- If you enjoy listening to MEP, please let others know about us! Tell a coworker, loved one, friend, or share it on “social media”. We might reward your love for us by sending you a free koozie! Tell others how much you love us, then send us an email to email@example.com with the codeword that we’ll mention during the show, plus your mailing address. Each show has a different codeword, so keep an ear out!
- MacroFab and Mouser Electronics have teamed up to create a free monthly meetup in Houston (April 26th) for anyone involved with hardware & electronics engineering or manufacturing. Hosted on the last Wednesday of every month, these meetups are designed to build a community of professionals who want to learn from one another, gain new insights on emerging electronics technologies, and expand their network.
- Sign up here!
- What to expect
- Fireside chats with Q&A
- Individual project sharing and discussion
- Door prizes
- Parker has been working on the next step of the RPI3 LVDS project.
- Pinout the SODIMM connector for the RPI3 Compute Module. See Figure 1.
- Part number 1473149-4 by TE Connectivity
- LVDS DS90C365A IC is net listed up
- Trying to use the PWM function to control the backlight of the LCD.
- GPIO40 is PWM0 on the RPI3
- Still to be done
- Power 1.8V, 3.3V, 12V
- SD card
- USB and a USB Hub
- Break out all the pins
- Test points for clk signals and data streams
- Pinout the SODIMM connector for the RPI3 Compute Module. See Figure 1.
- Stephen has a challenge for the listeners.
- Goal: Design for a ring of LEDs that surround a control knob on a synthesizer and light up according to the position of the knob
- The circuit must connect to a potentiometer (this can be a dual gang version) example: rv16a01f
- The ring of LEDs must be centered on a 12mm radius around the potentiometer.
- Whole circuit must not be larger 1.2” W x 1.2”H x 1.2”D
- PCB can solder directly to pins on potentiometer
- Multiple PCBs is acceptable if needed
- Color of the LEDs does not matter although green or red is preferable
- Number of LEDs is 16 minimum and 32 maximum
- LEDs are arranged on a 300 degree arc (potentiometer has 300 degrees of rotation) starting at 240 degrees and ending at -60 degrees
- The circuit will receive external 5V power so it must have through hole pads for power and gnd
- CHEAP CHEAP CHEAP! Do this for as cheap as possible
- Rapid Fire Opinion (RFOs)
- Revealing Capcom’s Custom Silicon Security – Hackaday
- Capcom developed their arcade boards to die with their secrets through a “suicide” system. System name CPS2
- Basically all the game code was encrypted and the key was stored in volatile ram kept alive by a battery.
- Reminds Parker of the reflow oven that MacroFab currently use.
- NXP chip checks your booze – Electronics Weekly
- The Tag uses NXP’s NTAG 213 Tag Tamper technology which checks the origin of the bottle, detects if the bottle has been opened, and creates an URL for the bottle.
- Stick it on, and if someone tries to remove it, the NFC chip will separate from the tag, rendering it unusable.
- Example Tag that uses the NTAG213
- IoT Startup Bricks Customers Garage Door Intentionally – Hack A Day
- Garadget remotely bricked an unhappy customer’s WiFi garage door for giving a bad Amazon review and being rude to company reps.
- “Technically there is no bricking, though,” the rep replied. “No changes are made to the hardware or the firmware of the device, just denied use of company servers.”
- What happens when this fine firm to go bankrupt? Do all their openers stop working?
- Revealing Capcom’s Custom Silicon Security – Hackaday
Special thanks to whixr over at Tymkrs for the intro and outro!
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 3 00:10
Hello, and welcome to the Mac fab engineering podcast. This is Episode 61, where your hosts Parker Dolan, and Steven Craig, if you guys enjoyed listening to the Mac fab engineering podcast, please let others know about us tell a co worker, a loved one a friend, or share it on social media in quotes yet, right. So we might reward your love by sending you a free koozie. So listen in for a code word that we will have going on during the podcast if you if you hear the CO code word, which we will call out and you email us the code word, we can send you off a koozie some for some mac fab swag. And so the email address is podcast at Mac fab calm. And make sure to include the code word. That's right. And your address that Yeah, that's right. We can't email physical goods yet. Almost there. Yeah, our developers are working really hard at that. So yeah, podcasts at Mac fab calm is is our contact, if you want to just say hello, or tell us about something that you've been a project you've been working on, or suggestions on the show or just you know, say hi, hello. Alright, anyways, so last week, yeah, I was talking about the Raspberry Pi LVTs. Project. Yep. And so I got that working, as I said last week, and I started working on the compute module now. So this is like a step up more, you know, embedding more of the goodies into a PCB basically, yeah. So I, I designed a 200 pin. So DIMM connector, which is what they use for laptop memory. And that's what the, that's the format that the compute module for the Raspberry Pi three is in. And so I built the whole footprint in Eagle and all that stuff made the symbol that stuff at that all done, that must have taken a while. Ah, yeah, a lot of copy and paste a lot of copy and paste. And then I dropped it into the schematic and I made all base I net listed it out. And so that's all done. And I've actually
Host 2 02:28
Spent an hour just reviewing the netlist coming off the sodium connector and verifying with like, three different Raspberry Pi, three sources to make sure it was all good, the official source and alternative source. And then I actually use the datasheet that they use for their, like compute module IO board. I made sure they all matched and everything was good. So we're good there.
Host 3 02:56
And for that, so dim connector, I'm using the TI connectivity parts 1473149 dash four. So, so look it up. If you're in charge, go to browser and type in I want to go see a 200 pin connector. Our listeners are the only people who would actually like go type that in be like, Ooh, what is that? What is that connector? And then for the end that put the LVDS chip on it, which is the DS 90 C 365. A, which I talked about last week. So that's on the board to have that own netlist correctly with the correct pins on the compute module. And it was it was a lot of ease, it was pretty easy to do that from what I currently have, because they call the GPIO pins the same thing on both, except I think it was pins was pins 27 and 28. On a Raspberry Pi three are called they're reserved pins. But you connect to them to output your your pixel clock and data enable for LVTs chip. Yeah. And so I had to actually go and dig and figure out what they actually call that under the hood. And it's GPIO zero and one
Host 1 04:14
Oh well how convenient yeah
Host 3 04:18
Yeah wonder how they chose those one? Yeah, no, it took a bit but basically dug under the hood and found those pins sure and connected those up and hopefully that works. Like the first time around. Yeah, well the first board worked the first time around right how Lucky's just plug it in? Yeah, that never happens. It actually was it was seriously like I like I changed enough of the config dot txt code in the Raspberry Pi to basically tell it to use these pins as the data display interface, RGB 666 mode and I didn't mess at all with the timings and it worked. Like displayed on the screen. I'm like, that's impossible. That's just crazy. Whenever circuit now like the first time I fired up, I just have the expectation that I'm going to have to do something to it. Yeah. But But no, you got lucky, you got real lucky. We're lucky. And all I had to do was basically I went to the datasheet and pulled the right timings and plug those in. And the screen looked even better. Yeah, yeah. So you had to do something to it. But it's just making it better than it. Oh, yeah. It was just code, though. It wasn't like I wasn't changing anything on the board. So yeah, the LVDS step is all netlist set up correctly. That's all good. And the connectors for it. That's all good. In the control the backlight on the on my test board, I just pulled the backlight enable pin tie. Well, I'd like to be able to control it through the Raspberry Pi. So I can dim the display some kind of PWM thing. Yeah, so I'm going to use the PWM function that's in the Raspberry Pi, you can PWM a GPIO port. Nice. And I think it's GPIO 40. That uses like PWM, seven, one PWM pin to to IO that can have dedicated that's like their alternate function. Yeah, yeah. So we're gonna, I'm going to try to use that and then bang it that the work, then I want to write like a script that runs in the background. And so you can use a function key to dim the light on a laptop basically. Right? That should work. Yeah, it should. It shouldn't be too hard. And then other things I need to do on this board as I need to do the whole power circuitry, design all that. So 1.8 volts, 3.3 volts, and 12 volts for the backlight. Oh, that shouldn't be too bad, though. No, I'm the 1.8 volts isn't too bad. It's the 3.3 volt line needs to be hos power supply, because it needs to be able to supply the screen. Basically all the transistors and that, you know, 1366 by 768. Display, how much how much juice is a screen poll about an amp on that line? Quite a bit. Yeah, that's, that's pretty beefy. Yeah, so I need to make basically a beefy linear power supply on the board for that. Yeah. So it's nice and quiet, I need to add an SD card. That's easy, because I already had that footprint built an HDMI ports. And I know I'm happy I have lbds on this. But just in case, I want to use this board for something else. Sure. I have HDMI port, so I need to make that connector, USB connector on it. And then put a USB hub on it. Because the Raspberry Pi only has one USB channel, USB host channel. And so you have to basically put a hub on it hub chip, right. So you can have more stuff connected to it. Breakout all the pins I don't use because you never know. Yeah, and put test points on clock signals and stuff that you want to click onto. Are you gonna break all the pins out to just like through hole? Holes? No, I'm gonna do to, you know, 2.54 millimeter headers. Okay, oh my god, she's gonna break but eventually not populate the headers right? To save cost? Yeah, yeah. But for this first board, this is just like, I'm not going to connect GPIO 40 right to the PWM. I'm going to build this board put together, make sure it see if I can get a PWM signal out of GPIO 40. And then just basically jumper that over to the PWM signal and see like to control the backlight. Oh, okay. Okay, so you make it a dead board? Yeah, I'm making a extensive bike. You know, dev boards are designed to be like all purpose. Yes. Is like there's like, three things I need to test that with this thing. Has a highly specific dev board are three things I need to test. And if they do more, right than I could do the next version, which is basically roles that all into a
Host 3 08:45
Strip, strip everything off and just hard connect debit card Connect. Yeah. So hopefully by next podcast, I have the schematic done. That's a lot of stuff to do by next podcast. Yeah, well, the power stuff most of it's designed in the part wise, because a lot of times most most of the thing is designing the parts. That takes the longest and my opinion for most of my part, you mean like footprints and patterns and things like that. It makes your symbols are set up, right? And all that stuff is good, right? Actually net listing everything together. That's pretty easy. Well, if you have an idea for what you're doing, then then drawing it out is just a matter of just a handful of clicks. Really? Yeah. Handful clicks. That's the week's code word. Handful of clicks. Good word, a handful of clicks. So Stephen, yeah. Stephen, you have a challenge. That's right, based off your synthesizer. That's right. So as our listeners are probably getting sick of hearing I ever talked about this thing for like 30 episodes. Honestly, it's been a lot of episodes. So yes, synthesizer is still doing that. It's still making it made made some good progress. This Week and feature creep feature creep creep pretty hard this week is a bad day. So here's the thing. It's a feature creep, but it's also a, it was a lot of things spawn out of Parker and I just sitting in the engineering department talking about just random crap. So this idea that I have here kind of kind of spawned out of us just just chatting. And then I was like, why not just pose this as a challenge to our listeners, and if anyone is interested or willing to participate, then we can involve them. So what I've been doing in the last like, two weeks or so is kind of building the synth, part by part, but also designing the front panel that the user is going to interface with. And I've spent a good long while making it look pretty cool. And then I had the idea of, instead of having knobs that just point to, you know, dashes around the knobs, what if there was a ring of LEDs around every knob that as you turn them up, the the LEDs illuminate it sweep up, they sweep like a bar graph, but a ring bar graph, and then I quickly sent him a link to a dev board that does this, or yes, the only the only issue with that is that it is these none of my knobs are rotary encoders, I'm not getting digital information off of these, every single one is controlling an analog signal. So the design challenge I have is to create a ring of LEDs that surrounds a analog control knob, which is basically just a potentiometer. And they light up according to the position of the knob.
Host 1 11:47
That's one of those analog control knob. Set protect geometer Yeah, yeah, that's a that's a, that's a that's a wink word in, in the industry, isn't it?
Host 3 11:57
Also, was it? Um, the adjustability of it? That discreet isn't? Oh, yeah, that datasheet? Uh, gosh, that was like, in the first couple episodes of like, three or Yeah, we found a data sheet that said the the, the accuracy of a potential mentor was, and this is word for word, theoretically infinite. Yeah, that's amazing. I bout fell out of my chair. That was the total whiteboard. So okay, so I have a list of requirements here. And we'll post these requirements. If anyone is interested in jumping in and designing one of these things, we would love to see what you have, or just ideas. Yeah, ideas, I mean, so I have a design that I've already basically created for this, but I would love to see what other people have to come up with. So let me run to the requirements real quick, the circuit must connect to a potential ammeter. And a can or most likely will be a dual gang version. So two potential murders. And in one, it's not where you get a clean signal, and you don't interfere with your signal path, right, because the signal that I'm that I'm pushing through is is random. And I need to be able to access one of the potentiometers. So the other potential ometer, that's in this dual gang version can control the LEDs. So the second requirement is that the ring of LEDs must be on a 12 mil millimeter radius, the whole circuit cannot be larger than 1.2 inches by 1.2 inches by 1.2 inches, a cube that size. The PCB can solder directly to the pins on the potential ometer. If so be that's that's easy. can use more than one PCB if needed. The color of the LEDs do not matter. Although green and red is preferable. The number of LEDs across this sweep 16 is a minimum and 32 is a maximum. So whatever works out, and then the LEDs are to be arranged on a 300 degree arc. Because the potentiometer doesn't turn a full 360 It turns 300 degrees. So the lowest potential ammeter is at 240 degrees and the highest potential ometer in terms of fully clockwise rotation. Is that negative 60 degrees is that with a normal Cartesian plane, that's right. That's right. Yeah, no, I chose I chose those because that's what people are probably used to those those degrees. Oh sure. So the the circuit is going to be powered by an external five volt power supply. So all that needs to happen is just some pads on the board that I can supply five volts and ground to and then the last requirement which might be the most important is do it as cheap as possible. So, Mac fab house parts are actually pretty cheap. That's a suggestion there. So if anyone is interested in jumping in on this, they we have the design requirements there. We'll post them up on the show notes which you can get at on Back fab.com you can find in the blog section, there's the podcast stuff. And if you have any questions, you can hit us up at podcast at Mack fab.com. If anyone's interested in designing this with me, I will certainly be using this in the synthesizer. So could be fun. One thing I'm looking at is like minimizing the number of parts. Yeah. And total square inch of fiberglass used would be that we more interested in? Well, yeah, I mean, if we get more than one design, maybe we can go through and kind of pick them apart and see what what people were thinking. Because there's 1000 ways now. Yours is Oh, yeah, mine is mine is garbage. My idea is garbage. Let's just put it that way. Here's the comparator. One. Well, I thought about that, because that wouldn't I don't know I go, I go to the analog solution first in my mind, and I don't get me wrong if I wanted 32 Or I'm sorry, 31 LEDs in in the ring. And that would require 31 op amps. Yeah. And, of course, my mind is like that would be fun, but not practical in any way shape, or take the packages or just take up so much space. Yeah, yeah, no, it's it's a stupid idea. But it's still I like the I like making the OP, like discreet op amps that come in packages that go in like, you see all these like, the DFN style packages that go into cell phones and stuff. Yeah, if anyone makes discreet op amps like that? Oh, I'm sure they do. Yeah, I'm sure a quick a quick search on Mouser would pop up some offense. But when what I found is when offense getting stupid packages like that, there's they're usually really specific and they usually really expensive, like super high fast, op amps or something like that low noise or noise or like their gain bandwidth is like out to 5000 gigahertz or something like that. I know something does another question on this dual game potentially amateurs? What kind of signal cross talk do you get on that?
Host 3 16:57
You know, I think actually the the there is a datasheet that calls that out. So the way that I've kind of designed it is that the signal goes through one of the potential limiters. Yep. And then I have a zero to five volt DC signal on the other. So really doesn't matter. For signal crosstalk. Because you got DC on one side, you say that though, but if someone uses like, a really fast led, no driver that's using a switching frequency, that's going to put that frequency on that five volt line back up and could be injected into your signal. Well, whatever circuit you have to determine where the position of the potential ometer is, should probably average out noise and have some history says it so that it's not, you know,
Host 1 17:43
There's not LEDs bounce all over the place. There's on PWM mobile led, so not all lit up. Well,
Host 3 17:49
I was talking about two adjacent LEDs next to each other. You wouldn't want it bouncing between No, no, I'm talking about just if you have all them lit up. Yeah, it's going to be the driver and LED driver is constant current. So it's turning the voltage on and off to keep that current the same, huh. So you're driving it like, you know, 20 kilo hertz or whatever. Yeah, however fast you need to. I doubt that's gonna be a problem. It's turning on and off. Oh, yeah. Okay, I you know, I did not put that as a requirement, I just realized the current draw on this thing. So, right now I have a plan to put 21 of these into my synth module. I do not want them pulling like an amp each. So let me arbitrarily choose. It cannot pull more than 60 milliamps. Let's just put it like that. So the whole module, all the LEDs cannot pull more than 60 milliamps. But they're all lit up. Yeah, if they're all lit up, the doors do that. No. Mind Mind pulls under 60. So they I just realized that putting that stipulation makes the design way harder. I should have put that in there. Okay. Yes. cannot pull more than 60 milliamps as a whole at any one point in time. So yeah, let us know if you're interested in taking the chance. Do Stephens hard work for him? Well, I've already done it onto the RFO RFO. So the topics this week are revealing cap coms custom silicon security that we found on Hackaday and XP chip checks your booze found on electronics weekly, and IoT startup bricks customers garage door intentionally found on Hackaday so kept gone. Yeah. So this was back in the heyday of arcades so late 80s, early 90s. Yes. So they developed a new the background is they basically developed a new arcade system called the CPS two that had a suicide circuit. And it's so it basically it basically prevented people from tampering with the hardware. Yeah. And if you tamper with the hardware, it basically killed the, the system. Right? Yeah. So and how they did that was basically the game code was encrypted on the on the prompts. And at runtime it would decrypt the game code. And it stored the key I think it was like a 64 bit key or a 32 bit key something like that. Yeah, it stored that in volatile RAM. That had to be powered up with
Host 1 20:41
A with a quartz coins. Oh, better. Yeah. I think it was actually a soldered on lithium battery or Oh, that okay. But a long life like, like a 10 year battery. Yeah,
Host 3 20:51
Yeah. And it also prevented there was a there was a six pin connector on on the board that looks like you know, a programming header or a tag header, right. If you plug it into it, it automatically bricked the system. Capcom was proud of their games. And they did this because back then. It was really rampant in terms of of piracy was terrible in terms of in the arcade industry. Basically, someone would buy one game, rip the EEPROMs off, and then stick the EEPROMs into a cheaper cabinets, like a game that wasn't doing as well. And then slap new graphics on it. And now they get the brand new game that everyone wants, you know, okay, so funny enough, a guy at at my first job, he gave me a Street Fighter two arcade cabinet. Well, I shouldn't say gave his wife said that it had to get out of their garage. And at the time, I had a big workshop. So I said he was bringing to the workshop. So we put it in the workshop. And I played Street Fighter on that a bunch. And then one day, I opened the back of the cabinet. And they said Miss Pac Man on the inside. Somebody completely repurposed the entire cabinet and redid the graphics. Oh, no, they probably just put different problems in it. Or they put new a different board set. Probably, I mean, I looked at the board said it didn't have like socketed Yeah, so they probably ripped it out because it has PAC man's and Atari game. And so they probably put they probably ripped everything out. And they probably use a similar monitor. And similar controls. Sure they just slammed in a street fire two boards. Yeah. And rewired everything. Yeah, that stuff was actually pretty common back then. And so to prevent all that stuff they made the suicide system is what has been called a hardware suicide hardware suicide. And there's a woman about I think fall last year they can't A guy came up with I can't remember the guys name. He found out a way what to use Arduino and a lot of code and some hardware tricks to basically unbrick the system. Basically, you you have to rewrite over some of this stuff and then read put it in a key into the volatile RAM and then bam, you're good to go. How do you know the key though? I can't remember how they figured it out. Interesting. Okay. Cuz that keys stored somewhere on there. Yeah, so and volta RAM, right. I'm gonna yank that out. No, you can't if it's bricked, because that's not there anymore. Yeah. Someone wants to leaked it or something like No, no, they just cracked it. Oh, wow. Okay, so So towards the end of the life of the system around 94 ish. They were they that basically some researchers figured out how to stream the unencrypted code off. So as it was decrypting they were able to read the data in and that's how a lot of basically if your system bricked, you went and bought, I guess pre programmed EEPROMs or you programmed the problems again with unencrypted game code, same game, and you can get your game working again. Gotcha. There's a called Phoenix scene aboard. That's what they call that's an app name. Yeah. But it's just interesting that you know, we don't really have too much of that nowadays. Where it's it's hardware isn't really protected like that anymore. It's more about the software stuff. That's right. Yeah. But the fact that they go that far out of the way especially even with a programming header making that such that it breaks it Yeah, betcha it was like a special you had to do something when you first plugged in with a you know, right voltages or something. Oh, back in the day when you had to like hit it with 12 volts first and then drop it down like negative five or whatever. God I hated. That was one of the some of the first pics were like that.
Host 1 24:52
Yep. Yep. And it sucked. Yep. That's why you had to pay for their like $300 Programming Board
Host 3 24:57
Or E prom or EEPROM programmers. Yep. With those fancies if connectors, so that's crazy. That's, that's crazy to come. But it reminds me a lot of our reflow oven that we currently use that macro fab. It's the Speedline electro vert that we have. Well, the fact that you cracked into it. Well, no, what happened was the, because we bought this oven used, but two years ago, something like that. Oh, yeah. And so we're using three flow boards. And the onboard lithium battery that runs the PLC. You know, just crapped out, it ran out of juice. Yep. And so yeah, basically, one day, we would turn it on, turn it on, and nothing showed up on the screen. And so looking and basically going through the debug stuff on the we got the manuals with the with the oven and stuff. And it said, you have to, he says you have to replace the battery every so often. And it's like, well, we didn't we didn't read the manual. We didn't know that apparently, yeah. So you have to replace the battery every so often. And so he popped the new battery in still nothing. Hmm. So on that in that. So looking into that PLC model, and its ladder logic is stored in volatile RAM that's backed up by that. You lose every we lost everything. Oh, fortunately, after lots of calling Speedline because we kept getting sales people. And they're just they have no clue hang up, and eventually got through to an engineer. And he was like, Yeah, I'll just throw it up on the FTP. And so he through, even through thankfully threw up the the software package that we needed to actually program it to. Because we had to we had to had the the actual configuration data. And we had to have the software to do it. Yeah. And he sent us a link to work by a cable. Wow. Yeah. So here's a nice dude. Yeah. So we got it up and running. Because he knows he knows what it's like to be there. Yeah. But this was it was awesome. Like, well, that. Engineers awesome. But in this situation reminds me of that basically, like, think about if you own one of these and turn it on one day and the battery's dead. Yeah, it's game over. Yeah. So a lot of people on the Hackaday article complain like, oh, yeah, Capcom was greedy. And like, Well, you look at it. They still serviced. Like, sure. They're only greedy if you have to buy another unit from Capcom. Yeah, that would be that would be ridiculous. Basically, the battery dies. You have to buy, oh, you tampered with it yet to buy a new one. Which wasn't the case. They actually service the CP s two system well beyond its original lifespan. Yeah. So they would Unbreak your system for you. I think it was purely just to keep the piracy out. Yeah. Which is valid. But I mean, it is valid where? Yeah, people have the right to protect your IP. But should that impact the longevity? Because now have you shipped the CSP to board to cap on the real like, what is this? But should that actually limit the lifespan of your products? Artificially or non artificially? Well, I mean, it can but it's worth whoever's purchasing it. It's worth them knowing about that before they like the reflow oven, as I said to do it, but it didn't say why it was important. Right, right. And actually, the funny thing is when you go in there by Okay, okay, so this is actually a little dig at Speedline I guess is there is an option to set a voltage cut off on the lithium battery to warn you, but it's defaulted off. Hmm. So if that was on it would have beep that us that replace lithium battery
Host 1 29:11
And we would have been like ah, I'm not gonna replace it right? Why should I replace that thing? Next week? Crap. Oven is down for a week. It was down for four days Oh wow. That's a long time to have a reflow oven done wow.
Host 1 29:28
Yeah, it was way earlier if we were down for days now that would be like everyone would just be like just jumping off buildings.
Host 3 29:35
Oh yeah. Wouldn't not be good. That's our lifeblood right yeah, it's pretty cool. Hit up the link in the podcast description. It's pretty cool stuff. Yep. All the history behind that system is very interesting. I topic to NXP chip checks your booze.
Host 1 29:56
This is pretty cool. It's one that you found. Yeah, yeah. So A near field communication, NFC, NFC technology applied to
Host 3 30:06
Bottles of alcohol. This this was a this was an article that we found on electronics weekly. So the this this was it's sort of a speculation article. It's not like a direct, you know, article saying that this is happening, it was more of like an idea could happen. It could happen. Yeah. So using NSPS. And tag 213, which is near field communication, little antenna, it's similar to an RFID. Chip, right, basically fancier but you ping this thing at high frequency, it powers up over a distance and it and it does some things. But the whole point that this article was talking about was having it such that you can detect, like anti tamper on bottles of alcohol. But the thing about the anti tamper is we already have. So the anti tamper they're talking about is still like basically, when you tried to remove the label, the label gets damaged. Right, and the NFC chip stops working. Yep, it's one of those but we have physical confirmation that this label is ripped, like we already do. So it doesn't actually fix the things. I guess it's harder to replace. Well, and that was the point that you could serialize the NFC chip. And if someone breaks it, well, then they don't know what the NFC serial number one. That's right. And that's kind of what they were going and you can encrypt the serial number. And so that basically no one can copy it. Right. Right. So it's it's a matter of tracking, it's a matter of being able to detect tampering, you know, actually, so what I learned, I didn't know this, but alcohol piracy is a thing. There there is there are companies that own a ballroom That's right, exactly. Literally Yoho and a technological bottle of room. But I don't know, I think I think that's pretty cool. Because so the idea is, is embedding these antennas in the label. You can't you can't tell. But you can scan your bottle and see where it's been who's had it, that kind of stuff. Basically, you'd have to connect some kind of network to be able to do this, but But still I don't that's kind of alcohol. That's right. Cool idea. I always like seeing more of this kind of stuff. Yeah, I assume that we're going to see a ton more in the future. I mean, this is going to be in everything. Basically what they keep driving the price down for this stuff. Eventually, if they can get basically it's only a little bit more than a paper label. There's no point once you have a printer that does both at the same time. Oh man, some pie in the sky stuff. That's right. It prints and embeds this stuff in there. A 3d No, a regular flatbed printer that also lays down like an NFC chip.
Host 1 32:54
It's a fab and a printer. A silicon fab and a printer in one box and one box get on that Parker.
Host 3 33:00
Yeah, sure. Someone billion dollar idea right there. Yeah. You can have it for free. Just remember you got it from macro fab engineering podcast. Given his like acceptance speech for like, Nobel Peace Prize. That's right. I remember the name, the Nobel Peace Prize, because because we can track booze now. Yeah. Last topic, last topic. IoT startup bricks, customers garage door intentionally? This is crazy. Yeah, another Hackaday article. I don't say it's crazy. It's one of those. Another thing to think about when you go to an IoT device? Well, actually, I think it goes a little bit deeper than that. Another thing to, to think about when you purchase something in that, you know, you do not necessarily have control over over your own. Yeah, right. Yeah. Well, you own the physical side of it. But it's actual functionality. You do not own correct. They call it hardware licensing nowadays, right? That's a software licensing. We're like, No, I'm I'm sorry, not licensing. I'm hardware subscription service. Said a software subscription. Like the new Yeah, well, how they're doing with Eagle and Autodesk. And they're all moving to subscription services. Yep. Which is great, because you get updates for everything. I still don't know how it really works for hardware and making hardware better. The software side, it kinda makes sense. Certain software. I mean, I understand why some people don't like it because you don't, you know, you can't keep a version forever. You know, I heard someone argue the other day that video game consoles, you don't actually own them. You're you're sort of renting them and and I was like that, that sounds wrong. That's, I think you buy them. I'm pretty sure that you buy. Well. That was the whole thing when the ps3 got rid of their other OSS functionality. We couldn't install Linux on it. Yeah. That was a whole big lawsuit about that. And whether or not you owned or didn't own the ps3, right, right. But yeah, but it's not it's, I haven't gone to the user license or anything like that. But I'm pretty sure there's nothing in there that says, you are renting this or borrowing it for a cost. I think it's basically you have to you own it, unless you do these x things, and then we can sue you. Yeah, of course. That's whatever user license agreement is. Okay, so what happened in this? So the company is called garage, it was the thing, his garage and gadget garage, that garage, it's something slammed together. Yeah. There was an unhappy customer, who was complaining that his garage door or Wi Fi controlled garage door, didn't work with his iPhone. And he sent very nasty emails to the company, like chewed them out while he was also posting on the forum or on the forums and did it bad Amazon review? Yeah. And basically the company. What sounds like the company was basically fed up with this customer. Yep. And basically turned off his IP off the server.
Host 1 36:12
Right? Yeah. Well, he was he was borderline being a troll. Mate.
Host 3 36:17
I don't I don't know about that. But it sounds I'll put it this way is it sounds like the guy was being, you know, an assets a little unreasonable, low unreasonable, because I don't I don't know if this this this device, this, the Wi Fi garage door thing, even works with iPhones. But if it didn't work with iPhones, and you tried to explain that to an international customer, it wouldn't work. Yeah. Anyways, what they did was also completely unreasonable and terrible. They because they basically blocked his IP. So his garage door couldn't talk to the internet, and so that he couldn't actually use his garage door. Right? Yeah. So he couldn't open his garage door. So you know, and that actually brings up a good point, if you scan it, like, let's say you buy this at Home Depot or whatever, you scan it, or at the time that you scan it and you purchase it. Are you signing a license agreement to say they are allowed to block me at any time that they want? Yeah, it's it's, it's also another thing think about is like, what if instead of this instance, okay, so basically, all they have to do is block your IP. And your garage door doesn't work? But sucks? Yeah, right.
Host 1 37:37
Well, what if the company just went out of business? Now? Everyone's garage doors don't work? Didn't we talk about that a while back? Yeah. With what was it a air conditioning controllers? Oh,
Host 3 37:48
Was it nest? No, it wasn't nest, it was a nut. You're right. It was a no as a home automation company. Right when our business and basically any other unit didn't work, right. They shut down the servers. So it goes back to the point where if you're buying an IoT device, there should be a local network option. Well, at the same time, like be really wary about buying an IoT device for something that's critical critical for your life. I mean, garage door isn't necessarily critical, but it's a really big annoyance. Most people use at least twice a day. Right? leaving and coming home. Yeah. So it's, it's pretty, it's pretty high on the list. It's pretty high on the list of stuff that should just work. I mean, what if what if you had IoT controlled locks on your house, on the server, and the servers were in the company was like, sorry, when they went out of business, you know, or whatever, just Comcast just had service interruption. Right, right. Yeah. And you really have to go to the bathroom,
Host 1 38:45
Outside, and then there was an IoT toilet that just would not open and it could not flush. And it's just oh my gosh, no, that I get all this IoT, crap,
Host 3 38:54
Man, toilet, and the service went up the only service that allowed the flush.
Host 3 39:00
And so that's horrible. So So the moral of the story is not the moral the once again, Parker and Steven shit all over IoT. Like that's, that's that's like, how stuff would be on the show. Yeah, exactly. I don't know. I think that's kind of crazy, though. I mean, giving giving someone else the capability to shut you down at any time. If they don't like you. I mean, he was being a nuisance. Let's be honest, this guy was still isn't deserve you to take away the guy's product. Right? Right. Right. I mean, he owns the hardware. But at the same time, the funny thing is the that what they did was they shut down his access to the server. That doesn't necessarily stop him from posting online reviews. He makes it worse. Yeah. So I don't know what they were thinking here. Yeah, I don't know. I don't think they were thinking, No, that's, uh, well, I'm not gonna install it. IoT garage door opener on any of my Will you ever install an IoT Berberine setup? Only if I have sole capability of controlling it? Yeah. Then the internet goes down and you're like no and I'm like turn on propane. Yeah. Propane in a bathtub. Ah, so that will wrap up the RFO for this week. Yep. And with that, that was the McWrap engineering podcast. This was episode 61. We are your host spark don't. And Steven Craig. Later everyone take it easy.
Transcribed by https://otter.ai