- Parker ordered all the parts for the SAIM this week. The v-slot rails and brackets from OpenBuild,iES-1706 servos from Leadshine, and 3D printed parts from Shapeways.
- MacroFab uses 3D printed parts to make custom spacers and tooling for assembly of PCB boards.
- Parker will be adding acceleration to the motor controller code to make sure the SAIM does not jerk around too much. Maximum Smoothness as Stephen puts it.
- Stephen has been working more on the FX Dev Board. The board has a couple different power rails. It has +15VDC, -15VDC, an adjustable +9VDC and a half rail that tracks the +9VDC rail. The half rail uses a voltage divider for the reference voltage and is powered by an opamp. See Figure 2.
- The 16in/16out board that Stephen made has a 16bit A/D and a 16bit D/A on it. MAX5217BGUA+ and MAX11100EUB+. One uses I2C and the other uses 3 wire spi which Parker thinks is unfortunate. See Figure 2.
- Stephen has some old mylar 60’s era caps that his mentor in junior high gave to him. They are 500pF caps.
- The Super Simple Power Supply has not had a lot of work done on it this week. Parker has started working on the front control panel. Block diagram of the design is done. See Figure 3.
- SSPS with IoT? Who would follow a twitter account of a power supply? Parker would.
- Big Ben clock in London is tilting? Parker measures the picture the newspaper posted and it shows the tower tilting at 3 degrees instead of the 0.26 degrees the engineers say its tilted.
- KORG and Noritake release the Nutube datasheet. Stephen is stoked to say the least. It is not vaporware as Parker thought.
- Parker suggests they design an iPhone case that has one of these tubes in it for a headphone amp.
- Raspberry Pi 3 was released this week. Built in WiFi, Bluetooth 4.0, and a beefy 64-bit ARM A53. Runs at 1.2GHz. Parker says if a computer can’t run beefy web apps it is worthless. RPI2 was enough power.
- Stephen likes DOOM PI
- Students at MIT develop a kind of Sensor Tape. It has a ATmega328P, a couple sensors, and blinking LEDs. LEDs are important says, Parker. Arduino friendly and all those things. Parker is going to look into what it will take to make some.
- Hernando Barragán wrote an article about the early history of Wiring and how it led to the Arduino environment. Parker really likes the hardware aspect of the article and what could have been for Parallax and Microchip if they had open source tool chains in the mid and early 00’s.
Special thanks to whixr over at Tymkrs for the intro and outro theme!
About The Hosts
Parker Dillmann is MacroFab's Co-Founder, and Lead ECE with backgrounds in Embedded System Design, and Digital Signal Processing. He got his start in 2005 by hacking Nintendo consoles into portable gaming units. He also runs the blog, longhornengineer.com, where he posts his personal projects, technical guides, and appnotes about board layout design and components. Parker graduated with a BS in Electrical and Computer Engineering from the University of Texas.
Stephen Kraig began his electronics career by building musical oriented circuits in 2003. Stephen is an avid guitar player and, in his down time, manufactures audio electronics including guitar amplifiers, pedals, and pro audio gear. Stephen graduated with a BS in Electrical Engineering from Texas A&M University.
Host 1 00:09
Hello, and welcome to the macro fab engineering podcast. I'm your host, Parker Dohmen.
Host 1 00:14
And I'm Steven Craig.
Host 1 00:15
So I would like to thank all our listeners. So far, it's been a pretty interesting five weeks so far, I personally have never done a podcast before.
Host 1 00:25
This is the first for me to,
Host 1 00:27
It's been a lot of fun. The feedback we've been getting from our listeners has been great. And I've been hoping we get more feedback so we can make this show even better.
Host 1 00:35
Yeah, thanks a lot for all the listeners, guys.
Host 1 00:39
So during this week, as I mentioned, in my last on our last episode, I've been working more on the semi automatic inspection machine. We got the parts ordered from open build, I ordered some servo motors from lead shine. And these aren't your typical hobby servos. They're ginormous. They're like steppers except I think their internals are slightly different. And they allow for a lot higher resolution and positioning. And they do have a higher achieve at a higher RPM of rotational speed. And they have a lot flatter torque curve. So it's better for a high speed movement, and that kind of stuff.
Host 3 01:21
Well, the application for this is, at least for this model is going to be an optical inspection, right?
Host 1 01:27
Yeah. And actually servos aren't actually really good at that, because they work by sampling its position. And so it'll basically rock slightly back and forth. So you can stay on. It's its location, its
Host 3 01:42
Average position is in the middle of where it needs to be. Yeah.
Host 1 01:45
And so it's kind of the wrong use case. But I basically looked at that sheet and saw how much it would move. And it's actually under the deflection of the belts. We're using Jeetu GT to timing belts. Yeah, that's actually under that deflection. So you actually won't see that and the camera, right, just just the natural stretch of the belt. Yeah, we'll
Host 3 02:09
Take care of that. Well, and the thing is you onto the trolley, the thing you're actually moving is basically just a camera, so you don't have much of a load. So you can probably get it going pretty quick.
Host 1 02:20
Yeah, it should. It should fly. Yeah. And so the the servo part number from lead shine on that is the IES 17 06 Which is actually kind of new. It's an integrated servo encoder and driver. So basically, you just send it pulse signals for direction and pulsing for each step. And it takes care of the rest. It's pretty cool. Yeah, pretty cool little little guy there.
Host 1 02:49
You know how much those costs.
Host 1 02:51
They were like 150 bucks all in. Okay, per unit, so I had to get three of those. And then we got some 3d printed parts. I ordered those from Shapeways. The I basically I printed them on my home printer first, but I wanted a higher quality print. Yeah, and decided to go to a professional service to get that done.
Host 1 03:13
Well, I guess for people who don't know. What is Shapeways?
Host 1 03:17
So Shapeways is a 3d printed or a 3d Basic company that will print your parts for you. It's good way to put it.
Host 1 03:26
Sure. But they they have pretty high quality machinery.
Host 1 03:30
Yeah, they don't use a I think the FDM printers they have they have a powder printers. So your quality, you don't get layers and your prints. And you get a very solid piece of hunk of plastic that's almost injection molded quality.
Host 3 03:47
Yeah, we've actually used them for some of our jigs that we use for building customer. Yeah, that's a macro fab and then they come out great.
Host 1 03:56
And we've had custom spacers for basically through hole parts that keep them a certain distance off a PCB. Yeah,
Host 3 04:03
Sometimes we have like a height specification for for components. So we made made some spacers. And those worked out great. Yeah,
Host 1 04:10
Those were great. And basically, for those of you put them underneath the part it holds it a certain distance before the selective solder solder isn't right. Because before on that, that certain project we were actually hand soldering those those parts on, which took forever.
Host 3 04:26
Yeah, yeah. And then you get to measure that by hand and it's not very efficient.
Host 1 04:31
Yeah. And then the and then I've been working a little bit more on the firmware code base than any start adding acceleration. So right now there's a linear ramp acceleration. It's mostly implemented. I'm gonna work on it.
Host 3 04:47
Are you wanting to change that from linear logarithmic or exponential? Or
Host 1 04:51
I think eventually it's, I think it's gonna be linear at first and then I'll add a an exponential.
Host 3 04:57
Yeah, linear is probably more than enough. Yeah, linear
Host 1 05:00
Is probably enough. But once you get to a certain speed, you want that. You want that exponential actually deceleration. Acceleration doesn't really matter too much. I've seen in positioning systems for cameras. Sure. It's mainly getting that deceleration. Right? And so you don't get a jerk when it actually stops. So you don't have to wait for your chassis to settle.
Host 3 05:24
Yeah, yeah, I got you. I, it's, that's the way to make the maximum amount of smoothness. In in the travel.
Host 1 05:32
Yeah. Um, so Steven, I so I've been working on that, what do you been working on?
Host 3 05:39
So this week, I've actually been touching more on my FX dev board. We kind of mentioned it a little bit on the last podcast, but can go a little bit more into depth. So So I've designed a development board, which, which is basically just a PC board that that includes two solderless breadboards. It has, it has a some variable power supplies on it, which it's a plus minus 15, then a variable nine volt supply. So it goes from 1.25, up to nine volts. And then it also has a split rail on there. So basically, it takes the nine volts and cuts it in half, in case you want to use single supply for op amps and things like that. Yeah. So all of that is on board and readily available and accessible through the rails on the on the actual breadboard themselves, I have jumpers that you can select the voltages that go to all the rails on there. In addition to that, you can run it off a nine volt battery, or, and it has audio connectors on it. So you can the intent was to make it a development board for audio applications. And even a little bit more towards like guitar amplifiers and guitar pedals
Host 1 06:58
And synthesizers, synthesizers and things like that and that kind of stuff. So basically,
Host 3 07:02
You can literally plug your guitar directly into my dev board and then plug the dev board into a guitar amplifier and then build a stomp box or build a preamp or, or build whatever you're looking for. And all the peripherals are right there, including actually onboard potential amateurs that you can connect directly to the breadboard. So it's an all in one package. effectively.
Host 1 07:24
So Steven, so you mentioned that split rail, it's like it's plus minus 15 volts, right?
Host 3 07:30
Well, that's so you have the option of doing plus minus 15. But pretty much every stop box guy from now until you know back to the the 60s have run stompboxes off of nine volts. Yeah, nine volts. So So you need a virtual ground. That is that is halfway between nine and zero. So it has a 4.5 volt rail. But that 4.5 volt rail tracks the nine so so it's always half, it's always half. That's cool. Yeah, it's always half. So if I have a variable, nine volts, so you can reduce that nine volts to kind of simulate a dying battery, not a stompbox guys love to do that kind of stuff. But saying if you wanted to run your whole circuit off of five volts, and have a two and a half volt rail, it automatically takes care of that for
Host 1 08:14
You. So do you do that through a you do that through a voltage divider? Or how are you doing that?
Host 3 08:20
Yeah, so so it's honestly, it's a voltage divider that goes into a buffer op amp.
Host 1 08:27
Okay, so that's how you get the the voltage divider isn't providing the current, it's just providing the voltage reference, I prefer
Host 3 08:33
To do it that way. So that you have the really high input impedance of the op amp. Yep. And then you can drive, you know, 10 milliamps or so off of the half the half the half rail. Yeah. So so your nine volt rail has, you know, up to, you know, probably 102 100 milliamp output capacity, but the half rail isn't meant to source
Host 1 08:54
And yeah, it's just it's basically a reference. It's a voltage,
Host 3 08:57
It's a virtual ground. It's what we what I typically call it, okay, so that you can run, you know, AC signals on single supply. Okay, cool. Which which is really popular in the in the stompbox. World, and it's right there if you need it on the board.
Host 1 09:13
No. And you were using a year. Earlier this week, you were also working on actually building a circuit on that dev board. Yeah, I'm using your Mac's I think 1110 something 16 bit ADC? Yeah. Yeah. Well, okay, so much DAC. So,
Host 3 09:32
Now that I have my FX dev board, I I'm starting to kind of develop a handful of example circuits that go on to this. And I've always wanted to have a, an outboard is what I call it. In fact, the name of this one is 16 in 16 out, it's literally just a daughterboard that has a digital analog converter on it and an analog to digital converter on it. It's the interface in between the analog world and the digital World. So that's all it has on it. You can plug whatever your, your favorite micro into it and and be able to get analog signals in and spit analog signals out. Yeah. So yeah, I chose some some, some maximum chips because, well, they're, they're cheap, but they have really great specs. And, and it's, like I said 16 bit and 16 bit out. So that worked out really well. to interface with an Arduino. Actually, I was just running an Arduino to can to control the 16 and 16 out. And what I was building on the FX dev board was actually a function generator that is stuck in a phase lock loop. Okay. So it's a sawtooth generated kind of what you would see in a synthesizer that we generate an analog sawtooth wave, convert that into a square wave, and then feed that back into the, the Arduino. So the Arduino will spit out a control voltage to the function generator, and then read back the square wave that comes in and compare the frequency of that square wave to what's coming in on what's being spit out. And it automatically adjusts such that the frequency of the output is exactly
Host 1 11:19
What you asked for. So you built in some feedback to that. That's
Host 3 11:23
Yeah, exactly. And it's all running with the 16. In 16. Out digital in outboard. Yeah, so it's kind of a marriage between the the analog function generator capability, so I can easily get higher frequencies out of it, as opposed to just doing a wave table and spitting out to my DAC. Yeah, I kind of, I thought it'd be fun to do the the analog side of things. So I've been playing around with that and getting not significantly high frequency outputs, but But honestly, I was just trying to shoot for the audio range. So I've got up to 20 kilohertz at the moment coming out of it. And with with some tweaking on the circuit, I was actually able to get it up to about 170 kilohertz,
Host 1 12:08
You were actually I'm also a basically tweaking it for making sure your edges are sharp on that triangle wave, because you are getting this weird flat section. So that wave Yeah,
Host 3 12:20
So So when it comes to a, an analog function generator, trying to produce a sawtooth wave, you have to have a linear ramp. And the goal is as fast as possible to take that ramp from whatever your max voltages down to zero. Yep. And in the analog world, they just don't like to move that fast. So so it's it was a whole lot of play in with switching in caps. And effectively what you're doing is you're you're shorting the output of, of the op amp, yeah, into the input. And so picking the right transistor and picking the right drive signal to that transistor and all the all the timing and feedback. It takes a little bit of hand selection. Yeah,
Host 1 13:04
The first first breadboard that you did have it was just wires everywhere. It's like a was like a rat nest out of a fishing rod. Yeah. And eventually you had it all nicely laid out with flat wires and everything. But he had you had two caps that still had their trims on Yeah, and there were these big box foam caps I think. Yeah. And and actually the mylar Oh, mylar caps. Yeah. And, and I actually mentioned I'm like, Hey, so you know, you probably clean up that signal a bit more if you read the source leads a bit. And I you said there they were came from your mentor. Yeah, so these are like the old the old caps, these
Host 3 13:45
Caps. They were manufactured in the 60s. They were the old red mylar caps that have they have six dots on the top of them. And basically the color of those six dots determine their value and their tolerance and stuff regardless, when I was I was a junior high at the time and I was getting into electronics there was there was a really cool guy I knew who was he was way into guitar and way into doing all the all the guitar amps stuff. And he gave me these two capacitors and I'm holding on to him and they're like, and there's nothing there's nothing unique about them. They're just 500 pico farad mylar cap. Yeah, but I've never trimmed leads. I'm not gonna solder him into anything, but they're in my breadboard.
Host 1 14:29
Yeah, they're there they are some of the most tarnished leads. Oh I have ever seen horrible they're not they're not 10 solder or solder anymore. They are. They're gray and they're not shiny anymore. Yeah, just the patina is very nice.
Host 1 14:46
Hey, they're sentimental man. Yeah.
Host 1 14:48
And one more thing about that about that that in outboard? Yeah, um, it's kind of a personal question at least on my end is so those those chips I think the most fit 50 to 70 and the max 1110 things what they were Yeah, those run on i square C. Actually
Host 3 15:09
The let me the D to A runs on I squared C, the A to D is SPI
Host 1 15:15
Oh, that's unfortunate. Yeah, it's
Host 3 15:19
Just how the game is played with these two.
Host 1 15:22
Yeah, it would be nice. If those both were I squared C. And then you just basically put them on both on the same bus. And then that's so you only have two data lines going over to your microcontroller instead of four. Yeah, or I'm assuming it's two wire spires or three wires by three. Okay, so five wires? Yeah.
Host 3 15:39
And thing. The one thing that I find annoying about this, and I, I can't stand when maximum does this. Both of those come in their specific you Max package?
Host 1 15:51
Yeah. It's like really why T stops basically what it is. Yeah,
Host 3 15:55
Yeah. Yeah. Well, we're in a really fine pitch. Yeah. So it's, it's kind of annoying. But hey, I had him I had a made a macro fab. So it worked out pretty well.
Host 1 16:05
Yeah. How much that whole board cost?
Host 3 16:08
I you know, I think overall, the for that entire daughterboard. I can't, I don't think I spent over 10 bucks on it. And it's pretty crazy. Yeah, that's, that's cheap.
Host 1 16:18
So the last thing we've been working on this week was the more work on the super simple power supply. Not a lot of work done on this week. I started designing the front panel. But we did do a block diagram, finally, of all the subsystems and how everything should line up. What should be connected, I think the next step on that block diagram is actually write a pseudo circuit level design. That's just a rough outline like so sure. The digital signs got to control the analog side, but how does it control it? Do we need to use digital pots? Do we need to PWM certain pins? Yeah, that kind of level? It's kind of wide open,
Host 3 16:59
Because we have so much we have anywhere we can go with it. We were talking about? Do we just spit out high voltage levels from the from the digital side of things? Or do we spit out low levels and apply gain in our output stage? There's just a lot of variables to consider.
Host 1 17:17
And then we were actually talking kind of jokingly, at first, I think we were reading some article about the whole Internet of Things is not secure, and whatnot. And then we came up with the idea of actually adding Twitter capability to this power supply. And so it would tweet, like every five minutes or every minute or, you know, however, how long have what the power supply is doing. So it's like I am outputting five volts at two amps. And so it'd be kind of silly, but then and then you could have it like broadcast failure modes. Like if someone actually tripped the current production on it, it would automatically send out immediately sounded like so and so messed up their circuit and blew it up.
Host 3 18:03
We should have it we should have it where we have to log in with a password to says so it knows which one of us would accidentally trip it or short the box. Yeah. I think it'd be pretty funny. That would be that that's okay. So it's those kinds of things that makes the Internet of Things kind of ridiculous. Yeah,
Host 1 18:23
Um, because who's gonna follow a Twitter account of a power supply?
Host 1 18:26
You never know there might there might be some
Host 1 18:29
That I would follow it
Host 1 18:32
Is already one.
Host 1 18:35
Host 1 18:39
So on to the rapid fire
Host 1 18:40
Opinions, the RFO. So, the first article was from the metro UK, which is I guess it's a newspaper over in London. And basically engineers claim that Big Ben, the big clock tower is tilting, and this is kind of a civil engineering thing. Not really our specialty. But it's just you know, it's interesting that a tower that big is tilting, I guess. So it says it's a tilting 0.26 degrees, which doesn't sound like much, but it's 1.5 feet from vertical. Interesting thing about that is they're in in UK, which is on the metric system, and the newspaper was reporting it in a imperial units.
Host 3 19:31
Okay, actually, the interesting thing to me here is the Big Ben was was built in 1856.
Host 1 19:40
Do you actually know that date?
Host 3 19:42
I read it, okay. No, no, I didn't just know that off the top of my head. He did the research. I did a little bit of research on this. No, no, no, that the biggest thing that that that hits me on this is it was built in 1856. And in 2016, it's, it's leaning over by point two, six degrees. I'd say they did a good jab, they did a good job, I'd say they do it and
Host 1 20:02
Now actually attributing this tilt from recent construction, and they put a subway line recently in that area. So the fact that it probably wouldn't have been tilted at all, if London didn't keep getting bigger and more
Host 3 20:16
Advanced. So the soils just moving on. Yeah, well,
Host 1 20:21
Whenever you do anything underground, it's gonna shift soil around that's been, you know, compacted for two centuries now.
Host 3 20:27
Yeah. That Beckett, back in college, the the civil engineers used to always say, water plus there equals mud. Yeah. And that's the only thing you have to know in civil engineering.
Host 1 20:42
But also interesting thing is they have a picture of the Big Ben tower. And you look at the picture and it looks really tilted. You're like, wow, you know, that tower is crazy tilted. And but they have a quote from an engineer that says, you have to look at the tower the right way, and then you can slightly see the tilt. So I'm like, I wonder how angled that picture is. And so I actually opened it up in paint and and drew a line and then moved the line till it was parallel with the tower. Those guys doctored that photo, or tilted the camera or tilted the camera, any anyways, that picture shows it tilted at three degrees, which is actually less than the Leaning Tower of Pisa. The Tower of Pisa is four or so degrees tilted. Wow. So no, no undue alarm there. Well, they've been tower
Host 3 21:36
That sounds like fluff journalism. I just needed something. Big Ben bleeding. Yeah.
Host 1 21:44
Yeah. Okay, so this is one that's gonna hit close to your heart. Yeah. Is the nor talkie and Korg a long time ago, said they were going to make these new dip style tubes called the new tube, then that's and You Tube. Yeah. And YouTube, not an E W. So that was released, I think, a year and a half ago, something like that. And it kind of went dark. And for my mind, it was kind of like, oh, they probably release it. No interest. And so this became vaporware. But apparently, recently, a actual data sheet came out of this tube.
Host 3 22:28
Yeah. So they, gosh, it was the NAM show, which is the big audio shows that happen every year. They they, they broadcasted that I while ago, like a year and a half ago, something like that. And then like you said everything went kind of radio silent for a while. And just a few weeks ago, they they released the datasheet. And it's freely available to download as long as you give them their your contact information. But But regardless, I've been waiting for this for a while because I'm a tube nut. I love vacuum tubes. They're fun. I've been doing them for years. They're absolutely ridiculous. But it's it's really cool to see that after 50 years, there's a slight deviation slightly new technology has slightly Yeah, no, it's actually no The funny thing is it's not even new technology. No, literally just a different package.
Host 1 23:25
Yeah, it's actually what looks like a nor talkie VFD package. Yeah. With
Host 3 23:31
They just got it an OLED display. Yeah. And put put a tube in there. Yeah.
Host 1 23:35
Well, that was probably the cheapest they can do for tooling. Oh, yeah. Yeah. But so the thing about this, so this is the interesting thing I think about tubes is or it compared to modern transistors and stuff. So if you pull up the datasheet for two n 3906. It's like it's still about 20 pages long a datasheet for a single transistor. Yeah, this tube datasheet is two pages. Yeah, and the second page the second page is half of its mechanical diagram. And then there's a chart the the anode characterised anode characters, so and then the front page, half of it is description. And then you have three tables of actual data. Is this typical tubes?
Host 3 24:23
What's what here's the thing that's awesome about it, if you go pull up an old German to from 1940. The data sheets are just like this to like they stayed true to the old, the old school data sheets. So this is very typical for a sheet. Except the funny thing is with with old two data sheets, typically what you find is this and maybe one more page that would show a typical application.
Host 1 24:52
Okay. They don't have that in this actually. I don't think there's even a pin out. Now there is not a oh yeah, there Yeah, in the middle. Technical diagram, they got a pin out
Host 1 25:01
It just it's a bad one.
Host 1 25:05
Yeah, it says pin one, F one, pin two, F one, pin three and P.
Host 3 25:10
Right. In other words, if you didn't, if you had had no experience with tubes whatsoever, their pin out would make
Host 1 25:17
Zero sense. There's not even a schematic diagram for this thing.
Host 3 25:21
So yeah. Yeah, no, it's not it's not super well done.
Host 1 25:26
No. But it is exciting to see something like this come out. Finally. It's yeah, it's, it's, it's cool
Host 3 25:33
In terms of the tube world. So as a kind of a an overview of what this guy is. It's just like, every old school tube like 12x, seven little bottle tubes that you see out there that you can use for low level signal amplification. It runs identical. It's a triode configuration. In fact, inside the package, there's two separate tryouts. Okay, so it's basically think of it this way. It's kind of like two J FETs. In in one box. Okay, cool. But the thing that's, that's cool about it is it uses, like, 1/1000 of the amount of power then that a typical tube does.
Host 1 26:14
Yeah, that's actually interesting thing is the first sentence, it says can operate at low voltages, and with low power consumption. And I actually drew in here, an asterix for tubes.
Host 3 26:26
Yeah, tubes were never really known for their
Host 1 26:32
Power savings, power saving. So
Host 1 26:34
Are there a fish? Do
Host 1 26:35
They have a power saving pin that you can pull? Oh, that kind of like, chills, chills now? What
Host 3 26:40
Did you just say polo on a to? Parker is a very digital guy. Let's just, let's just put it that way. But okay, so what's cool is this guy runs on significantly lower voltage, typically, your two circuits are going to run in the two to 400 volt range somewhere. That's a wide range to 400 volts. Yeah, somewhere in there. I mean, it's up to the designer, you can you can pick whatever you want. Yeah, this guy runs in the 10 to 80 volt range. So it's still by today's standards, that's high voltage, high voltage, but for Two Guys, that's, that's sacrilegious voltage.
Host 1 27:17
Yeah. And the current seems to be pretty low on it to
Host 3 27:22
Very low, very low. Well, okay, the filament. So just the the, the current required to heat thing up is 17 milliamps as compared to 300 milliamps with its predecessor. Yeah. So
Host 1 27:33
I wonder if I guess they probably changed either materials or something inside of it, to enable that filament heats to be I guess, more efficient at heating the tube up?
Host 3 27:44
Well, there's that and I and the, the actual elements they're using are significantly smaller. Okay. And the spacing between things is also much closer. Yeah.
Host 1 27:56
Probably because they have better manufacturing nowadays, compared to the 60s. The design these tubes. Yes, they actually look at the dimensions of the sky. It's 46 millimeters by 17 millimeters. And what, you know, that's still pretty big. But the big thing for tube is it's only 5.6 millimeters thick. Yeah, so
Host 3 28:15
We can go into a real small package. Yeah.
Host 1 28:18
I'm thinking like, cell phone style size. You know. preamp for your for your cell phone. Yeah.
Host 3 28:29
Yeah, for the for the audio enthusiast.
Host 1 28:32
Yeah, an audio enthusiast iPhone case. Where all
Host 3 28:36
There streaming all their audio digitally across the airwaves, and then putting it
Host 1 28:42
Out of your outputs through the jack on it. Yeah. into one of these tubes.
Host 3 28:47
Well, I guess as long as it touches the tube somehow, then it sounds good,
Host 1 28:51
Right? Yeah, I guess so. But, um, so it says the new tube delivers exceptional linearity. So is that true?
Host 3 29:02
Actually, okay. So according to the anode characteristics, the linearity is awesome. Okay, this this thing is actually, it looks great. In terms of linearity. Most of its other stuff, its transconductance is pretty low. Its gains pretty low. It's anode. Resistance is really high. All of those are not good things. Yeah. or I shouldn't say not good. But they're, they're not as they're not very desirable.
Host 1 29:29
Yeah. And then, even when using distortion circuits has an impressive response distortion tones that retain the characteristic pleasant overtones of a vacuum tube. So can you actually measure that kind of stuff?
Host 3 29:45
I I'd have to pull up my my BS meter and how much is going off on their
Host 1 29:51
BS over dB? Hmm,
Host 3 29:53
Yeah, that's that's exactly. marketing hype. I think that's what we can chalk that up to.
Host 1 29:59
And you know, it's It's made by not what's gonna be made by a nor nor Taki. So, um, is gonna be made in Japan. It is probably gonna be a good tube.
Host 3 30:08
Yeah, I mean, I'm excited. We I've actually contacted the guys and they got back to me. And we can we can expect to see these available in singles sometime this summer.
Host 1 30:20
Oh, that'd be cool. Yeah. Should we I think we should definitely build something with it. Oh, I'm
Host 3 30:23
Getting guaranteed. I'm gonna get a couple of these. Yeah.
Host 1 30:28
So this week, the Raspberry Pi three came out. And that's kind of a big deal. So we went from a tube to something that's the size of a tube that also runs at 1.2 gigahertz though.
Host 3 30:40
Yeah, a little bit more technology in in the in the PI three.
Host 1 30:45
So the most important thing about the Raspberry Pi three, at least in my opinion is it's got built in Wi Fi. And built in Bluetooth. Yeah, that's really cool. So basically, you can hook up your, your keyboards and and connect to the internet without having to waste your precious, four USB ports that the Raspberry Pi two had some other things. It's got a beefier processor on it, which personally I thought the Raspberry Pi two was enough processing
Host 3 31:10
Power was something like was it 900 megahertz? Yeah, it was
Host 1 31:14
900 megahertz. I always thought the Raspberry Pi one was a little weak. And especially because Raspberry Pi one chugged with Internet browsers. Yeah. And they're like, if you can't access an internet browser, I'm pretty sure that computers worthless nowadays. There's so many apps and so many sites, basically are so heavy that you you kind of need a little more oomph. Yeah, some grunt. Yeah, to actually do that.
Host 3 31:45
Well, I was reading up about it and and everyone's kind of touting this as the first Raspberry Pi. That actually counts as a real computer. So
Host 1 31:55
I thought the Raspberry Pi two was enough.
Host 1 31:58
I used it for a while myself, it worked. worked fairly well.
Host 1 32:02
Yeah. And I think the great thing about the Raspberry Pi two is it was the first time I mean, I did over the Raspberry Pi one but it was really slow and chunky. But the Raspberry Pi two, you could SSH over with x 11. And so you can actually, you know, basically SSH tunnel your your desktop in. And it actually was pretty snappy. And you can actually would, you could play simple video games actually. Over an SSH tunnel. And it was fast enough to do that. Um,
Host 3 32:32
I did the whole Doom Pi. Pi. Yeah. Doom on my PI. And it worked. But man, it was brutal. And play was it was laggy
Host 1 32:44
I have never tried that. But yeah, it worked pretty good. So yeah, it runs a new arm. 64 bit ARM I think
Host 3 32:53
1.2 gigahertz with a gig of RAM. Yeah, the
Host 1 32:56
Raspberry Pi two was a giga RAM. But uh, yeah. What was interesting is comparing this guy to like the other system on chip boards, because there's been a lot of basically ever since Raspberry Pi one came out, there's been a huge influx of the system on chip cheap computers. One of the more popular I guess, maybe not popular, but well known would be the CHIP computer, which was the first $9 computer. And then right after that was the Raspberry Pi Zero, if you can even buy one because I've never seen one for sale. $5 Yeah, computer, supposedly $5. I've never seen one for sale. And then there's the pine 64, which has a Kickstarter. We'll see if they even deliver because Kickstarter is usually don't. Oh, joy, which is a reputable company that makes system on chip boards, like the Raspberry Pi. There's a lot of them out there.
Host 1 33:55
Yeah, they're starting to pop up all over the place.
Host 1 33:58
So you have all these, I guess what you would call Raspberry Pi possible killers. And the interesting thing is when people compare your product to Raspberry Pi killer, or it's an X killer, that really just solidifies the dominance of whatever it's supposed to kill. Like how dominant the Raspberry Pi platform is. In its circled space.
Host 3 34:22
Yeah. The thing about it is though, it kind of seems like any of these guys could have been the Raspberry Pi.
Host 1 34:30
Yeah, he's just who came out first. Yeah. Who's adopted?
Host 3 34:33
All of you know who owns most of the market? Yeah.
Host 1 34:38
The it's really the markets. I actually, I don't think it's more market space. I think it's the community building. Who did a better job at grabbing the community first and building off
Host 3 34:49
And pies completely? Oh, yeah. steamrolled everyone.
Host 1 34:53
Yeah, if you look at I think chips got a pretty decent one. And we'll see where Pyne 64 goes and Oldroyd seems to have in more technical space, I guess. They don't really, it seems they don't really gear to the low end.
Host 3 35:10
Yeah, it was something I noticed myself with with the pies that I wish was better is better documentation and easier access to the GPIO pins. Because they're like, you've got 40 GPIO pins, good luck. You know, it's it's not as simple as saying.
Host 1 35:29
So it's that simplest way is you write to a file. And then eventually the GPIO changes. Yeah. Right. There's no low level stuff.
Host 3 35:40
The first time I heard of, of Raspberry Pi, I was like, Oh, my gosh, this is blowing my mind that I can have an O S, and actually access the the hardware pins direct at me that that was amazing. And then I realized, it's not that simple, not that directly.
Host 1 35:57
And it might have changed since last time I looked at it. But that was how it was when I was looking at
Host 1 36:02
It, it's probably easier now.
Host 1 36:06
So going on the MIT or some students at MIT, they developed some something that they call the sensor tape, which looks like a roll of masking tape. It basically is a sensor network that's on a tape that has a 80 Mega 328 P, it's already obsolete, because they're using the 328 P and not the 328 P b, which we actually covered earlier in one of the podcasts. It's got a couple sensors on it, and some blah, blah, blah, it's got LEDs, and that's important. It's accessible over Bluetooth and all other Wi Fi stuff. But the interesting thing about it is basically they printed, I think it's about a one inch by one inch square, and all of them think it's more like an addressable microcontroller strip, or an addressable IoT device, kind of like an addressable led RGB strip. Hmm, that's what seems like to me in addressable intelligence. Yeah. It's kind of weird. Um, it's one of those also like, what kind of use cases they're looking for for it. They say that it can, if you attach it to something, it could, it can figure out the shape of the device says attached to Whoa. So it can figure itself out in 3d space of what it's what it is. Or at least the shape of the tape. I guess. That's kind of cool. Yeah. So I can see that if you put it on, like a diaphragm or something, then, and then it lets the diaphragm expands. You could read that or wrap it
Host 3 37:49
Around a pipe and you could get the information on the diameter of the pipe or something like that. Yeah. It's interesting. Yeah, that's, that's kind of cool. Well, I guess you just stick it wherever you want and run power to
Host 1 38:00
It. Yeah. And it communicates over that. Oh, it's kind of cool. Yeah. And apparently it's Arduino friendly code. So as is everything Yes, everything that comes out. They have their their thesis I think it's a thesis online. I didn't see any design files in it, but I'm gonna keep digging a bit and see if I can actually find something and build one of these strips. It looks like it's on a flex circuit. So it's gonna be kind of interesting. We couldn't be made. Yeah. But well, I'll give it a shot. See if we can build some disor goofing off.
Host 3 38:35
Actually, that brings up something talking about arduino real quick. I saw something earlier earlier this week where a guy submerge an Arduino Uno in liquid nitrogen. Oh, yeah. And he overclocked it up to 65 megahertz
Host 1 38:51
That we that's over three times as fast as it normally goes. Yeah, yeah. Yeah. So they have I guess if you overclocked it. So at least in the FPGA world or in an actual my Controllers The faster you go you eventually start running into basically your i o can't switch fast enough you start running into the actual electrons can't You can't switch them off fast enough or on fast enough and so you start getting clock slew on all your data. Yeah,
Host 1 39:24
Right. They electrons don't like to turn corners that fast.
Host 1 39:28
At least on and off fast. They like to keep flowing. Well, that's inductance for you,
Host 3 39:33
Right? Well and that's the whole thing the Arduino the the I O ports. They were never designed to go anywhere near that fast.
Host 1 39:42
Yeah, so I wonder if if I guess I'll take a look at that what that guy posted in CV actually posted the like the waveforms of the pin the GPIO on it. If not, I guess that'd be pretty cool thing to try out is actually a couple take a couple of different microcontrollers and basically just make them output basically just a square wave as fast as they can. Janice main loop words on off on off all the all the just do like order a sequence of the pins, sure. And overclock them in liquid nitrogen cooled them and see how fast you can get to the point where you can't actually either till the point where the microcontroller starts function stops functioning, or you don't get a usable waveform out of the IO, huh?
Host 1 40:33
Yeah, that'd be a fun little project and fun little project. Oh.
Host 1 40:37
Speaking of Arduino though, I think it was came out today there was a blog post by Hernando Barragan that's an Italian name. And I'm, I'm a Texan. That's the best I can do. Or Nando Hernando. Um, so this guy is the original writer of the writing library that Arduinos libraries are based off of. I did not know that the wiring
Host 1 41:08
Library. Yeah, wiring like that, right.
Host 1 41:13
And so this article goes into all the politics of Arduino and all that stuff. The arduino.cc Verse arduino.org. Personally, I kind of ignored all that stuff, because I don't use Arduino stuff. But was interesting thing about this article, was he details and actually great detail, the history of the hardware that they chose initially. Yeah. All the revisions that they went through. Yeah. And, and also going through the mindset of what they wanted this project to be in the end, which is really cool.
Host 3 41:52
It's actually kind of interesting, because it seemed like they had a vision for what they wanted it to be. Yeah. But large portions of that vision was kind of cloudy. Because they they didn't know that people were going to adopt it. as hardcore as people have. Yeah. So it's kind of everything seemed a little bit trial and error, where it's like, hey, we'll build this board. And then people use it. Like, this is absolutely amazing. Let's make another hardware vision. Yeah. So
Host 1 42:22
So I guess the the first prototype, which was a shock to me, was it ran off the parallax Javelin stamp microcontroller, which was, that's an unusual choice even for 2002 I think is one that happened. Yeah, yeah. But looking back, I guess actually, in that time period, the stamp actually pretty the basic stamp from parallax actually dominated the the low end hobby market pretty well. Yep. Because back then you basically had a VRS pics and that, yeah, those were the hobbies. And
Host 3 43:00
Mainly because those guys didn't You didn't have to buy a $500 programmer.
Host 1 43:04
Exactly. Yeah, you didn't have to buy this giant programmer to program your controllers and it didn't need weird voltages the program
Host 3 43:11
Right You didn't have to pull something at 12 volts for a period of time and then bring it down to five and
Host 1 43:16
Host 1 43:18
But they did he ended up not using it Hernando did end up not using it because it had proprietary tools. And one of the things in his thesis is about how the entire project needed to be completely open source from the get go. Yep. At least all the software cuz I'm pretty sure you can't get any of this stuff. The internal hardware stuff at all. Like the actual masks that make AVR or or pic or whatever.
Host 1 43:50
Well yeah, I mean, then somebody is gonna make money on it. Um,
Host 1 43:55
But yeah, so the second choice was they started working with a pic and it was the same problems at the time microchip didn't have an open source solution for the tool chain it was all proprietary tool chain.
Host 1 44:06
I wonder if they're kicking themselves for that now? Yeah, it's interesting.
Host 1 44:12
If because it's what's interesting is is microchip now has an open tool chain and Prop has an open tool chain
Host 1 44:18
Probably be clear locks prop have this kind of history. I don't know if it's
Host 1 44:23
Maybe microchip is flex. I wouldn't know but I'm pretty sure prop parallax propeller is open for different reasons cuz it came out that open came out like recently within the last two years, okay. It's not like it was like 2005 And they did it. Okay, okay. I don't know when microchip opened theirs up, either.
Host 3 44:43
So the first revision was a parallax parallax stamp. Yep, a javelin. Then pick was the second Yep. Both of those didn't cut it because they weren't open. Yeah. And
Host 1 44:53
Then they went to Atmel. And they use a giant. I think it was a 1891 Are 4008 Yeah. And then they eventually settled on the 80 Mega 128 to make it less expensive. And that would be the original hardware, so to speak, the wiring library, right? And IDE because I think there was an ID based off of processing. The processing is an ID.
Host 3 45:22
It's in parallel with processing. Yeah. If you've ever used processing, it's it's kind of like computer programming using an Arduino language. They're so similar.
Host 1 45:31
Yeah. And it's actually interesting. He also goes into detail about Brian Dean's work with who's the guy who behind AVR dude. And that actually was coming out around the same time does this stuff too. So that'd be like, 2003 2004.
Host 3 45:45
Right? And just icing on the cake for AVR. Guys.
Host 1 45:48
Yeah. And the crazy thing about that, though, is the Arduino IDE still uses AVR. Dude to this day. Yep. To upload to uses, like, I think the original one, use the SDK 500 protocol. And now that AVR do supports a lot of different protocols. But yeah, interesting that it's a cool progression to see that happen. Yeah, it's really cool to see this this history from the perspective of as students who wrote the original wiring library,
Host 3 46:20
Actually. So on top of all that, I'm sure this I believe this is available on Hackaday. This this whole article here?
Host 1 46:28
No, it's actually on his own website.
Host 3 46:31
Oh, okay. I'm sorry. So on his own website, if you go there, his master's, master's thesis, it's on there as well is on there. So if you want to read it, it's it's large, it's huge. But you can read the basically the conception you can go through all the way through all of his thoughts on wiring and how he got his master's in engineering through this. Yep. So it's a it's a pretty cool read.
Host 1 46:56
Yeah. Well, I think that's gonna I think this was a long episode, a lot of content. But you know, last week's was a little short, so we kind of tried jam packed a lot of stuff in this one. So I guess that's gonna do it up for this this episode. Awesome. So I'm your host, Parker Dolan. And I'm Steven, Greg, and catch us next week for the next macro fab engineering podcast. Take it easy
Transcribed by https://otter.ai