCad yet.


I spent more time understanding the different parts of the circuit than the actual schematic wiring itself since this is basically me re-doing what Phil did but step by step so I understand what's going on.
Had to refresh myself on a few concepts but it's starting to come back to me.

Spending time working on next subsystem which is power. Will post photo once done.

[usfbullz]

I finished the Power subsystem. I'm working on the peripherals subsystem which will consist of a PWM driver for controlling the servo, EEPROM and Flash memory.

Like I've mentioned before,  most of my time is being spent to understand the subsystem rather than actually wiring it up. One particular bit I was finding difficult was the thought process behind the pin assignment for the STM32 that was used because I kept looking in the application notes of that specific STM32 for a pin layout to no avail which was pretty frustrating (did I mention I've never worked with a STM32 before?) but I found another video from Phil's Lab (the same person who made this design) who made a tutorial video for desiging STM32 MCUs in KiCAD. Turns out, I was meant to use the STM32 Cube IDE as a reference because its very difficult to be able to do it from the datasheet which was a literal bruh moment for me.

I've also been spending some time to properply understand the serial communication protocols better and I started with I2C. I read up a Texas Instrument document and it made a lot of sense. It's been so long since I used it as I was a freshman when I learnt about it but we did no projects with it and never used it again throughout my undergrad (I take responsibility for this fault as I should have been a more proactive student with projects)
 
In the mean time, I actually picked up the Go Board by @nandland so that I can do some FPGA projects since I wasn't able to do any physical hands on stuff with during undergrad due to COVID and it was too expensive for me to afford, plus I sucked hard at Verilog. But I intend to go through the tutorials of nandland plus also redo my undergrad project for my didital systems design class.

Whilst there were countless times I wanted to bash my head into my wall due to how confused I kept getting, I'm glad I stuck it out in the end and its made me more eager to go deeper into understanding despite how much gaps in my knowledge there is.

[audiotubes]

I am not EE and so have nothing useful to say about that. But it is important for people to be competent and feel competent at what they do, and not everyone is suited for every job. Sometimes it's because their minds don't work the way this or that job requires, or because their personality isn't suited. Given we spend (waste) so much time at work, the best advice anyone can give somebody starting out, is that you should aim for a job related something that you love, that suits your personality, and that suits the way your mind works.

I don't know what opportunities you have, but engineering sales can be a fascinating and rewarding job. It's rare for people to find the perfect job, whoever does should consider himself extremely fortunate. But you should still place yourself in a position for success rather than failure.

[usfbullz]

I am not EE and so have nothing useful to say about that. But it is important for people to be competent and feel competent at what they do, and not everyone is suited for every job. Sometimes it's because their minds don't work the way this or that job requires, or because their personality isn't suited. Given we spend (waste) so much time at work, the best advice anyone can give somebody starting out, is that you should aim for a job related something that you love, that suits your personality, and that suits the way your mind works.

I don't know what opportunities you have, but engineering sales can be a fascinating and rewarding job. It's rare for people to find the perfect job, whoever does should consider himself extremely fortunate. But you should still place yourself in a position for success rather than failure.

Thank you for your kind words, it is something I will consider as I have applied for some roles in that field. The feeling of compotency... something I've never really felt before other than doing basic PCB design and anything involving mathematics (very comfortable with it throughout college) interestingly enough. It's also definitely my pride and stuborness as well that plays a big part in my reluctancy to try for more of the sales roles but something I'm coming to terms with.

I do definitely feel the concepts are clicking more now that I'm out of the pressures of an academic environment.

[audiotubes]

One of my oldest friends is an EE who worked for a few decades as an engineer and then had a long engineering sales career with IBM.

The guy is very good in math and all the practical sciences but also has a very jovial personality and good attitude. It turned out he was even more successful in the people dimension and enjoyed it more than he expected. He was able to combine his interests with his natural people skills, flew millions of frequent flyer miles and got to see many places he never dreamed of going.

So, try to go with the flow and you will probably succeed in ways you probably didn't imagine.

You don't have to figure it out day 1, and not even after a few years. But you should be enjoying what you do and that will lead to good things.

[PlainName]

I spent more time understanding the different parts of the circuit than the actual schematic wiring itself

Unfortunately that shows. The schematic looks nice but is not very useful - to see which LED is connected to which I/O pin (or even to find out it is connected to a pin in the first place) you have to read the label and then closely read a fistful of very similar looking ones to find a match. And maybe there's another label with the same name elsewhere which you can't see or haven't yet noticed.

The LEDS are right next to the port pins. Literally. Why not just join the former to the latter with actual wires and make it instantly obvious how things are connected with just a brief look? You can still keep the descriptive boxes if you want, but just join things up - it is called a diagram for a jolly good reason, after all.

[usfbullz]

I spent more time understanding the different parts of the circuit than the actual schematic wiring itself

Unfortunately that shows. The schematic looks nice but is not very useful - to see which LED is connected to which I/O pin (or even to find out it is connected to a pin in the first place) you have to read the label and then closely read a fistful of very similar looking ones to find a match. And maybe there's another label with the same name elsewhere which you can't see or haven't yet noticed.

The LEDS are right next to the port pins. Literally. Why not just join the former to the latter with actual wires and make it instantly obvious how things are connected with just a brief look? You can still keep the descriptive boxes if you want, but just join things up - it is called a diagram for a jolly good reason, after all.

Thank you for this, that is a very good point. I have been blindly wiring up the schematic in the exact same way the original author did it and didn't think too much about it as I was more focused on making sense of the schematic. I will take a more proactive approach to adding improvements and make it more effective as you've described.

[PlainName]

Apols if I came over a bit strong on that, and well done for taking it gracefully 

[usfbullz]

Apols if I came over a bit strong on that, and well done for taking it gracefully 

No worries at all, I always appreciate getting feedback as I want to keep get better at this so thank you in fact! I will edit my post with the original picture with a new one once I update it with your feedback + I realise there were actually still some ports missing

I'll also be posting a picture of the peripheral subsystem soon so feel free to critique as you see fit.

Also for anyone that's curious, here is a pdf describing this project from the original creator, Phil's Lab: https://github.com/pms67/HadesFCS/blob/master/Documentation/HadesHardwarePresentation.pdf

It's honestly a lot easier to understand than I'm making it out to be 

[audiotubes]

That's awesome! I have confidence that you will succeed.