This episode suffered from a recording glitch so only half the audio exists. Sorry :(
Parker is an Electrical Engineer with backgrounds in Embedded System Design and Digital Signal Processing. He got his start in 2005 by hacking Nintendo consoles into portable gaming units. The following year he designed and produced an Atari 2600 video mod to allow the Atari to display a crisp, RF fuzz free picture on newer TVs. Over a thousand Atari video mods where produced by Parker from 2006 to 2011 and the mod is still made by other enthusiasts in the Atari community.
In 2006, Parker enrolled at The University of Texas at Austin as a Petroleum Engineer. After realizing electronics was his passion he switched majors in 2007 to Electrical and Computer Engineering. Following his previous background in making the Atari 2600 video mod, Parker decided to take more board layout classes and circuit design classes. Other areas of study include robotics, microcontroller theory and design, FPGA development with VHDL and Verilog, and image and signal processing with DSPs. In 2010, Parker won a Ti sponsored Launchpad programming and design contest that was held by the IEEE CS chapter at the University. Parker graduated with a BS in Electrical and Computer Engineering in the Spring of 2012.
In the Summer of 2012, Parker was hired on as an Electrical Engineer at Dynamic Perception to design and prototype new electronic products. Here, Parker learned about full product development cycles and honed his board layout skills. Seeing the difficulties in managing operations and FCC/CE compliance testing, Parker thought there had to be a better way for small electronic companies to get their product out in customer's hands.
Parker also runs the blog, longhornengineer.com, where he posts his personal projects, technical guides, and appnotes about board layout design and components.
Podcast Transcript
Host 2
Hey guys, Parker here. We had some issues recording this episode. Basically by the next day, we realized that the entire our first section and Stephens, part of Steven section was missing. So this episodes only about 14 minutes long. Again, sorry about that. There's not much we can do. We're kind of ran out of time to, you know, re record this. Good news is the Star Wars Christmas special thing that we did last week, there's actually going to be a video up and hopefully it will be up today on YouTube. And so I'm hoping that will make up for this short episode. Catch on next time. Hello, and welcome to the macro fab engineering podcast. We're your hosts Parker Dolan
Host 3
Host 2
Welcome to Episode 47. Yep, first episode after our ginormous Star Wars episode from last week. It's kind
Host 3
of crazy to go back to just audio. Yeah. And
Host 2
well, actually, the video is not out yet. Hopefully when this comes out. The video is out on YouTube.
Host 3
Yeah, well, last week was a bit of a production a little a little a little bit.
Host 2
Anyways, let's go right into what projects we've been working on the past two weeks. Yeah, we didn't talk about last week. Yep. So the Jeep brand new updates. I've designed the circuits that mean, Stevens talked about last last time. Basically, it's a it's an op amp that reduces the gain by half mm DC offsets at the 4.6 volts, right in the middle of the voltage range of the that that I see that on the radio users? Yeah. And then I basically ordered it through McWrap. I think it was like $19. Nice. Yeah. So it's pretty nice.
Host 3
It's basically a textbook example of a level shift in a game change on an op amp.
Host 2
Exactly. And then only difference is, it's got a analog switcher by maxium. I think, Max 45440 Yeah, that's right. Yeah, I'm actually was playing with that on a breadboard and was get was able to switch the audio out of the Jeep radio. Nice with that. So that was working. And basically, those are on the board. And so it's an all in one little boards about an inch by inch and a half. And it's all on one side with a ground plane on the bottom. So it should be actually the quietest part of that radio. A good place
Host 3
to have it quiet though. Yeah. Cuz you're, you're pumping in your signal from there. Oh, yeah. And
Host 2
there's gonna be, you know, wires coming off the board into this board. And, you know, trying to reduce noise as much as possible.
Host 3
Yeah. And then you kind of crossed off.
Host 2
Yep. So that's ordered. Hopefully, within next two, three weeks, I get my prototype back and test it. And hopefully it works. And then I can finish finally finished that project. It's been like two months now. And then I started designing an amplifier board, or an amplifier for spooky pinball. Yeah. In my free time. And the the, basically, the big problem with what they've been having is supply problems with their current amplifier. Yeah. And it's not a very good amplifier here, either. It's a 2.1. And it's rated at 60 watts per channel, which is a bunch of bullshit. Yes, it
Host 3
Host 2
I'll put it this way. It uses a 12 volt, three amp power supply. So there's no way you can get 60 volts.
Host 3
It's just Yeah, that's the easiest way to tell. It's not even possible.
Host 2
Yeah. And so we're actually designing a true 1515 Watt per channel. Yeah, yeah. 15 Watt per channel. is a power supply. And we're gonna be supplying it with 24 volts that like six amps, and it's gonna be a hos. an amp
Host 3
basically the way it should be done. Yeah.
Host 2
And it's going to have a it's single supply, but it'll, it'll have a separate rail, all that good stuff. It's based off the TDA 7387, which is a basically a ginormous four channel power amplifier for car audio. Yep. Basically, like a lot of like Ford and Chrysler and all those guys. They use this chip in their main stuff. So cool. Yeah, it's a four channel. I won't be using three channels. But I was able to design the board and get it kind of put together. I haven't routed it yet. But it's running. It's gonna run around about 15 to 16 bucks and quantity 500 With just the board assembled, yeah. And then I got to put up a A big heatsink on it because we got sink like 80 watts.
Host 3
Well, right, yeah, max volume and max volume and dumping out a lot of stuff. Yep. And then
Host 2
I need to be able to spec the power supply. And that's the trick is we're trying to make this whole thing under 35 bucks. And I ate 15 of
Host 3
that. Well, okay, so 35 bucks gets you the amp, the heatsink and the power the power supply, but not the speakers, not
Host 2
the speakers. Correct. Okay. Because the power of the heat sinks gonna be about 10 bucks. Yeah, for something that's big enough. The problem is the power supply because I only have like $10 left. And I can't really pump out that much power for 10 bucks. So the bill of materials is going to have to, well, there's no way you can shrink anymore. There's like no extra stuff. And like, everything that's mission critical is specked out should and everything else is like use the cheapest possible on all that stuff. Yeah. Yeah, it. I'm hoping next week, I can place an order for it. So I got routed over the Christmas break. Well, that's
Host 3
a fun little project. Well, it
Host 2
gets me away from the family.
Host 3
Classic engineer. Yeah. cuse me. I can't come to Christmas dinner. I have to router an amplifier. Yeah, this is really important. On the on the gonna go on a lot of pinball machines
Host 2
on the couch with my laptop open, just like, you know, rounding the board, ignoring the you know, Christmas parade and stuff that's partly on Gmail TV.
Host 3
Oh, well, I don't blame you in that case. Yeah.
Host 2
The only thing I need to add to it is a a speaker level input. So it's got line level input right now. But they want to add speaker level inputs. So they can basically sell this to pinball users that have
Host 3
older machines, where they can compute power even more. What do you mean by speaker level? And
Host 2
so all the old machines have like integrated sound, a board that already has the game applied to it to go to a speaker?
Host 3
Wait, that when you win the game, you mean like, the the the level that would actually be at the speaker? Yes. So So in other words, it already has an integrated amplifier?
Host 2
Yes. But the problem is, a lot of them are basically really weak. And they want to be able to drive their speakers harder. Ah, okay. So they want to get rid of their old speakers that are, you know, weaksauce put in badass speakers, and then put this amp in, in between.
Host 3
Okay, okay. So okay, it was something that goes along with that. So the topology of how the amplifier actually operates depends, there's, there's a handful of different ones, there's a, b, b, c, h, d, there's a ton of different topologies that you can choose. Now, some of them work fine with an infinite load effectively, yeah, you can take the output of an amplifier and put it into something huge, all the way down to about four ohms. And you find now when you're talking about speaker level
Host 2
input, yeah. So it's going to have to look like to the old system, like a four ohm load.
Host 3
Well, does it? And the reason why I'm questioning it, because if if those old machines have amplifiers that can push an infinite load effective. I don't I don't think that you don't think so that overheat? Yeah. Okay. Because because a lot of times I've done this in the past with, with amps, you can you can have, like a line level output that's in parallel with the speakers. So the speakers still see the or the amp still sees the load of the speakers, but then you get a signal output in parallel with and you can drive that into anything else. Yeah,
Host 2
I think the best, you know, it's going to have to look like 408 ohms. At the ask, you know, what, you know, what can we put it might be we have to have a toggle switch on it or jumper, the switch referring for an eight ohm load. And then the we got to be able to have a way to make sure that it doesn't clip coming in. Yeah, so we don't overdrive our
Host 3
AMP. Sure. Sure. That's, that's
Host 2
the easy part. I think I think the hard part is, at least for me, because I have no idea how to do it is make our AMP look like a four eight ohm load.
Host 3
Host 2
load. Excuse me, an infinite load like it is now but it's driven by line level and it goes right into an op amp is the first thing that happens well, you
Host 3
know, give it a shot because then you might be able to get away with that. You might it might be you might be able to work it out. I don't know. Cuz you see the typically a power amplifier is a current amplifier, it's not a voltage amplifier. So you're not really juicing up or gaining voltage. In other words, your signal doesn't appear to get larger. It just provides more power. Yep. So you got to kind of watch out for what's going on there. Yeah, cuz I'm
Host 2
mostly worried is we design this thing. And then someone opposes weights, a lot of people who own promotions have no idea how to fix them. So they put this thing in, and, you know, they're not comfortable soldering wires or anything. So it's all gonna have to be like, you know, plugging connections, right? Well, of course, yeah. And if they put it in some machine that we have no idea what it is. It could be like an old Gottlieb that uses a different power amp than anything else we've ever tested. And it blows it up. That board set that's in that machine is probably like one out of like, maybe the last 500 that exists. Yeah. And there's those parts were in like the late 70s, early 80s. You can't get a replacement part. Hmm. So I was like, Okay, if we just made this thing look like a speaker to the system, we solve that issue.
Host 3
Well, you could always just put a giant resistor. Exactly on the front end. That's that's the, that's almost guaranteed to be the cheapest way to do it.
Host 2
And the thing about it is this is a this doesn't have to be on the app most of the time, we can just leave it on pops
Host 3
for when it's for their. Their units. You don't need that at all. Yeah, okay. So just leave pads for like a 25 watt resistors, something like that. Yep. Okay. Should work. Sometimes it's not necessarily the most elegant solution that works. It's just the cheap and dirty one.
Host 2
And it will probably work the best. Yeah, yeah, exactly.
Host 3
Host 2
one part. Cool. So that's all I have ADD
Host 3
so far. Well, so in episode number 45, the one two weeks ago. Yep. I brought up the idea of doing a calibration resistor. Yeah, basically or making our own. Yeah, we need to come up with a name for
Host 2
Host 3
Host 2
Yeah. Working work. And yeah.
Host 3
name isn't working. Okay, so Tweet us,
Host 2
if you have a good name, hopefully will actually explain what it is. So they have an idea of what the name is.
Host 3
Okay. So we'll go through some stuff pretty quick here. I don't want to get too bogged down in math here. But let's let me let me make it let me make it really simple here. Take Take four resistors. Okay, all the same value, if you have two of them in parallel, okay. And then you take the other two and put them in parallel. And you take those two sets, and you put those in series. Okay. Okay. So, effectively, what you have is a two terminal device you have basically you've made a resistor that is made up of four resistors. So I'm looking after our manufacturing operations directors dog, and he just jumped up onto my lap. Hey, how's it going, Billy? Okay, so that's where all the sounds coming from. Okay, regardless, back to resistors, the cool stuff. So okay, if you take these, these resistors, and you make them all the same value, you can actually create a component, you can create a or composite component of multiple components that has better specs than any of the single components by themselves.
Host 2
So basically, the tolerance starts to decrease
Host 3
correct, while you keep the same value of the component. So effectively, you can add up to infinite amounts of resistance to get a resistor of any value you want, effectively, and you can have you, you can have control over the tolerance of that resistor. So what we were thinking was creating a calibration resistor, effectively, it's a PCB that just has a boatload of resistors on it, that are in series and parallel configurations such that they all act as one resistor, but with a really tight tolerance on them. Okay? So think of it like a an array of resistors, where you have columns and rows, okay, if all the columns in this array are in parallel, and then all the rows are in series, and the this array has to be a square, so yeah, so the number of columns has to equal the number of rows. If all of the resistors in this array of columns and rows, if they all have the same value, the resulting resistor that you get from this entire array is the value that you put in. So if you have 1,000,000,010k resistors and you put In this array of rows and columns parallel serial, you'll still get a 10k resistor but the tolerance gets better for every resistor you add effectively. So
