Embedded - 35: All These Different Reasons Why You Might Want to Do Something
Episode Date: January 15, 2014Want to learn how to get from idea to schematic, through layout, all the way to physical boards? Elecia spoke with Chris Gammell about his Contextual Electronics course to teach the missing steps bet...ween what an EE learns in college and what an design engineer's job entails. Chris is co-host of the excellent electronics podcast The Amp Hour and author of Chris Gammell's Analog Life. On twitter, contact Chris via @Chris_Gammell or ask questions about the course @ContextualElec. We mentioned UT Austin's online embedded systems course which starts soon as well. Contextual Electronics includes some in-depth KiCad instruction. Some intro (and free) KiCad tutorials: Chris' Getting to Blinky Series teho Labs Wayne and Layne Curious Inventor
Transcript
Discussion (0)
This is Making Embedded Systems, the show for people who love gadgets.
My guest today is Chris Gammill of The Amp Hour to discuss his online electronics course.
Finally, someone to tell me how to do a double-E's job.
Chris, thank you for being on the show.
Yeah, it's great to be here. Thanks for having me on. In a show with Phil King, I might have spent some time whining about my inability to see a hardware project past a hand-drawn schematic.
I get frustrated with the schematic capture.
I get lost in layout.
I get nervous about fabrication.
And here you've got a way to help me.
Well, I have one really simple way to just fail a lot.
And then eventually you feel a little better about it but uh yeah until then uh yeah i do have i have a new course called contextual
electronics it's uh you know it's it's an eight-week course that's two eight-week courses
basically the first one's kind of teaching how to how to design something and then the second
part's teaching how to build it.
So the design is what's a resistor, or do I need to be past what's a capacitor and what's an inductor levels?
That would be helpful. So the people I'm kind of targeting and hoping will join the course
are, what I'm saying is advanced Arduino users. So know feel a little bit comfortable coding uh feel a little
bit comfortable using leds and feel a little bit comfortable you know like leds resistors capacitors
that kind of thing like passive components maybe have pondered the the intricacies of an op amp or
a 555 timer before and uh you know and then from there we can kind of do the rest. We'll jump into the rest.
Okay, so basically having read the first Forrest Mims book
or having fussed a bit with electronics,
but not necessarily understanding how an op-amp works inside.
No, oh my God, no.
I think I'd have like two or three people that would sign up.
Because that's the thing, there's like this weird dichotomy.
It's like if you try and get people that are too advanced, no one will ever join the course.
And then if they're not advanced at all, if they haven't touched anything, then you're going to be spending a lot of time.
Here's a voltage divider.
Here's a resistor.
Here's how you read the bands on a resistor.
But really what you should get is a microscope and some smaller resistors.
Yeah.
Yeah.
So eight weeks, how much does it cost?
Eight weeks, it is, so there's different tiers, actually.
Yeah, this is interesting.
So this was all because I wanted to make sure that people could afford it and then there's different levels.
Basically what it is is you're paying for my time.
That's kind of what it comes down to.
So there's a-
Isn't that what we all really want anyway?
Right, exactly.
So like the lowest level is what I'm calling audit.
And basically it's just the videos.
It's just there's a bunch of videos
for picking out components and putting them
into schematic and doing layout and stuff like that. A lot of the time, it's real time kind of
watching me do it. So there's not as much of a summary videos and stuff like that. So it's really
a lot of that stuff. It's all of the stuff you need to see someone going from the beginning to end of building, designing a widget, right? And I call
that technology voyeurism. So that's kind of what the audit, really the whole course is built on
that, where it's, you know, you're watching someone build something and there is value in that.
And then there's a middle tier that is basically, there's a community built around it. You're still watching all those videos,
those other videos of, you know, real time design and watching all the decisions. But then there's
also there's group activities, there's tasks built in, there's, there's summary videos,
more background videos about the internal guts of everything. And then finally, there's a top
tier where it's kind of like I'm coaching. There's different coaching sessions with me.
You get prioritized forum access and stuff like that.
Basically, if you need more help, if you're not feeling super comfortable, it's a way to get a little bit more of my time in order to get more focus and really learn this stuff.
But the prices are pretty wildly different.
I mean, the lowest level
was less than 100 bucks and that's all the videos yes right and uh yeah so it's 75 225 625
and not only that there's actually an entire course on how to use kai cat also the uh
the open source cad program, which is my favorite.
So there's all that stuff too.
So is that included with the lowest level is the KiCad?
Do you get that too or is that separate?
No, you get that too, yeah.
Okay, so wow, for $100 you get a lot of video.
Oh yeah, it's like 150 videos right now.
Wow.
Yeah, so it's like 50 cents a video if you're looking for value i
suppose but i mean the real value i think is you know in the material and learning stuff if you
don't if you're just looking for like a price per video then it's like whatever but it's like if you
want to actually learn stuff then i think there's there's even more value than that oh come on don't
you want to put them all in vhs and have like a whole bookcase full of them? Like an Encyclopedia Britannica?
Yes, each one labeled.
Yeah, you get volume one for 30 cents, but then if you want to buy the next one, it's $475 per volume.
Exactly.
Don't you miss those commercials?
I do. I miss a lot of those commercials.
And this, I mean, now I want you to redo the introduction of this as that commercial.
Or how about like, what was his name? Billy Mays, like the ShamWow or OxyClean.
Exactly.
Billy Mays here!
For the low, low price. You can pass this on to your children.
That's right. Yeah, and you know, it's tough. It's tough as an engineer trying to like, you know, because I want this to be successful.
I want people to join and I want people to, you know, participate.
But at the same time, then I have to do, I've done a lot of the upfront technical stuff
and now I had to shift gears into the selling mode and not even selling mode, but just,
you know.
Marketing.
Marketing mode.
Yeah, like you have to do this as a consultant too, right?
I mean, you have to, you could be the best technical person in the world.
You have the best technical material in the world.
But unless you get it out there and people know about it, then it's just going to sit on the shelf.
Well, you have a pretty big network.
And with consulting, we get, I would say, 95% of our jobs from people we've already worked with. So the amount of time I actually go to IEEE meetings or even embedded systems conferences
to get contacts is pretty small.
I tell myself that's why I do it, but really I do it more.
Really, you go for the beer and the hanging out.
I go to complain about other people and complain about companies.
Yeah, so I'm in the same boat.
I go for the dev kits and stay for about companies. Yeah. Yeah. So I'm in the same boat. I go for the dev kits.
Oh, that's true.
And stay for the beer.
That's right.
Yeah.
So you, you did, the more expensive levels have regular get togethers and group discussions.
And that's all through forums and Google chat, right?
That's right.
Yeah.
Like Google Hangouts.
And you are, okay, I'm going to say marketing because we just used it.
You are marketing the social aspect of this as a key component.
But doesn't that require that he stay on schedule with everyone else?
Ah, yeah, that's a good point.
Yes, basically it does.
I mean, it's not necessarily realistic for everyone
i mean and and basically people get lifetime access so if they can't then they they can't
but uh yeah that's that is the idea is to try and stay on stay on topic otherwise it's probably best
to just you know do the audit videos if you're not going to participate in the forums or anything
like that just do it at your own pace and you're going to run this class pretty soon.
I mean, it hasn't started yet.
Right.
Yeah, it starts on the 20th of January.
Okay, so it's going to start really soon.
And then everybody's going to go week to week.
Do you think people are going to say,
wait, we're too far behind, wait a week,
or something like that or i had this online courses
in real time don't go together for me right now yeah no i understand uh we'll see i don't know
some elephant yes yeah i mean um there will be catch-up time between so like the other thing
is like so i split it out into two two courses. That was part of the reasoning was from session 1A,
which is like the design portion, to the session 1B,
where it's going to be like building it and troubleshooting
and stuff like that.
There's going to be a period in there for some people to catch up.
The logistics of this have been a nightmare,
just trying to figure out how to do hardware remotely for everyone.
It's going to be crazy.
We'll see if it works.
And remotely for everyone.
I mean, you spent a lot of time talking to Dave, who's in Australia.
I think you're going to be talking to all over the world.
Yeah, right, right.
My Saturdays for the next eight weeks are well for the eight weeks of the
course so basically i i looked at that i'm like well i have to do i have to do two video sessions
because i'll have to do like a a 10 a.m you know uh basically the u.s will be able to participate
both like a 10 a.m and like a 5 p.m because one will be for the aussies and and that whole
hemisphere and the or sorry no the 10 a.m. will be for the
Europeans. Yeah. Because then it'll be like 3 or 4 p.m. on a Saturday. And then 5 p.m. or so is
9 a.m. on Sunday, the following year over in Aussie land. So yeah, it's fun. I mean,
everybody's got to deal with that these days, though, with globalization and trying to work with remote people.
Oh, absolutely.
You're videoing these live?
You're videoing the recitations live?
Yeah.
What part are you videoing live?
Yeah, like Q&A kind of stuff.
Yeah, talking about the week's videos and and seeing seeing what people
have problems with and stuff like that mostly the social aspect of it um but yeah all of the build
videos are basically done i mean i've i've spent five months making 150 videos so
so but you also have a full-time job and the amp hour isn't a full-time job and this isn't a full-time job, and the amp hour isn't a full-time job, and this isn't a full-time job.
That's correct, yes.
What do you do with your day job? What is your day job?
I work at a big industrial company making electronics for big industrial places.
I don't really talk about which one it is, but yeah, it's analog measurement stuff.
That's kind of my
my mo so big you know analog basically embedded systems um some kind of analog front end with
a to d's and stuff like that talking to talking to a micro talking to a large network that kind
of thing cool yeah how do they feel about you moonlighting uh they would probably be okay with it if they knew about it they know about the amp hour uh the amp hour is a uh that's actually a known event i mean like so i
that was that was actually interesting i i you know went into the interview i started putting
it on my resume actually that was a a bold move but it seemed actually worked out okay
you know talking about it in my interview and, you know,
I mentioned that I never talk about my companies. So that basically, uh,
the company I'm currently working at, I never talk about it. So it's fine.
I guess.
Yeah. I mean, the Amp Hour is really neat. It's a really awesome resource,
but it is, it is kind of hard you know as an employer
sometimes occasionally i have been on the hiring end it would be like well what are you gonna say
about me and that is a little scary yeah yeah it does take trust right i mean you gotta trust your
people but uh it's worked out well i mean i've i've never had any slip-ups i've you know i talk about
old employers now um well yeah they're fair game they can't fire you right exactly and and i try
and keep it i mean positive too i mean because every job has positive and negatives right you
don't want to just just be negative so i think it's i think it's actually could be a good thing
as well so i think that you know i i've talked positively about keefly and samsung as much as i've talked negatively about them so um that's where i used to work so
and why do contextual electronics i mean i understand the amp hour right as as my toes
in the podcast world this is kind of fun i agree yeah but contextual electronics is i mean you're building a little university course out of this
why why good question my wife asked me that a lot what the hell are you doing down there
don't you like me i mean i can see how your wife would be like, what about dinner and the lawn?
Yeah, the dogs.
Yeah, I know.
Fixing the house for the first time in the past four years.
Yeah, no, she's awesome.
So it started, this started on a plane ride to Maker Faire a year and a half ago.
A little bit more than a year and a half ago.
That's actually when electronics started.
That's when the idea came up for it.
You know, just thinking about what kind of, if I was going to start a business, what kind of business would I want to start?
And it's like looking at what was necessary out in the marketplace and thinking about
it.
And then it kind of came to me like
thinking about my own educational experience and thinking about how, how it was really frustrating.
I mean, how, how did I, how did I learn what I learned? And it was a lot of struggle, which is
okay. Struggle, learning is struggle. But at the correlation between my education and my actual work life and how disparate they were, right?
I mean, like, some of it I didn't expect to be the same.
Yeah, I mean.
Right, exactly.
I mean, like, it's such a wide field.
Well, and you learn the theory of electrical physics and not the practice of building products they're totally different
you barely even need to know one that's right the other
okay yeah i mean ideal op-ams get that crap out of here right
okay so you had so this this is why i buy a lottery ticket about three times a year. It's the,
well, if I had all of the money in the world and all of the time to do it, what would I want to do?
What do I want to do next? And then when I don't win the lottery, my question for myself is always,
why am I not doing what I said I wanted to do? That's just dumb. Except for the six months on the beach in Tahiti, I know why I'm not doing that. But all of
the things that are, I want to build X, I want to run a company, I want to do Y, I want
to experience Z. These things, a lottery ticket lets me daydream.
It's, you know, for a buck I get a facilitated daydream that then I need to go back and figure out why am I not doing that? And so you went to Maker Faire, you said, this is what I want to do.
And then you, and then what happened?
Actually, this is on the plane to Maker Faire, so there were gin and tonics involved in a book.
Oh, that's good.
All the best decisions are made that way.
Oh, yeah. Well, it's a long flight, you know. Yeah. And so it was, it kind of sat around for
a while then. And, and it actually in a similar way to what you're talking about. So, so it's,
it sat around for a while. Actually, the first part that came out of it, I had written, I wrote down, it would
be really useful to not only have an educational course around this, but also to have someone
suggest parts that are difficult for me to find as an engineer, right?
It's tough to find out about new parts unless you have vendors coming in and telling you
about, oh, well, this is the new pick part, or this is the new MSP430 part, right?
That kind of thing.
And so that's when I started Chip Report TV, which was a fantastic failure in my first
venture into video.
I stopped doing that after six episodes.
But that's actually the first piece that came out of that daydream, right, of contextual
electronics.
And then basically, I was working, working, I switched jobs to my current job. And then I started talking to a
friend about doing a startup and we started working on a startup together and everything
like that. And basically we started getting it. And, and basically, um, the point where I'm like,
you know, I can't, I can't commit to this because I don't have,
I don't have any kind of other way to support myself other than a full-time day job.
And so then I started thinking, you know, well, what could I, what could I do about this? You
know, what would fit well in with building products on my own? If I want to design my own
stuff and do a startup, how would I do that? And thinking about venture capital, yada, yada, yada, yada, started thinking, well, what if I built this course and built up a library of, of stuff that's
interesting, useful to people, right? A course of, of how to build electronics. And then eventually
that I can help balance that against, you know, things that I want to build on my own and, you
know, do a course one for, you know, know a couple weeks and then do a do building on
something for a couple weeks and then of course for a couple weeks and so really that was kind
of the goal and but the thing that first needed to happen was i needed to build build this course and
so i spent about five months and uh late nights and built contextual electronics
i don't you so so you you want to go you're using this as a mechanism to allow you to do additional
engineering work is that right yeah yeah that's kind of fall in love with teaching now
well that's i kind of figured might happen as well as that the uh the the vehicle for getting
there might also become the end goal and uh
and that has happened i mean i really enjoy doing this stuff and the other benefit is that um
in teaching these courses and building something with people watching um i'm actually designing
stuff i mean so this is i'm designing it's open source but it's uh it's stuff that's useful to
people and it could eventually be sold on its own if i wanted to and uh yeah so it's kind of the means the ends are the means right
yeah sometimes uh i think that's great but how did you did did conceptual electronics fall out of
the amp hour did you really and you And you mentioned this other TV thing.
How did...
Oh, yeah.
Why did you decide to do this instead of consulting or one of the other make money on the side
while you do what you really want options?
Ah.
Okay. So that's a good question. So if we're just talking about money here,
I kind of analyzed it and I ran into the same thing that a lot of consultants run into
and thinking about like, okay, so say I wanted to take a break from working, right? So say I need
to go on a vacation or I want to go work on a side project
for charity or I want to work on a startup, that kind of thing. And looking at consulting, it's
like the best you can ever do with consulting is your hourly rate. And you can charge, you know,
there's 128 hours in a week or 168 in a week, right? And you could work 168 hours times your
hourly rate if you, you know, smoke crack or drank a lot of coffee. You could charge 168 hours times your hourly rate if you, you know, smoke crack or drank a lot of coffee.
You could charge 168 hours in a week, right?
I mean, or something less than that.
Likely, more likely 30 to 40 hours a week if you're doing really well of charge time.
And then you look at, well, where does the value come in?
Sorry, how do you end up making money past your hourly rate?
And it's basically by adding value, like nebulous value to something and then having it easily
distributable.
And basically, that's kind of what contextual electronics is.
It's taking knowledge that I could try and teach someone one-on-one.
I could go tutor someone at 20 bucks an hour, or I could, you know, add extra value
by integrating a community aspect
and integrating actual hardware builds
and stuff like that.
So that's the value added.
And then also it's easily distributable
because it's basically, it's digital content.
And so it's, I don't want to say it's limitless
because there is a limit on the market, but it's, you know, if you look at it from a money perspective, then it's comparing I do get a monthly check and it's been pretty consistent.
And, you know, it's more than going out to dinner.
It's less than I can live on.
But I could see how it's the sort of activity that you get money back for a long time,
where a long time may only be three years but hopefully you will recoup your time investment
over a longer period which then lets you use that it's like you're banking time and and it's
earning interest yeah that's actually a really good way to think about it um but then again like
i said i mean it's not just about the money.
I mean, it's also that it's turned into this thing that's been really, really interesting.
And it's allowed me to kind of dig into what I was always frustrated with when I was learning electronics is, you know, all the stuff that they don't teach you in school and sometimes can't teach you in school.
And I think I'm covering some of that stuff.
Yeah. Money might be an aspect, but the helping people and the getting excited about all the things you can do with this path is, yeah, money ends up being sufficient but not necessary to do anything useful.
Right.
I'm not going to get rich off this thing.
It's going to be, if anything, it'll support me enough to just work on this. And you've put in money for the video gear or did you already have that?
Or do you have to pay your videographer?
I do pay my videographer.
Are you seeking a whole bunch of cash in this?
He's pretty cheap.
Okay.
No, no. My videographer is very cheap. He's me.
Oh, okay.
He's, yeah.
I guess I have seen the...
It's pretty easy once you get into it.
I did see the Blinky, so I guess I could have figured out it was you, but I haven't seen any of the contextual electronics videos, so I was curious.
Gotcha.
So, okay, well, that's really...
But that explains why people should pay you to actually do this because
you this is time this is everybody's like the internet it should be free
what about the online courses the udacity and the coursera those are all free
but they're not really free actually okay so well yeah i mean there is that um you know i and so i struggled about this a lot too because i
always say like you know hardware hardware folks are cheap right i know a lot of hardware cheapskates
because they and and the argument was always the same you know a lot as well oh yeah all the
hardware guys you want to make sure that they're paying for their own lunch, damn it. You don't want to split bills with some
of the hardware engineers. But yeah, it's true because unlike software engineers where
all I have to do is type some more and that's not really cash, hardware engineers actually
understand that if you want to build a product, you have to put money up and yeah they just tend to be a little more skin flinty
sorry phil it's true though phil would have said sorry i know you listen so i have to
make sure to torture that's good i like i like the hidden digs that's good
yeah okay so um yeah so but i i tortured myself over this too like why would people pay for this and
and basically it's it's because this i mean this doesn't exist anywhere else you know you look on
udacity you look on coursera um and some of those are paid courses for like the like the python
classes and stuff like that but But I don't know.
It's just, and the other thing is that you can,
I think people, this time that we're living in,
it's very possible to take this knowledge
and then go and build a widget,
put it onto Tindy or similar
and make money selling those kinds of things.
I think it's very realistic
and it's very
directly applicable to making money as well.
Oh, you could have another tier that says over your life if you sell anything, I get 1% of your total
sold goods. You know, buy stock in person. No, that's silly.
That would need an application process probably.
Well, I don't know.
Sometimes you just win the lottery that way.
You just go ahead and take that from home.
But yeah, you actually want cash now in order to live.
Well, you want cash in six months in order to live on startup stuff.
Have you taken any of the online courses?
The Udacity's or course stuff?
I've signed up for a bunch.
And actually one that I'm actually really excited about is the, the,
the UT,
I think it's UT Austin.
They're doing an embedded course coming up in January.
It's starting,
I think the same day as,
as contextual electronics.
It's a,
what's it called?
I think it's embedded system, something, something. So I think it's uh what's it called i think it's embedded system something something so
i think it's uh i'll find it what is it through so i know yeah it's it's i know it's through ti
so i know larissa over ti and it's using the stelaris uh booster something something and
it's uh but it looks great i mean it's by two of the ut austin
professors and it seems like it's a little more from scratch actually because it's like here is
a resistor you know i still look into some of the intro videos but uh i think it's coursera
and it's a it's a it's a good looking course yeah i i i worked a little bit with udacity we
talked about making a course around my book and around the embed processor,
but it didn't all work out.
There were a few too many people involved.
Oh, really?
It was really neat.
It was really quite cool, the idea of doing the course.
So I'm kind of excited about having a course that teaches
the piece that i am missing which is the i have a concept now how do i go from here to getting
prototypes and maybe the next set of classes will get getting prototypes made here and then
how do you go to china and actually work with manufacturing engineers to put this in lots of stuff and your course isn't going to cover that is it no mine's up through
build your own stuff i mean if you can't build your own stuff you probably shouldn't have someone
in china build it for you because then they send it back and it's all walk or draw and you don't
know what you're looking at yet so uh exactly Yeah, it's all about kind of getting to the build your own stuff point
and then maybe in the future.
I mean, I've built stuff in different countries but never China,
so I wouldn't teach that course anyways.
But do you hope this will lead to more courses?
Or where do you want it to go?
Yeah, I've got – I have a bunch of ideas for other types of courses.
I mean, other things I want to build, other platforms to build upon.
I mean, so this first course is, you know, I say advanced Arduino users.
It's built on top of an Arduino just because I wanted to be able to integrate a common component that people would feel comfortable with.
But, you know, BeagleBone, Raspberry Pi, building on top of other platforms,
because I think that's a very common thing these days
to take an embedded platform
where you don't have to design it in a BGA
or FPGA or anything like that.
Take this commoditized embedded platform,
build something onto the pluggable headers
or however you integrate with it,
and then have,
have a useful product at the end because you actually have all of the,
uh,
it's not just having the input output.
It's also having something connected to that input output.
That's interfacing to the real world.
So are you going to be talking about how do you program the Arduino and,
and how do you write code?
And are you starting at the,
what is a semicolon level?
Or are you starting, you said advanced Arduino.
So I suspect your users should be able to code a little bit.
Yeah.
The real thing, so this is going to be, so the thing we're building in the first session
is called BenchBuddy.
And it's basically everything you need on your bench in order to monitor it.
So there's a thermocouple input.
There's a voltage input.
There's current output.
There's a fan driver.
There's a LED output for lighting your bench.
There's a relay.
There's a big relay for turning on.
So the application that I always kind of say is like you could monitor the temperature of your soldering iron uh with the thermocouple input you could light your bench
and then if you leave if you leave it on too long if you monitor your temperature and your
your iron's on for 20 minutes or something like that you flip the relay and you turn off the power
to the to the soldering iron uh and then there what's the other thing that's on there uh there's
really oh and then there's actually a power output that's on there uh there's really oh and then there's
actually a power output as well so like a regulated power to go to your projects
i can see how i can i i can completely envision all of the software you would need for that i
have no problem with any of the arduino but now you're talking about actual power. And I'm like, yeah, okay. Relays.
I know what a relay is, but I really try to avoid using them because I figure it's just going to cause a fire someday.
Which is how I feel about a lot of electronics, especially those that I built.
Right.
And so it's, yeah, and it's all about, I mean, honestly, this whole course is based around, I remember that
fear, like that, well, what the hell do I do next feeling, right? You know, like that when you're
like, okay, I know I want to build something, but I just, I don't know, do I do this? Do I do that?
And what it ends up coming down to is you hope that you have a friend or coworker that'll sit
down and show you this kind of stuff, right? And that's kind of what the course is. It's like,
I'm that coworker. I'm sitting there showing you how to do layout. I'm sitting there showing you design decisions that you need to
make. And it's stuff that I think about, you know, the co-ops I did, right. That that's a very
helpful experience because you are sitting with someone and you're, you're, you're kind of
absorbing their design decision and their, and their, their thinking process as they think through
these ideas. And you're going to end up using that when you, when you go and make your own
decisions then. Right. And so it's kind end up using that when you go and make your own decisions then, right?
And so it's kind of a viewing of all these design decisions,
processing them, and then being able to hopefully
make those decisions later on, on your own, with more confidence.
You said that this has made you build something
you wanted to build anyway.
Has it also made you more thoughtful as an engineer
as you are designing things? You think
about, well, how would I explain this to someone? I know with coding, that would make me a better
coder. You know, I don't know if it's made me any better. But I had worked with a lot of young
engineers in the past. So I think I may have already gone through that stage of trying to
explain it to people.
I don't actually know if I'm any good at it yet.
I haven't had the feedback yet,
so maybe I'm just not actually good.
But yeah, I agree that if you need to teach someone something
and you're thinking about teaching,
you're going to be much better as a coder
or as a design engineer in general.
And switching topics, this is all in KiCad,
which is the open source CAD tool, right?
That's right.
I wish I could see you nod.
It's natively Linux, but it's been...
Yeah, right. I'm nodding right now.
It's natively Linux, but then it's also ported to Windows.
There is a Mac build that you have to do,
but it's pretty messy,
so I usually suggest that people do a virtual machine
running Ubuntu or something
if they're going to run on a Mac.
That's the only messy part about it so far
is that the Mac stuff is a little less great.
Oh, it's not like firmware engineers use Macs either.
No, I guess so.
Oh, I have a Mac, but I boot camp to Windows simply because I destroy all of the computers and this aluminum has held up very, very well.
But back to KiCad.
So are there other KiCad tutorials that you would recommend?
Yes.
There's actually a couple of good ones.
There's Curious Inventor, Scott Driscoll out in Atlanta.
He's got a great one that actually I learned of when I was using KiCad a couple of years ago when I was starting to learn it.
That's actually what I used.
There's Tiho Labs. There's a couple on YouTube as well that are okay. But it was kind of this
disparate, you know, there's a lot of the beginner stuff. Oh, sorry, Wayne and Lane is another good
one. Those guys have some good stuff as well, but they're all written tutorials.
Some of the YouTube stuff is good as well, but some of it's not so good. Some of it's like the actual actions are great.
They're showing a lot of the important stuff, but they don't talk.
Then some of the stuff is like it's just hard to understand them. I think there's some people, not native speakers,
who are, you know,
they're doing good videos.
It's just hard to understand them.
What other tools would I learn
with contextual electronics?
I mean, what else do I need to have?
Saturn Iron?
DigiKey account?
Ah, okay.
Well, let's see.
So there's a couple things.
Cacad's the big one.
That's the biggest focus because most people don't know that to start with.
There's a little bit with LTSpice that we go over,
just talking about modeling components and stuff like that.
But that's not a very big focus.
I think it will be in future stuff, but it's not a big focus right now.
Is that another open source thing?
And then the physical side of things.
Is it LTSpice?
Spice is open source engine.
Okay.
Linear Technology built a really nice wrapper around the open source engine.
And the program itself is not open source, but the engine underneath is.
That's good.
Okay, what other tools? Sorry.
Let's see.
So the other thing is,
so this is where the logistics I mentioned before
kind of come in.
You know, trying to figure out
how the heck do you get, you know,
50 people working on something in different locations?
How do you how do you try and how do you start troubleshooting?
You know, so, OK, we say, let's go build these parts.
How do we how do we take this board that we've designed and how do we get through to a working a working piece of of gear?
Right. And it's it's a terrible mess. So let's do a thought exercise here.
If it's just me and you working on this board together, right? And I've designed this board
six months prior, and you're going to be following along and building it on your own,
and you're going to have all of these slight changes, right?
Because there might be a different design decision.
You might move a trace as you're routing the board a little bit differently.
You might decide to go a little bit differently.
But basically, it's like a chaos theory kind of thing, right?
I mean, you can have one small change that ends up affecting your entire build. And what the only way I could end up figuring it
out is what I do is as I was going through this, this design process, and you know,
saving different, different versions of my board and everything like that, I needed to say,
okay, you have you have a save game effectively, like like in video games, right? So we can have, so say you're following on a video
and you kind of have a little problem
and then you end up,
it's unrecoverable, right?
You don't want to end up
backtracking your design
all the way to the beginning, right?
And trying to start over
and copy everything exactly.
It's like, no, you need like a save game
where you can go back to that.
And so what ended up happening was I discovered
github and I know you don't like github but I actually love github and so basically control
so it's not the version you're using it for version control it's great yeah right so basically
I have I have a save point after every video I have I have a have a GitHub commit and it's all basically anyone can go and
download it and start from there. And then there's also, you can diff between point A and point B,
and you can see what's changed between the two if you wanted to do it on your own. Or if you want
to just kind of jump back on track to where M in the videos, you just go download that commit number
and you're back to square one or alternately you know someone
can go and make commits on their own and backtrack to their own save game point
no that makes sense a lot of the software courses are done that way because you do build up something
and you you might have a small bug in your version one and it would kill your
your lesson 20 if you didn't and so the teacher puts up the canonical
code every week and uh and you learn you use that which is good because you also learn some
uh subtle design things like naming variables is a good idea they should not be a b and c
they should be like input output and coefficient or something yeah input from fpga to yeah gotta do that
that's my vhdl days coming back to me uh so that yeah that makes sense so that's one part of it
right okay so then you have so then the other part of it is we're going to go and try and build this thing together.
And your design, even so say even you follow along perfectly, right?
And you have your own design.
It looks like it's going to work.
There's the design that I did.
It looks like it's going to work.
Well, then we start building our stuff and we're building up alongside of it.
And both of us turn the switch on at the same time and mine works and yours doesn't and it's like okay well let's go through and then you say oh
well i have a different i have a different pcb and and so and i me is someone trying to help
you troubleshoot it's like oh well crap what were your design decisions here like wait let me see
your pcb files and everything like that and so it's like when you get to the point when you start building if anything's
different that could be what's impacting you right i mean even if it's like trace crossed over another
trace it could be creating these these small variations right you might just get in the
smallest glitch that is a problem on your board but isn't the problem on anyone else's board
and so the solution there was to say okay you've designed
this thing alongside me and maybe you have your own version but i'm still sending you my version
of the pcb and if there's a problem with my pcb we're all going to find it together and we're all
going to the bodge wire together we're all going to cut a trace together we're all going to do a
jump together that kind of thing because there's value in learning that stuff as well. Well, that lets you control the cost of the course.
I mean, you gave the course cost for your participation
and for the tiered levels, but you're building hardware,
so there's going to be more cost.
Right. Yeah, that's a great point.
So using a popular PCB service like Oshpark, right?
So right now, the pcb we're designing
is 15 square inches and it's a four layer board um and if you order that from osh park osh park's
great for boards of all types but if you order that from osh park it's going to be 150 bucks
for three boards if i order that in a bulk order it's going to be about five bucks and so i order
it and i ship that out to people and that's part of the course it's included. And basically that allows for cost
savings for everyone. And it also allows for having a similar, a similar design.
And is that the assembled board or is that the board and a baggie of components?
No, no, that's just the assembled board. Because the other part
that's important about this and important in hardware is the part, learning how to not just
pick parts, but also pay them and handle them when they come in and, you know, learn about sorting
everything like that. That's such a huge learning experience that there's nothing else like that
out there, where you know you have to go and shop around for the correct components uh another thing was shopping when when we were looking for components as well
when i'm doing this design cost constraint was a huge thing not because it wasn't just normal
cost constraints of like oh okay we want it to be low cost it had to be low cost and low quantities
as well which is an entirely different design constraint than than high volume design where
you know you're going to high volume design where you're
going to talk to a vendor, you're going to go talk to TI and say, well, I'm using 100,000 of
these a year, get your list price out of my face, give me the real price for 100,000 units.
And we can't get that as individuals. So we designed for low cost at low quantity.
And then we tried to... So the other thing I did was try to design in swappability, right?
So maybe we're using a LM324, right?
There's parts we can buy from TI.
There's parts we can buy from, I think Linear Tech has a version.
I mean, like everybody makes a version of that, right?
I mean, there's tons of versions out there.
Well, the other problem we have is that because we're all over the globe not everyone's going to buy from digikey they might not buy from
their local distributor they might not be able to get a part because they're in a country that
doesn't have an import on it and so we need to have uh compatible parts as well and so there's
so many learning opportunities in there it's like i that we we had to do that where people are buying their own parts
so so everybody is buying their own parts and you're giving them you're giving them a blank pcb
yep and they're going to take it home and solder it together yep wow yeah it's gonna be intense that kitting process is is tough i mean yes it is worth doing
but being able to maintain for everybody is gonna be tough how much is the kit gonna cost
in the u.s i mean say all i ordered all of the parts you did from digikey or wherever
bomb cost i can pull this up actually.
I have a spreadsheet as every
engineer does.
I believe it was like,
so I wanted to make sure that the final
cost was less than $100.
So if I use the rule of 2.6,
you ever heard that rule before?
You take 40%
for yourself and then 40% for distribution.
Oh. No. You take 40% for yourself and then 40% for distribution. Oh, no, no.
But I do always assume that the bomb is only 30% of what the final product will cost.
Yeah.
So I think I learned the same thing, but I just ignored where it went.
Okay.
Okay.
So, yeah. Okay. Okay, so 2.6, your bomb cost is X,
then your final product cost is 2.6X?
Yes, that's correct, yep.
And so I wanted the final cost to be $100,
and I ended up hitting that.
So what is that slush?
Let's see.
I don't know why I didn't have the final cost here.
That doesn't look right.
Well, let's see. If it's supposed to be $100 divided by 2.6.
I was doing that. Yeah, so $38 for one set of components right now.
I think that's... I'm not sure if that number works out right.
It might be a little bit more than $100.
But again, that's not cost reduced either.
So that's at a single piece price.
So if you're going to buy one op amp, it might be $1.
But if I buy 1,000 op amps, then it's going to be $0.20.
So it's going to be 20 cents, right? So it's going to be different.
If you buy one set of parts,
it's going to be more than if I'm buying 1,000 parts
and then going selling something.
And so I think I actually optimized for the 1,000,
buying like 1,000 parts and building 1,000 boards.
That was the ultimate goal, you know,
if I ever actually sold this thing.
Okay.
And you do have groups, you do have have a forum and there's going to be
some part where you say here are all the parts go buy them and then there's got to be a delay
of a week or two so nobody has to do overnight shipping your groups if they were co-located
could easily uh buy 10 and thereby save probably 10 percent uh yeah yeah are you going to be
talking about doing this in quantity and making them work together to try to get
extra couple bucks out or is that just kind of bonus probably not worth it i mean i mean the
logistics of that are just are not what what's worth it i mean because i mean it's another thing
about hardware right i mean part of part of working on hardware and prototypes is understanding the trade-offs of
low quantity low quantity stuff is going to be more expensive now what i did instead is that i
took so session 1a session 1b they're both priced at the same 75 225 625 right um but what i did is
if you signed up for session 1a i, because you're going to be buying parts
and you might have to buy equipment,
it's only going to be a third of the cost
if you signed up for session 1A.
Because I wanted to make sure that, you know,
we're able to afford the parts and able to build
and do all the other valuable stuff.
Okay, that makes the price a little weirder,
but we'll just link to it and it'll all make sense.
Well, if you signed up for Session 1A, Session 1B would be either $25, $75, or $225.
Oh, so if you sign up for both, it's cheaper than if you sign up for one or the other.
Well, no, it's like if you already chose Session 1A and then you decide to sign up for Session 1B, you'll have like a coupon code effectively.
Oh, okay.
You can just sign up for the next one at a much lower cost.
By two, they're cheaper.
Yeah, effectively.
Because the other thing is,
so you mentioned before too about equipment, right?
So that's another logistical nightmare
because if you have 50 people all build their boards
and then they're like,
well, I'm measuring the voltage here, but it doesn't look quite right.
And I'm like, okay, well, how are you measuring?
Send me a picture of what you're measuring with.
And it's like, okay, well, you're using a DMM from 30 years ago that hasn't been calibrated in 30 years.
That's why your voltage is wrong, right?
And then you multiply that by 50 and that's really difficult and so the other thing is for people that don't already have that equipment or even if they do have that equipment
there's a suggested list of equipment that we're going to be using the low cost dmm low cost
oscilloscope which is the shaky part and then a low cost yeah uh you know a reasonably priced
soldering iron a decent soldering iron so a power supply well the thing is the bench body is a power supply so we don't need that
well yeah that's circular ish yeah but uh well where is it going to get power from we're not
going to deal with all of the different forms of ac power in the world are you no it's actually it
uses an atx power supply so there's a low cost ATX power supply and we designed an extra filtering for that
and then it has a built-in connector and everything. So yeah, that's how it's being
powered. And then basically we take that power, we filter it, we modify it, and then we can actually
use that for other things if we needed to. You've had to do a lot of work trying to make
sure that international is going to be successful. I mean, it was a lot of work trying to make sure that international is going to be successful i mean
it was a lot of work just to make a class and wow yeah i was i was back on my old podcast the
engineering commons and as i said on there there's there's been a lot of like oh crap moments while
i'm like standing in the shower thinking about the course and be like oh how is this gonna work
that's never gonna work
so yeah it's been a lot of thinking about it Wow well I wish you the very
best of luck with it because I think you're gonna need a little bit a little
yeah there's been a lot of yeah yeah there's there's been a good number that
signed up not really saying how many yet but it's been it's been a good number that signed up. Not really saying how many yet, but it's been better than I expected.
Well, that's great.
Yeah.
So I think the amp hour helps.
I think being on the show here might help as well and being on other podcasts.
And hopefully people trust me from the amp hour and such.
But if not, then they probably wouldn't sign up anyways.
Well, you have some credibility you've
been explaining these technology things for a while and sharing your enthusiasm with them so
yeah you you've built up some stock there yeah that's that's kind of the thing is trust right
yeah that's what a book does too that's like your book that's that's a very very good trust building model. Yes. Yes.
Although,
well,
someone,
I don't know if you heard last week's show where we kind of read aloud one
of the chapters and that was,
I was listening to that.
It was good.
I'm not sure the spider should have really been involved,
but
so I have one more question for you and then we probably should get back to our
our regular silly scheduled saturdays of which you have very few left i haven't had any in six
months so yeah um what's with the name why contextual electronics oh um yeah i don't
so i i had bought a bunch of domains and i didn't really like any of them Oh, yeah. I don't.
So I had bought a bunch of domains and I didn't really like any of them.
It's like hardware to software, HW to SW dot com and like all of these other things. And like, I don't know.
It was so basically.
But I always talk about context because it's, you know, you think about how you're how you learn stuff, right?
You're never learning stuff in isolation, but a lot of, a lot of college classes try and
do that, right? Where you're, okay, well, let's look at the math for an ideal op amp or a voltage
divider, right? And it's like, okay, that's good. But unless you know, unless you know why you
should even care about it, right? If you don't, if no one ever tells you the beginning,
you want an op amp because you have this little signal and you're trying to make this big signal
and you want the big signal because it helps when you're trying to put it into an A to D and measure
it and yada, yada, yada, yada, right? And there's all these different reasons you might want to do
something. The why is never, never really a part of instruction. And it's super critical in
electronics because it's, it's difficult, right? You might not know
why you want to be, if you don't know why you want to do something, you're not going to try and power
through when you're learning it. And so that's kind of where the name comes from. I wanted to
be able to explain the why and show this final widget we're building to, right? And because I've
already built it, we can already look at it at the beginning and say, okay, well, this is what we're building to. And here's,
you know, let's actually go through all of the different decisions, the design decisions,
the why we're choosing parts, the why we're deciding to route here and there, the why we're
using open source software, the why we're making open hardware, right? I mean, like, all of these
things, if you don't have the why, then you don't hold people's attention, I think.
And so that's where the name comes from.
Yes, that makes a lot of sense.
Putting why you're learning into the proper context and making electronics not just about electrical theory.
Right. And you know what?
The same thing happens for firmware too.
I always use pointers as an example.
Like once you finally grasp pointers
and not just the how,
because that was tried to,
someone tried to drill that in my head many, many times,
but the why, the why of pointers is just,
that's what allows,
that's what makes you want to use them then, right?
That's what lets you go through all that crap you have to makes you want to use them then right so that's what that's what
lets you go through all that crap you have to do in order to use them and if you're just doing it
from this high level software view of oh well you use a ampersand and an asterisk it's like that
doesn't mean anything it's like no you need to do it because you want what's in that memory location
so that's my that's my firmware knowledge. And done.
Any last thoughts you'd like to leave us with?
I like your podcast.
Please keep doing it for a long time.
Thank you.
Yeah.
It was so good when finally I found out about it.
And there's not enough.
If other people want a podcast too, talk to us.
We'll tell you about it.
Because honestly, there needs to be more technology podcasts. If other people want a podcast too, talk to us. We'll tell you about it.
Because honestly, there needs to be more technology podcasts.
I mean, there's got to be Ruby podcasts.
Ugh, ugh, Ruby software.
Hardware's better.
So hopefully your audience wants to join.
If they do, I think you'll have a link that they can use, right?
Oh, definitely.
There will be a link in the show notes,
or you can just search for Contextual Electronics,
or the Amp Hour has a link. If you can spell it, yeah.
That's true.
Well, the show notes, I promise a link there.
And you can search for Chris Gamble,
and I'm pretty sure that it will come up come up somewhere among his google links yeah yeah it's on the front page of my site so
right number of l's and gammill that's right there'll be links it'll be fine you'll find it
i'm sure and calm down people calm down you're gonna be fine exit to the left link on embedded.fm and you can always
just ask
yep
my guest this week
has been
Chris Gamal
creator and star
of contextual electronics
he's also part of
the dynamic duo
that hosts
the amp hour podcast
where he's invited me
to be a guest
in March
yay
so we're looking forward
to that
yes
thank you for being
on the show Chris
yeah it was great
good talking to you
and thank you all for listening thanks to Christopher White for producing the podcast So we're looking forward to that. Yes. Thank you for being on the show, Chris. Yeah, it was great. Good talking to you.
And thank you all for listening.
Thanks to Christopher White for producing the podcast. And thanks to Confucius for my final thought this week.
Learning without thinking is useless.
Thinking without learning is dangerous.
And while I usually like being dangerous, I think maybe we'll both be thinking and learning.