#115 – An Interview with Dr Greg Charvat - Watcher of Wraithlike Walls

Chris Gammell: This is the Amp Hour Podcast, recorded September 30th, 2012. Episode 115, with guest Dr. Greg Charvat, watcher of Rafe-like Walls.

Dave Jones: Welcome to the Amp Hour. I'm Dave Jones from the EEV blog.

Greg Charvat: And I'm Greg Charvat from Butterfly Network Incorporated.

Dave Jones: Hey, Greg. Thanks for joining us.

Greg Charvat: Oh, well, thanks for inviting me, Dave. It's great to be here.

Dave Jones: And let's make fun of Chris, because you said his tubes are still warming up in his lab there.

Greg Charvat: Yeah, and his router. He's waiting for the tubes to warm up. That's why he's not on the air yet.

Dave Jones: Yeah, he's been having serious issues. He's trying to dangle his mobile phone out the window, trying to get a cell connection, as it's called in the US, and then get a Wi-Fi repeater into his dungeon basement there. And we don't think it's working too well. So Chris may or may not appear for this episode. Feel free to jump in, Chris, if you're there. No. No. Fail. All right, Chris, you're from where? Tell us the story.

Greg Charvat: Okay, so let's see. Where am I from? Well, I just moved to Connecticut, okay? Why Connecticut? Well, I was working at MIT's Lincoln Laboratory for four years, and I was recruited aggressively to join a startup in Connecticut. Aggressively? Yeah. As a second founder. Did they send guys around in trench coats? Oh, no, they didn't. That was my last job. There were guys in trench coats. But no, this one, we had, people have been reaching out to me for various startup opportunities, as I'm sure they talk to you as well. And I was, I can't reveal who our investor is, but he's a fairly famous guy. And he wanted to bring me in. And it turns out he and a student of mine started this company together. And the student was in my radar class at MIT. So they, you know, from that class, you know, they wanted to bring me in because I build stuff fast. So that's my shift.

Dave Jones: Ah, that's your speciality, huh?

Greg Charvat: Yes, I make things very fast. And so they want to build a series of prototypes to test a concept, which I can't talk about, unfortunately, because we're working on IP stuff. But, yeah.

Dave Jones: I'm surprised that Valve haven't snapped you up, because you're aware that, like, I think three or four of our guests have ended up at Valve. Really? The software company, yeah.

Greg Charvat: Yep. Do they make hardware there?

Dave Jones: They've got a hardware group. They've got a hardware startup group there, and they've sucked all of our guests. What are they building these days? What are they making? They're in, what are they building? They're gaming hardware. Oh, so they're making, like... They're not allowed to say it's the next generation gaming hardware. I see.

Greg Charvat: They can't tell you anything, just like my thing.

Speaker ?: No, they can't.

Dave Jones: That's right.

Greg Charvat: No one can say anything in this startup business.

Dave Jones: All this IP crap and, yeah. Oh, man. Don't you hate it?

Greg Charvat: Well, I don't know. We'll see. I might like it later.

Dave Jones: So tell us about Lincoln Labs. We were talking about this before the show. What's... How many people, and what's the budget?

Greg Charvat: Oh, man. Lincoln Lab, first of all, it's an amazing place to work. It's... They work on primarily government, military things, and FAA and NASA stuff, but mostly for the military, the U.S. military. Mm-hmm. And it's... I think the budget is something on the order of not quite a billion, but close, last I recall. A billion. Before I left. And I think that they were well over 3,000 employees on their way to four. And that was kind of where it was about a year ago when I joined the startup. And they're growing by, you know, pretty... By leaps and bounds. And they're doing a lot of important work over there. It's quite a place to be, actually. It was amazing.

Dave Jones: Unbelievable. 3,000 employees. I don't think we have 3,000 employees working in the entire Australian electronics industry

Greg Charvat: anymore. You must have a few people, though. You make some pretty good stuff, though, these days.

Dave Jones: Oh, well, we try and make some stuff, but geez, yeah, it's down in the noise. As far as those sort of numbers are concerned, it's ridiculous.

Greg Charvat: Well, yeah, we were just a lab of MIT, you know, just... Yeah, I know.

Dave Jones: It's just one lab in... Crazy. So you've formed... You've joined a startup now. That's right, yeah. Can you tell us anything about that at all?

Greg Charvat: Well, I can say that we are... We're not trying to be a... We're not a social network or Web 2.0 startup. So I figured that'd be a relief for you. We're actually making hardware, and we're working on next-generation imaging and diagnostic equipment. And we have got a very small team at the moment, but we're trying to staff up. And so we're actually trying to hire digital engineers and FPGA embedded designers and stuff. So if you know any... There you go.

Dave Jones: If any of our listeners are into that sort of stuff, go for it.

Greg Charvat: Yeah, yeah. And, oh, I wanted to say, if any of the listeners here are into that stuff, you can write me directly, but don't send me your resume. Send me pictures of boards you've designed.

Dave Jones: Ah, there you go, folks. We've talked about that one before. That's more impressive than CV. Yes. That's, yeah. No one cares about the CV.

Greg Charvat: No, no. Well, I mean, I think that HR folks might care about it, but all I care about are the boards and the things you've created. I think that's the coolest part about engineering.

Dave Jones: Absolutely. And you've created some very cool stuff, which we'll link into the show notes, no doubt. But you're into radar, primarily. You're into tubes.

Greg Charvat: Oh, yeah, yeah. Tubes are great.

Dave Jones: Okay. We'll have to talk about that.

Greg Charvat: They're so easy to design with. Really? They're so easy. Oh, yeah.

Dave Jones: See, I've never designed with tubes. I guess I'm too young. You don't know what you're missing.

Speaker ?: Right.

Dave Jones: Oh, boy. Now, you know, I'm not too much of an analog guy, let alone a tube guy. So, there you go. Please extol the virtues of the tube.

Greg Charvat: Well, okay, so I have to say, I think it all started when I was a kid. My parents were very supportive, and they'd pick up these old television sets for me to take apart. Excellent. Back in the early 80s, everyone was throwing out their tube stuff. It was all dying off. So, I took apart a lot of tube stuff, and I think that it was just wired in early on that I had to do things with tubes. But the audio, you know, so I got into making tube audio gear for fun and got out of hand with it by making the vacuum tube home theater system that I have. That just was a little crazy.

Dave Jones: And I've seen it. It's a huge, what, 40-foot rack, four-foot high, no, six-foot high rack or something.

Greg Charvat: That's right. Yep. Crazy. Yeah, it's full. 23 tubes and maybe like, actually, I can't remember, 16 or 20 amps of filament current and, you know, the plate voltage is 535 volts DC. You know, look, I like to tell people it's about efficiency. Okay? So, you know, where else could you both get amazing and loud home theater sound, high-quality sound, and heat your living room, all for one kilowatt.

Dave Jones: All for one kilowatt. Measley one kilowatt.

Greg Charvat: For one kilowatt. It's a great deal. Oh, boy.

Dave Jones: Because these are like 400 watt per channel amps, right?

Greg Charvat: Well, that's peak. So, I… Oh, right. Okay. Someone asked a really good question on the comments, which was, you're really getting 300 watts? And I'm like, no, no, no, no. I think there are different ways of characterizing output power of audio amps in any amp fire. So, that was peak, and really, RMS, I think, is on the… There's a four-channel amp. It's 80 watts per channel RMS, and then it's 55 on the single monochannel amp. So… Right. That's RMS. Now, if all four are going at 80 watts, the power supply starts to sag.

Dave Jones: Okay.

Greg Charvat: And then they'll start clipping probably around, you know, after a while, probably 50 watts, 40 watts, and somewhere thereabouts.

Dave Jones: Well, you need a beefier power supply with big copper bus bars running throughout the thing.

Greg Charvat: Oh, no. You don't need that with 600 volts DC. You just need, you know, little wires. Of course. Yeah, yeah. That's it.

Dave Jones: Hence the high voltage, yeah.

Greg Charvat: 400 milliamps at most, you know. That's it. But, you know, the power supplies sag with tube stuff. It's weird because we're so used to, you know, in the solid state world, digital world, you have regulated power supplies. They don't sag.

Dave Jones: Yeah, that's right. Well, as long as you design them, right?

Greg Charvat: Well, they don't sag.

Dave Jones: They drop out, you know, and then they can do all sorts of weird and wonderful things when they drop out. They can oscillate and do all sorts of stuff. But, yeah. So they sag, and that adds to the warm ambience of the sound, does it?

Greg Charvat: Well, I don't know about that. I don't know. I don't know. Sometimes that stuff can be bogus. Yeah, right. You know, when it sags, it definitely, it really just lose peak power and it'll start to distort when it sags. But I've thrown some parties in college with that system before, a number of parties. And we've had that thing going, you know, early into the morning, very, very loud without any problems. So no sagging going on.

Dave Jones: Speaking of college, you were saying before the show that how long did it take you to figure out how to get girls, how to meet girls in college?

Greg Charvat: Yeah, well, okay. So, you know, I love the hobby of amateur radio. I've been into it since 13. But, you know, gosh, you know, when I went to college, I really wanted to meet some girls. And it took me five years of college to realize that the girls, generally speaking, not all the time, but they're not, most of the time, the girls are not in the amateur radio club. You got to go. I realized, oh, well, you have to go where the girls are. Oh, boy.

Dave Jones: And it took you almost your entire PhD to figure that out. Yeah.

Greg Charvat: Yeah, it did. You know, and finally I decided to learn how to swing dance and, you know, then got, you know, disabled meets. Now I'm married. We met at a swing dance, my wife and I. There you go. Very good. Yeah. Yeah. So it worked out well. So now I can just go back to ham radios.

Speaker ?: Right.

Dave Jones: Now, what's this?

Greg Charvat: Hopefully she's not listening.

Dave Jones: Of course she's not. No, she won't. No, wives don't. No, they don't pay any attention to our hobbies, do they? No, no, basically not. No, they just, no, couldn't care less.

Chris Gammell: And sometimes they even.

Dave Jones: Hey!

Chris Gammell: Yeah, yeah, I made it. I snuck in here. Sometimes they, the only thing they do is they loan us their phones so that we can hook in with 4G connections with their cell phones.

Greg Charvat: Well, that's been very supportive.

Chris Gammell: Oh, yeah, it was. You aren't going to wind up with a bill of hundreds of dollars, are you? Oh, I'm probably, I'm watching it like a hawk. We'll see. Right. We'll see, guys. I take it we are live right now. We are live-ish. We are live. Yes, we're 15 minutes into it.

Dave Jones: Yeah.

Chris Gammell: Oh, man.

Dave Jones: You're way out, dude. Now, what's a .46% distortion that you proudly quote on your website for your valve amp? What is this? I thought, you know, the whole idea of having a tube amp is so that you get that warm, fuzzy valve sound. Well, you want distortion.

Greg Charvat: Oh, no. Isn't it actually warm? It is. It is. And, you know, honestly, I built them just to make a, to learn how to work with tubes and just build a high-performance amplifier. I wanted the lowest distortion possible. I designed the feedback control system, you know, undergrad control system theory and applied it to the amplifier and calmed its oscillations and, you know, got less than a half percent distortion. And how long did that take you? That was not trivial. Right. You know, I'll tell you what, most of your time when you design a tube power amp, a good one, is going to be spent actually compensating the network or making a compensation network to keep it from oscillating. Yeah. You know, the fun valve, you know, soldering the resistors together, point-to-point wiring, that's nothing. That's a week. Uh-huh. You know, measuring every single frequency plot manually with an old tectronics oscilloscope and an analog signal generator, that takes, you kill a lot of weekends doing that. Yeah, man.

Dave Jones: Now, here's an interesting fact, which is thoroughly fascinating. If you go on to Greg's, look at all the stuff we'll link in in the show notes of all the stuff he's designed and done. You've actually never laid out a PCB.

Greg Charvat: Never. Not once.

Dave Jones: Wow. Now, insert PhD jokes here, folks.

Greg Charvat: I don't know the faintest idea how to do it. That's why we need those board designers to join my startup.

Chris Gammell: I was going to say, didn't you say that that's going to be the interview criteria is if they have laid out a board? Yes. That's a prerequisite. That sounds a little hypocritical.

Dave Jones: So, you say, startups just filled with PhDs that can't actually hold a soldier in iron or lay out a board. Well, you can hold a soldier in iron, clearly. But, yeah, sorry. You can't do a board, though. We just love making PhD jokes here on the NPR. It's a staple.

Greg Charvat: Oh, you know, you'd be surprised what you could do without laying out boards. You could do a lot. Oh, yeah, of course.

Dave Jones: Point to point. Tag strip. Yeah.

Greg Charvat: Yeah? Yeah, tag strips. You'll feel those thermal strips with the lugs on them, and then you've got the proto boards, and then the wire wrap. Now, that's the best.

Dave Jones: Oh, boy. Now we're talking.

Chris Gammell: We have some pictures on previous episodes. There's some pretty epic wire wraps on Flickr if people search for that stuff. It's just like they're doing like digital, you know, 64 lines at a time stuff. It's just insanity.

Greg Charvat: Now, are these new boards they're making? Are they pro types, or they're just from old vintage? Oh, no.

Chris Gammell: These are very legacy, yeah.

Dave Jones: Wow. But that goes to show how diverse the electronics industry is. You can be a top-notch industry designer yet never have laid out a board. Oh, yeah. Or never have held a soldier in iron, for example. Some of them never do it. You know, they just – all they do is design chips or they – you know. And I know people, top-notch board layout people, never sold it in their life. Wow. And it's – yeah. It is so diverse.

Greg Charvat: It depends what your thing is too, you know. Like –

Chris Gammell: Well, I think a lot of companies want that, right? I mean, they want the – or even institutions like schools, you know, they – especially PhDs. I'm sure you can speak to that, Greg. You know, they want very focused work. And so if you don't branch out yourself like you do with, you know, hobby stuff, then how do people get it?

Greg Charvat: Yeah. I would say that if – had it not been for hobbies, I just, you know –

Dave Jones: You wouldn't have designed any electronics, right? Yeah. Yeah.

Greg Charvat: Yeah. Hobbies are key. You need to have the hobbyist interest.

Dave Jones: Some people don't, though. Some people have a job where they, you know, they don't want to do any electronics on the outside because they get their full fulfillment at work. And well, okay. You know. Yeah. But I've found that's rarely the case, though. Yeah. Especially the stuff you're truly interested in, you know. Yeah. It's pretty rare that you get a job that has, you know, smack on exactly what you want to do. So, you know.

Greg Charvat: Well, at Lincoln Lab, I was doing a lot of management after – managing of programs after a while. And so I would go home and, you know, fix a valve, an old tube radio or, you know, improve the feedback control system on my stereo or something. Yeah. Then get the engineering fixed. You know, you don't want to lose it. You can lose it easily. Of course. Got careful.

Dave Jones: Yeah. Oh, man. But your – one of your big things is radar, right? Tell us about that. How did you get into radar? Was that part of your PhD and your stuff? No. Well, no.

Greg Charvat: Actually, where I went to school at Michigan State, it's not – they're not really known for radar. But if they find a grad student who's ambitious enough to try to teach himself or herself, they'll give you enough rope to hang yourself, basically. Yeah, okay. Go ahead. Go knock yourself out. So when I was 16, I was working on a radio telescope for the science fair. And it was a really nice one. It has all analog swept frequency. It can track signals and things. It was very interesting. Awesome. I toured the Big Ear radio telescope at Ohio State University. This was 1996. And I – one of the volunteers there was an astrophysicist who co-founded this company that built radar systems, imaging radar – small imaging radar systems to image stealth airplanes, basically. Right. And it's called – Lintec was the company. And they were bought out by Aeroflex at some point shortly thereafter. And so he – so I was asking all these technical questions to the people at the radio observatory. Hey, how did you do this? How did you do that? Because I'm trying to make one, you know. How do you deal with this problem, that, and the other thing? And then so he goes, hey, those are pretty good questions. You want to intern? I was only in high school at the time. Awesome. So I'm like, sure, yeah. So they brought me down to this company and showed – you know, I was working with these radar systems. I was doing a lot of wiring, cables, connectors, boards. And then I was – then they showed me how, you know, to use the 8510 network analyzer. So I'd sweep amplifiers, this, that, and the other thing. Nice. And then I would help repair radar ranges. Like they had a couple – there's one range in Dayton I had to go up there and repair the actuator, the azimuth actuator. And so – so anyway, that's what got me into radar because I – you could actually make imagery with microwaves. I thought that was fascinating.

Dave Jones: It is. And you're now teaching a course, are you, at MIT about –

Greg Charvat: That's right, yeah.

Dave Jones: Actually designing your radar thing using, what, tin cans? Tell us about that. Yeah, a coffee can. How did that come about and how does it work?

Greg Charvat: Well, I was at Lincoln Lab. You know, Lincoln Lab is a really cool place and they had – there was this guy, this division leader, you know, his name is Bob Shin. He's a really interesting guy. And he sent out this note saying, hey, we need to – division head is a very high level at the lab. You know, you're – you know, it's a pretty flat organization, but that's one of the people at the very, very top. Right. So he wanted to put together a class for a – it's called IAP, Independent Activities Period Class. At MIT, they used the month of January to just be a sort of a free-for-all where the students can take whatever courses they want for fun.

Dave Jones: Oh, okay. Fantastic.

Greg Charvat: Yeah. And really, almost anyone can teach a course there.

Dave Jones: Oh, right. You don't have to be a – you don't have to be a tenured professor or anything.

Greg Charvat: Anyone can. And so – Right. Interesting. The laboratory. Dave is a guest lecturer. I could be a guest lecturer at MIT. There you go. Seriously. That'll look good on the resume. I'm sure they would take you in a second if you had – if you wrote up like a week-long class or something. You could teach it. There you go. Yeah. You apply online.

Dave Jones: Shame it's 12, 14,000 miles away. Oh, yeah.

Greg Charvat: Yeah. Maybe you could do it by teleconference.

Dave Jones: Right. Teleconference. Yeah. Dave appears on a big screen. Yes.

Greg Charvat: A big – like a huge one, you know. And they have a little camera there looking at everybody else. Oh, cool. Yeah. So, you know, this is how I – so the lab thought that – the lab was trying to, you know, improve their relationship with campus because the lab is in Lexington, Massachusetts and campus forces in Cambridge and the lab is doing the, you know, government, military stuff and campus is doing all kinds of stuff. So, you know, they're always trying to sort of improve the relationship and so they thought, well, this is pretty low-hanging fruit. We can offer a class on radar. Terrific. So, they came to me and I wrote up a little short proposal and I said, hey, you know, when I was – for my master's thesis, I built a imaging radar system, you know, from scratch. You know, I built lots of radar systems in grad school but my first imaging radar system was made out of, you know, some wave guides and stuff and that, you know, horns and things. But the cool thing about it was I made synthetic aperture radar microwave images with it by laying out a piece of measuring tape and translating the radar manually every one inch and then hitting on a lab view, you know, VI. And then it would pulse the radar when I hit enter. And I did that 138 times. And it would make imagery though. It worked. Sweet. Wow. So, I said, hey, you know what? What if we do this with parts that are at lower frequency where the parts are a little more available? And, you know, what if we did that and we had everyone make a synthetic aperture radar system, a microwave imaging system? Why not? And I said, why not? Look, it's been done before, you know. And they jumped on it.

Chris Gammell: So, do you now do the same thing with the enter or do you actually like script it now so that it actually...

Greg Charvat: No, no. It's worse, worse than that even.

Chris Gammell: But, you know... We have to like hit like a button or something. Oh, no, no. You get an enter key, you get like a push button or a dip switch.

Greg Charvat: Well, here's the thing. You know, if you're going to build this radar, it's got to be cheap because you're trying to give it away to the students. So, it's got to be very, very cheap. So, this thing was built out of... We use coffee cans for the transmit and receive horns. That's why there's two of them. We use some mini circuit stuff and we bolt it all to a block of wood, you know, like an old 1920s radio. Yep. So, it gave us cheap and also gave us something that was very transparent. You can look at it and see how it works, understand it conceptually. So, how we acquired data was we use the audio input to your laptop as the digitizer for the radar. And so, what you would do is the right channel would be what we call the video signal or the baseband signal. And the left channel was a synchronization pulse. Got it. So, on the rising edge, you knew the radar chirped. Yep. And then I wrote a little MATLAB script that would look for the rising edges and pull the data off, you know, the de-chirp data off the right and then bin it into a little array. And so, that's how the radar worked. And so, when you would form a synthetic aperture image, each kit came with a 50-foot tape measure. You lay that sucker... Well, first step is find a straight and level surface. Right. Step one. Yeah. Step two. Put this 50-foot tape measure out there. Oh, boy. And then step... You really got to want it. Okay. Right. Okay.

Dave Jones: You really got to want it.

Chris Gammell: Right. This isn't for weak links, right?

Dave Jones: No, no, no. This is not a flashing lead class, folks.

Greg Charvat: No, wait. No. This is... You got to want this image. Okay. Now, step three. You line the radar up the left side of the block of wood with the first mark, two inches. Okay? And there's a little toggle switch in the kit. And what the toggle switch does is it mutes the left channel where the sink pulses are flowing out of the radar. and so you you mute the left channel move it two inches unmute the left channel for a good one

Dave Jones: mississippi two mississippi three mississippi then you mute it i tech engineering here yeah

Greg Charvat: you slide it over two inches and unmute it and mute it slide it over two inches unmute it and of course you count your one two three mississippi and all the while you record a wave file

Chris Gammell: i feel like you should have a different count instead of mississippi it should be like one

Greg Charvat: fine man two fine men here something like that i guess we could have but i don't know i like it

Dave Jones: yeah you're right i like you keep keep things simple and this and this finally produces an

Greg Charvat: image at the end after all this massive amount of processing yeah it does it uh it will create an image it's very coarse microwave image but the idea is you want to differentiate targets in range and cross range so you know away from you and across the synthetic aperture that you just acquired and we would instruct the students to find really big metal objects and image those like for example the uh alexander calder um it's it's uh the great sail there's this huge metal statue on mit's campus called the great sail and uh when you image that thing with the coffee can radar it's this big metal thing there's lots of multi-path within the statue just bang bang bang and then you get this amazing image it looks like a wood screw after you process it

Dave Jones: terrific yeah it's really cool so how do you do you have to rotate you know i i know absolutely jack about radar okay so i'm a total noob here do you have to like you know rotate yeah i mean the thing like do you have to go all 360 degrees around the object no no how does it all there are different

Greg Charvat: may there are different ways you can do it and the way we do it was we have we we we make a linear synthetic aperture so we set it we set the tape in a line and usually you you're we tell everyone to find like a rail or something or you know you know put a couple of two by fours out there on some horses yep and you just go in a line and that's how that's how this you can you can circle it that's another way of doing it okay but it's easier to know where you are on that with a couple of two by fours and saw horses got it so it but this is mind you this is the last of three radar experiments you do the other two are not as rigorous the first is a doppler uh experiment and the second is just ranging and the third finale of the class is always the star image right so you go in a straight line

Dave Jones: pointing towards the center of the object so you have to adjust as you move out oh no from the line

Greg Charvat: do you have to adjust the angle or no you don't you just you just let it always points out so it's

Dave Jones: always pointing straight out even though it's not in line with the object that's right yeah right okay so how far can you go either side before well you know you're not seeing the object anymore yeah let's

Greg Charvat: see the beam width of the coffee can antennas is about uh 70 degrees oh okay right they're floodlights man okay got it you cannot miss it got it so you can keep going for a mile and you can go for a mile

Dave Jones: and you'll still see it right got it and there's lots of advanced processing that eventually stitches the image together that's right yeah right and that's all a matlab script yep and it come if you

Greg Charvat: go to the uh the open courseware site on uh mit's open court mit.ocw or something it will put in the

Dave Jones: so it's all there so people at home following along at home can build their own coffee cup

Chris Gammell: yes everyone can build one fantastic and so you said this is all kit form does that mean people can

Greg Charvat: buy the kits too is it can't you have to there's a bill material and you have to order everything there's several hundred parts it's pretty tedious but oh really it works it does work and that's the funny part you just have to have a little bit of faith wow but uh the the i get emails about once a week from new people who've actually taken the plunge and built this thing and you know they share their results and stuff it's pretty sweet a lot a lot of people built it's being adapted all over the country actually um uc davis actually took this concept this class is like a one to three week class in its current format this guy uc davis turned into a uh one semester long course on radar you mentioned

Chris Gammell: an email too about this experiential learning and and like you're really big into that stuff i mean like you said uc davis is doing that but i mean is this a wider trend you're seeing or is it more

Greg Charvat: limited so i think it's a wider trend i find there's a huge demand for this sort of thing um big time big time demand it's pent-up demand actually uh you know i when i architected this course i i set it up i i wrote the course to be how how would i would like to learn about radar you know yeah what what sort of what would be the ideal course and that's what i did and and the ideal

Dave Jones: course something that works build it yes yeah exactly who wants to go into deep into radar theory

Chris Gammell: it's probably boring as bad shit right oh man yeah the question is why it took so long for it to get to that i mean like why why did it why isn't it always been like that i guess is the question

Dave Jones: i have no idea i'm a huge fan of working backwards you know you start with something well here's you know you build something or you have something that works here it is okay now let's go to the theory and figure out you know how that works you know and not from the physics level like you know from the modular block diagram system you know that top down kind of not from the bottom up you know oh let's learn quantum physics and bloody you know electron physics and everything else and then we'll work our way up that's just bullshit you know how can how can anyone keep interest like that

Greg Charvat: i don't understand it i i don't know either i mean all all i can tell you is i was the most relentless university right well yeah i know many of them i never really paid attention in college or school ever really i just was throwing circuits on the margins of my textbooks the whole time so i you know i wrote it like that like i figured you know how how what is the right way to teach us and like just like you said we we did we would have high level lectures sort of explained to you the basics we would we would get into the physics right we'd say all right now we're going to build this and do this experiment and email us your results so they'd email us results and we'd post them up on the the big board there and and call on the teams who acquired this result and we'd ask them to explain why does it look like this and what happened is sometimes people wouldn't know understandably so we would start diving into the more deep in-depth radar theory that we understand at lincoln lab and people appreciated it at that point and absolutely they're very engaged you know i'll give you an example like you know it's in such a demand that we we had a mit professional ed version of this course so we took the ip one and we turned it into from a three-week course to a five-day course for professionals um and in in the when the first time we ran it which was august 2011 we filled the course with 30 people plus a waiting list wow that's the kind of demand for hands-on learning that's out there and we were the first hands-on learning professional ed course at mit that was open to you know professionals like yourself and myself yep yeah we we were the top ranked professional

Dave Jones: ed course at mit in 2011 there you go yeah we did really well there's that many people who want to in that many professionals who want to learn about radar is there oh yeah yeah in matter of fact it's

Greg Charvat: been spun off to the mitre corporation they now uh have their own version of it where they use the coffee can radar and our lectures and uh like i say uc davis and and we actually ran an internal there's internal one in lincoln lab and there's a there's one for the air force that they do there's you know

Dave Jones: it's sort of got it going which is awesome it's great is there a simpler way to do it because you talk about like there's a the bill of materials there's a couple hundred items and there's you know is there a simpler cruder way to do this and get something running i would say in terms of the minimum investment you need before you can get a a star image if you could print this design onto a

Greg Charvat: board uh it could become a lot cheaper i think you can get down to maybe the 200 price range right

Chris Gammell: uh with with surface i think that that might be a call for our listeners maybe one of our ambitious

Greg Charvat: listeners would like to do that go for it i mean we there is a huge demand for the uh kit in in a real kit form where you don't have to order the build material and if we had it in a kit form it would have even wider adaptation i'm sure there you go there's a project and it's all open right so

Dave Jones: somebody can take all the circuits and everything and come up with a presumably still open source because you can't just steal some stuff and close it so uh no they could come out with a a kit sure i guess and be the kit supplier that's right and they could build it better and make

Greg Charvat: improvements there's a lot of room for improvements i will be honest and uh you know i would love to see that because it's it it would greatly improve access to this this sort of uh learning experience

Dave Jones: hmm now you've got a you've been working on a radar which sees through walls that's right yeah tell us about it how does it work what type of walls how far can you just see a blobby image or can you actually see people's faces of course blobby image yeah there's no way you're going to see faces

Chris Gammell: or anything um so what what distinguishes a person from a big metal statue movement going forward yeah yeah the statue doesn't breathe it's true statue doesn't breathe i've i've heard about that with

Dave Jones: statues so you can actually physically see someone breathe and tell someone's breathing even if they're not

Greg Charvat: moving yeah you can if you want to we've done this before you can actually measure their breathing rate

Dave Jones: how how does that work please explain creepy yeah it's it's uh yeah is it is it their chest expanding

Greg Charvat: or what is it it's just it's just they're moving right they're moving very slightly in the right so

Chris Gammell: it's still based on movement yeah yeah right so the one of the videos that we that that you had linked us to though it looked it looked really blocky like how can you even see that it seemed like it was like you know like 20 pixels kind of stuff it didn't is as it progressed since since uh some of

Greg Charvat: the prior press on this thing no it's uh it's that's about it and the thing is it's it let me think about the resolution i think it's on the order of uh gosh maybe i think it's about a half meter by four inches or something like that it's kind of a rectangular block that's that's that's the resolution and it's limited to the the physics of the aperture you know the the antenna array which is the physical aperture and the antenna array is eight and a half feet uh wide so uh the center frequency is three gigahertz you can only do so well in in in uh resolution and and if you go up in frequency

Dave Jones: i'm assuming then you would have problems with absorption and penetration and all that so you can't just arbitrarily go up in frequency and get better resolution right yeah you won't get through you don't get through so what is the range of frequencies that can get through what concrete

Greg Charvat: presumably yeah concrete oh you know really i think any our any microwave signal will get through concrete just means how much right you know how much gets through and so we're at the upper end yeah exactly you know we the frequency of operation for our system at three gigahertz was at the absolute highest frequency people ever people attempt through all imaging really most people you know it's it's it's uhf you know 440 megahertz or right or maybe 1200 megahertz or somewhere thereabouts it's almost never uh three gigahertz got it and what we did it's we we chose that because we wanted resolution so let's say we we went down to 400 megahertz well we'd have to have an aperture almost 10 times the size of our eight and a half foot aperture to get the same resolution so that thing becomes ridiculous

Dave Jones: like there's no way it has to be bigger than than the room you're than the room you're imaging yeah

Chris Gammell: ridiculous yeah and i imagine this stuff like you're selling it to like the dod or something like that it's like well you know this thing can see through tons of the concrete but you know it's

Dave Jones: about 40 feet wide so you know they've got visions of you know all these you know soldiers coming up with their little you know with their little x-ray vision camera yeah that's not gonna happen

Greg Charvat: no no we're not gonna get that you know not with this technology i mean not with microwave imaging no way

Dave Jones: so those x-ray specs they advertise in the back of the comic books they're bogus are they

Greg Charvat: oh i i i cannot answer that oh i don't know i've never tried well that's good because our uh

Chris Gammell: our one of our listeners was very worried about uh people coming to get him i don't know if he uh has particular uh issues but he was worried about people seeing through walls well i wouldn't worry

Greg Charvat: about this thing being used adapted widely anytime soon it's a prototype and it's uh you know there's every time we would turn get it out of mothballs to use it something else would break and you know it costs a fortune and all that so it's all it's very much a laboratory thing still and you can't

Dave Jones: defy the laws of physics captain you know it just requires big antennas to do this it does you know and even then it's crude so you know yeah if somebody's parked a 40-foot shipping container out the front of your house then yeah maybe be concerned that there's hard to get nervous right from that no you know that uh black van out the front the moving company is yeah that's right

Chris Gammell: it's like a moving van or a flower van and all the guys are wearing sunglasses that's like yeah then

Dave Jones: you get nervous i don't think you have to worry but they're the ones shooting lasers onto your windows so they can hear everything you're you're saying right so they don't need our radar for this and they've got their tempest you know stuff which allows them to pick up everything you're viewing on your monitor and all that sort of stuff you know be afraid be very afraid i highly recommend

Greg Charvat: the tinfoil hat works well actually if you wore that with our no no with our radar you'd you'd show up more strongly with the tinfoil hat oh there you go so it would do the opposite

Chris Gammell: oh that's such a great idea because you know all the crazies out there

Greg Charvat: they're actually making themselves more visible yes way easier to pick up with the tinfoil hat and actually we found if you carry like anything that resembles a firearm you become more easily detectable as well yeah by about 10 times as much as it will be reflected in that so yeah because it's steel anything metals just really lights you up that is solid yeah but yeah it's a prototype but it's of course only for military applications would be the end idea but it's a laboratory prototype and it works quite well for like you know as blocky as that imagery is that's uh it's very good for through all radar if you look at a lot of the other systems out there it's it's good so what about this

Chris Gammell: uh so this is part of the course with the i i think but it has the wi-fi transmitters is that right

Greg Charvat: the other phased array is that something different than the coffee can yeah so that was the uh the the there is so a good question now so the we had the uh the coffee can course was successful so so lincoln laboratory wanted to have another radar course and so i proposed let's run the coffee can again and let's let's let's kick it up a notch let's build phased array radar systems cool yeah and what so what we did was we so what what is i don't know what that is sorry it's it's a radar that has multiple antenna elements and you know like let's say for example you know like the big uh red radars that you see at airports that look for the aircraft you know the the faa radars i think it's the asr9 yeah that radar is is what's called a real aperture it has this big huge dish and it spins and it scans the horizon now a phased array radar something like you'd see on the aegis uh system you know the the aegis weapon system which is in all of our destroyers you see it's those flat panels on the front yeah those are phased arrays now those things can scan all directions like that big red faa radar can except they do it digitally using beamforming yeah technology which is pretty

Dave Jones: advanced it is yeah i've worked on beamforming underwater sonar stuff and that requires massive fpgas massive parallel processing that's one of the beamforming is one of your traditional uses for the year ten thousand dollar high-end fpga chips yeah it's uh it's really advanced stuff so what we

Greg Charvat: wanted to do at mit was we want to have a course on phase ray radar and so instead of doing it with ten thousand dollar fpgas we're going to make fpg we're going to make phase ray radars out of pegboard and wi-fi antennas of course as you do explain how that works well what we did was we succeeded okay and it worked yeah okay class and it was a big hit so what we did was we i i designed the the the radar with another guy uh lincoln uh brad perry and we we found uh these really low cost um wi-fi antennas patch antennas and so we bought a few and measured them in the with the laboratory and found out realized yeah okay this will work you know it's only thing we can get so it's gonna have to work so many of my designs are there's no spec it's like let's just what see what we can do and make it work you know so anyway we we we found these antennas we came up with with an antenna layout that would allow us to operate in what's called a switched uh switched array uh mode where we took a coffee can radar and removed the coffee cans instead of plugging them into instead of plugging the radars transmit receive ports of coffee cans we plugged into uh antenna switches a four-port switch for transmit a four-port switch for receive and then each four-port switch was connected to four wi-fi antennas and we placed them in such a way that we can get uh half a wavelength um uh virtual aperture spacing so we switched okay only back up a little bit we'd transmit on one receive on another then switch and transmit on a different one receive on a different one we would do all several combinations

Chris Gammell: of antennas you know i think it was like was that like a mimo system yeah it was it was a mimo system

Greg Charvat: it was time division multiplex we didn't transit simultaneously okay so mimo being multiple input multiple correct for people that don't know that that uh so it was my we used a lab view um little uh usb controller to toggle all the antenna switches and pull in the pulses and yeah we could we had a real-time uh phased array radar system of a range of about 100 meters and you could watch people walk around uh the common at mit or you know or wherever you can watch cars drive by with the thing awesome it's

Dave Jones: really cool well you can also do that with the camera you know yeah that's true but it's not as

Greg Charvat: cool hey what if it was foggy yeah exactly that's right oh boy you didn't happen to use the uh the edis research stuff did you because we've had man edis on the show before no we didn't uh i i i've i've looked at that before for the radar problem and the the trouble i've run into and some people have taken radar courses have tried to use it for the coffee can radar the problem we run to is triggering and timing the radar is all about timing you transmit something at a known time and you receive you know at a known delay from that time so that is stuff sent to no linux then it tends to favor streaming you know it's it's it's it's it's radio so it's all streaming and so i think maybe you could get that to work with you know some clever hacks or something i'm sure someone can get it to work but it wasn't us we didn't have enough time to figure it out and do it so

Dave Jones: there's a interesting question from um erison who is one of our listeners do you know of any reconfigurable antenna technology perhaps built using mems technology he's looking for something the physical equivalent equivalent of an fpga for antennas yes i know about that uh they're called

Greg Charvat: self-structuring antennas oh and they are still in the academic and military space and the idea is you have this antenna and you have lots of elements on it okay and then you have little switches that that add those elements or remove the elements and you use elements in fractal elements perhaps no just just just just regular wires right okay imagine a large printed board maybe 12 by 12 and you have all these wires in like a grid or some sort of array and and where each one meets you have a switch and you

Chris Gammell: know expensive rf switches all that that kind of thing yeah yeah got it yeah the idea is that you

Greg Charvat: could take this antenna and throw it on anything and it would self-structure for optimum because we know that if you put an antenna on like a car or aircraft it it's different for every aircraft you know we know this from the military side things every time you put a new antenna on an airplane you have to you basically have to make a special antenna for every single airplane because the skin of the airplane goofs up the antenna pattern yep right so it's uh that's what airplanes everything's expensive

Dave Jones: in the aircraft world technically that pretty much exists at the academic level and yes and an antenna

Greg Charvat: based fpga yes yes there's uh there's a gentleman who worked in my lab at michigan state who built one and it worked quite well his dissertation was on that exact problem there you go yeah so if he emails me right place i can put him in touch with this guy so sweet yeah have you worked on fractal

Dave Jones: antennas have you done anything on those because they're obviously you know what have made our miniature mobile phones and everything no these days possible i haven't touched them i uh my my concern is just

Greg Charvat: my my interest in antennas extends to what what do i need to do to make this radar work got it and then i stop right there so i i've designed antennas before but only to make the radars work and then i can i never just of how they work and then that's it gotta move on to the next problem

Chris Gammell: i love it that's great who's got time for everything it sounds like it sounds like you're a hacker trapped in a in a professor's body i hope that i hope that sounds like a compliment because it really is it's meant to be very accurate it really is um what what so what is your your actual research here i mean is it is it it's in the radar systems area i mean is it well i i kind of you know i got

Greg Charvat: into the radar because i i worked at a company building radars when i was in high school so i thought it would be cool and then in grad school i taught myself how to build imaging radars i made a series of five different imaging radar systems in my garage i had a you know i had a microwave lab in my basement i had a milling machine in my garage a bridgeport i'd i could turn out anything in that garage at my house in grad school but it the reason why i did that is i just thought it was fun you know i ended up getting uh my my i was studying applied electromagnetics and i was very much interested in the applied aspect of electromagnetics so my dissertation topic i i actually got the funding myself for the dissertation topic it was funded by the naval research lab in washington dc i met um that of the radar division at a uh conference and at conferences i like to go you know hang out and chit chat with everybody and see what they're doing you know because it's fun and this guy said hey we're trying to look at this through all problem we've been struggling and you want to you know you sound like you've got some really good experience building these radars in your garage you want to give it a a shot so i said yeah why not so uh yeah they funded my dissertation which was on through all radar then later on we brought that work into lincoln lab and made the very advanced through all system

Dave Jones: which you see uh online do you think you said they funded it do you think you would have been able to

Greg Charvat: do it if they didn't fund it yeah absolutely i think i spent uh to build the so i had to develop uh as part of the dissertation i built a model to see if it's even possible to get the right snr to actually form inventory and so on so forth so i made a model uh then i came up with an unusual radar architecture that would facilitate uh low uh peak power pulses and and long waveforms so i was transmitting 10 millisecond long chirps into the wall and being able to pull them out which is difficult because most of your energy just bounces right off the wall and then saturates your receiver deafens your receiver and distorts everything so that's what you have to fight is that problem so therefore most people who approach the through all problem look at in terms of an academic sense they say well if if if i'm going to saturate my receiver with a big long waveform i'm going to send a short two nanosecond waveform what happens is if you send a two nanosecond waveform now you've got to dump 10 kilowatts into that two nanosecond waveform to get in and out of that wall right yeah well my systems and grad then the wall explodes yeah no it's fine it doesn't explode but 10 kilowatts now you need a twt amplifier you need all this expensive crap to make that happen right infrastructure goes way up okay and and not only that to digitize a two nanosecond impulse you need a good you know four giga samples per second or better digitizer and the performance of those digitizers are not as good as the hundred kilo samples per second digitizers and so um you know so the system becomes unbearably complicated and you also have to swallow a lot of data so so it's very difficult to do real-time processing and real-time imaging with that kind of data coming in even today it does so my approach was oh no you know what everyone's doing that in that field the field of through all radar everyone seemed to be going the impulse route i'm going to go the opposite i'm going to send a really long cw fm cw chirp at the wall and i'm going to figure out a way to reject the flash the reflected signal off the wall itself and only uh receive what's behind the wall only send what's behind the wall to my digitizer and i did that

Dave Jones: and how do you do that do you just shut it off during a time you shut your receiver off during

Greg Charvat: the time period so that's saturated or what no it's always on what i did was i i first of all i used a wide dynamic range front end and what often happens in radar systems is the the the electronics often have more dynamic range than the digitizers so you're right you're usually managing digitizer dynamic range so um i had a wide dynamic range front end that can just take it all in no problem and what i did was in the if uh i down convert to 10.7 megahertz and there i used a um a trick i used what's called a crystal filter which is what you use in a ham radio a single sideband transceiver i put a crystal filter there the crystal sounds pretty obvious actually it does doesn't it with fine side right no it costs like five bucks from digi key yeah yeah that's what i ordered from is it's from digi key and i threw that in there and the when i chirped the radar you know there's a function of you know the rate which you chirp and blah blah blah blah so when the smoke clears on the map what happens is the crystal filter acts like a range gate it oh when you have an fmcw radar range to target is proportional to beat frequency on the output of your your your front end of your radar so you multiply your transmitter by what you're receiving all the time and so fascinating that the product ends up becoming a frequency difference on the order of kilohertz okay you know 10 20 30 kilohertz let's say there's something 50 feet away it comes in at 10 kilohertz and then it goes and now you have something 100 feet away shows up at 20 kilohertz for example so what i did was

Dave Jones: interesting so it's a almost a it's almost a range to frequency converter it is yes precisely

Greg Charvat: yeah wow and this technology has been around since the 1930s this is how uh radar altimeters were created in the 30s they they of course yeah delta f delta yeah yeah that's right that's probably how

Dave Jones: they landed that rover on mars too you know they they had the radar ping in the surface and they

Greg Charvat: oh i'm sure yeah they probably had some kind of radar like that to close the loop on that i'm sure they did yeah and and they're so close they would probably they might be forced to that solution i don't know maybe it's an impulse but it could have been fmcw most likely because that's the historically radar altimeters are fmcw right so you got so now you have frequency is proportional to range well i want to filter out the frequencies that correspond to the wall and pass the frequencies that correspond to the people 10 feet behind the wall that's easy that's the problem right doesn't sound that hot it's not when you think about it from this perspective from a systems level right so i said well they have these you know my ham radio in the basement's got this amazing thing called a crystal filter in it it's been around since the 30s and uh it really does a good job of rejecting everything outside of its pass band very is very very uh tight characteristics so i thought

Dave Jones: steep curves and yeah either side so it's band pass filter nice and steep very very steep so i bought one

Greg Charvat: from digikey and i architected the radar so that instead of uh converting down to 10 kilohertz i converted to 10.7 megahertz okay so i can feed that that base band signal through one of these crystal filters and what i did was i adjusted the um lo on the uh on the radar so that local oscillator folks the local oscillator so that the the let's say the wall was coming in at 15 kilohertz and the people behind are coming at 20 kilohertz i set it up so that the wall would be buried in the noise and the people would be passed through and then i then i down convert to baseband after that and digitized with the 200 kilo kilo samples per second 16 bit a to d so i had a really low you know bandwidth uh a to d very uh low bit count for small amounts of data so you can get massive dynamic range on your adc there yes yeah cheaply and easily right and now with with only uh one milliwatt of rf transmit power i was looking through four inch solid concrete walls and i could see a coke can behind the concrete wall moving

Dave Jones: wow with one milliwatt one milliwatt that was in my garage in east lansing and this brings us to a circuit from a question from tom circuit he calls himself he wants to learn about low power sub 100 milliwatt radar systems but he finds that most of the literature literature is uh high power long distance military it's unapplicable it's totally unapplicable we should

Chris Gammell: mention that that's uh that's with that's of the uh sorry that's that's tom laments of the uh

Dave Jones: that was formerly on the show oh is it okay i thought that was interesting he's calling himself tom circuit

Greg Charvat: on uh reddit okay yeah i know him um so that it's a good question so so most book most radar books are are entirely theoretical you almost never see design examples and there are some good ones though that i can recommend like intro to airborne radar is probably one of the best but um they're they're long-range far-field radars and and when you when you get down to the milliwatt level your radar is often in the near field uh so you have to account for that sometimes especially if you're making a phase ray radar and uh you know you have to use unusual architectures you're not pulsing a 100 kilowatt pulse anymore you're you're you're probably cw because you don't have enough you know otherwise you have to pulse a nanosecond or something and that gets tough to analyze so yeah i'm writing actually i'm writing a book on small and short range radars to fill that gap so it will be the only book on

Dave Jones: small and short range radars when it comes out wow there you go and how's it going by the way uh because we've talked about writing books on here because i've been you know offered uh deals

Greg Charvat: to write books and stuff and yeah i i signed a contract a long time ago uh back in 2010 and

Dave Jones: we're creeping along and we're still going okay there you go folks more confirmation that you can't write

Greg Charvat: a book in under a year no no it's so much and if you're trying to get a startup off the ground there's really no time for the book yeah we're getting a few pages done uh a week and uh you know i could talk to tom and send him some of the uh i can give him a preview if it helps him so oh there you go

Dave Jones: yeah awesome so you've got a deal through a book publisher yeah crc press yeah oh there you go yeah look out for that folks estimated time how close are you gosh i think my deadline is in the

Greg Charvat: winter but uh it might slide a little bit winter of what year 2013 oh boy well subscribe to my blog and then you'll know it i'll blog i'll blog extensively on it when it does come out blog it to death i will blog it to death and and and the book is unique in that uh it's it's it shows design examples uh so we we talk about the theory of course and we we we show in in the book i show how to to characterize the expected performance of very small radar systems and then i show working example after working example after working example each one has a bill material with it it's got all schematics and and data i'll put on the website measure data that you can process matlab scripts to process the data and a youtube video showing you that radar in action and the data acquired during the youtube video will be on the website so that's that's the plan and

Dave Jones: and it's moving forward i think tom's uh salivating at the moment yes but unfortunately we're a little

Greg Charvat: behind but uh hopefully we'll get it out there eventually but that that's the vision for the book and

Dave Jones: we'll make it happen mm-hmm what's the target market for this book uh people like tom is it are you trying are you hoping to get it used as a course text or something is that no is that an

Greg Charvat: intention no that'd be nice i think it could be but i think that the objective here is to it's it's really filling a gap you know when i learned how to build these small imaging radar sensors i had the same problem tom had i would i'd get you know skull next radar handbook which is an excellent amazing book by the way but i would find a lot of its material is unapplicable to what i was trying to do in my backyard right so i just it's just like the radar class i wanted to write a book about radar how i would have wanted to learn it myself and that's the way all the best things come

Dave Jones: about what would i want and then boom you produce that and then other people go oh that's what i want

Greg Charvat: too yeah and you know and the best way to do it yeah and hopefully hopefully it'll fill that void and you know get more people into this uh this amazing and interesting uh hobby that can turn professional

Chris Gammell: so how big would you say i mean like radar seems like it's still pretty niche for you know like military for military and aerospace applications are you are you seeing in other other spaces are you

Greg Charvat: seeing interest from other a little bit you you see it uh automotive uh uses it here and there for blind spot detection and automatic cruise control and stuff um and of course you have the ubiquitous uh door opening thing that you see at the big box stores you walk up to the door it opens that's

Dave Jones: because there's a small radar up now oh isn't infrared i thought that was all the um infrared ir movement detector they're usually uh radar really yeah it's interesting i i could have sworn that though i i know you can get dual types of alarms now that are combination infrared and sort of radar like i'm probably talking out my ass here but you can get like these dual alarm sensors and i thought that's probably what they

Greg Charvat: are were at best now oftentimes they're usually uh what's called a gun plexer which is a little it's a type of microwave module it's been around since the 60s and it's it's uh it consists of two diodes and a ferromagnetic circulator what it is is basically like a radar tr module very cheap and it it transmits a 10 milliwatt cw signal and there's a little mixer diode right there next to the gun diode which is oscillating and it always down converts whatever it is reflected to it well that's going to uh

Dave Jones: really um make the uh paranoid tin hat wearers even more paranoid yeah they're everywhere i'm not walking in that shopping center it's pointing a radar at me yeah they do phone rfod

Greg Charvat: i've taken them apart before uh stanley makes one and and that's that's the one you see everywhere if you look at the name it's usually the stanley one it's really nice and you could then you know you can buy one probably i'm sure you could buy one is there a way we can mess with one

Chris Gammell: is it is there a way to like hack with them and mess with them like to try and get them from

Greg Charvat: oh yeah you could i'm sure you could i i bet you can jam it from you know across the parking lot

Dave Jones: just have it open the door ah right there's a fun project yeah yeah i'm sure you can do it i have no doubt it's

Greg Charvat: possible aha you call it the radar be gone the radar be gone if you're one of those people that wants to waste the heat you know of the big you know you can if you want to like bankrupt a big box store by wasting heat and just opening the doors all the time that there you go right yeah

Chris Gammell: of course it could be like a like a james bond bad guy where that's like your or like a simpsons bad guy where that's your plan for taking over the world you heat the world those little radars every

Dave Jones: door going so please please tell us how do we jam such a radar gosh i gotta think i want to know

Greg Charvat: because i want to try it it would be awesome there would be a couple ways of doing it i think that uh the most difficult way would be to try to lock onto its frequency phase lock to it no no no i just

Dave Jones: want to know i just spew stuff spew something what you'd have to do then is probably the easiest

Greg Charvat: way to do is to buy one of those things right and amazon.com and rip it apart take up the gun flexor and then make a circuit that uh that deviates the bias voltage for the um for the little uh gun diode and that'll cause it to modulate in frequency back and forth and that'd probably get it novel but

Dave Jones: i think over complicated it's got to be a simpler way come on oh there's no simpler way than that come on come on that's your phd talking come on i want the i want the ham hacker talking here

Greg Charvat: you know maybe what you could do that ham hacker way let's think about this oh you know what i just

Dave Jones: thought of it here you go here you go folks see but once he puts on that ham hacker hat you know

Greg Charvat: what i saw this thing comes in there i saw this thing at a lake house uh um a few months ago this this is a lake house sorry just a cottage on the lake oh a literal lake house okay right and these people had these little solar powered lighthouses on their front and you know right of right where their boat ramp was so they can come home and see where their boat ramp is oh okay and the solar power lighthouses were neat in that they had an led light of course which is ubiquitous with solar lights but they also had the um they had a motor in there that would rotate a mirror around the led so so it actually rotated like a police light or something so what i would do is get one of those things okay and try to retrofit a tin foil hat or some sort of larger reflector on there and have it rotate all the time and it'll always be fooling the doppler radar ah there you go yeah

Chris Gammell: do you think there's any uh radar application like we're we're seeing you know google just uh get that bill passed to have driverless cars in california is that part of the is that going to be using radar probably collision detection i'm sure it is okay so there there is a pretty practical application for you know like widespread stuff huh i mean like it's not not necessarily just you know seeing through walls now or detecting what it does is radar gives you a car's coming

Greg Charvat: towards you range you know unambiguously here's the range and angle of something and that's very valuable if you have uh like optics and cameras and things and there's some uh algorithm that has to take place to sort of back out the range but the radar it's it's given without any thinking really

Dave Jones: necessarily cars are going very fast so how a how fast can you detect the rate of change oh which is you know instantly right yeah yeah really so you can be traveling a hundred kilometers an hour sorry yeah got to speak australian here hundred kilometers an hour head on what's that mean to another car oh that's about 60 miles an hour oh you know you can be heading 60 miles an hour head on so two cars heading smack into each other six miles an hour problem yeah it can still detect that

Greg Charvat: two cars leave yeah yeah think about it 9 a.m fighter jet radars right those are true they go a lot faster than 60 miles an hour oh yeah yeah there you go okay

Dave Jones: fine yeah there's no reason why you couldn't do it does that work at any frequency really

Greg Charvat: yeah yeah just right in process and the pulse the rate at which you pulse the radar and the the you know your waveforms and so on so forth but absolutely yeah you can you should be able to pull out range and then uh some sort of vector um of what the target's doing pretty easily yeah okay so well if

Chris Gammell: if every car on the road has this at some point does that mean that it'll start like flooding the spectrum or is it such narrow band signals that it really and they're sweeping so much that's a good

Greg Charvat: question i i i would say well i'll give you a working example now the mit radar class we would have an entire room of 30 of these coffee can radars going at the same time they share the band with wi-fi and everyone has a laptop on wi-fi and we had no problems ranging people down the hallway and in the classroom itself so right um they're tripping yeah that's a good measure you'd have to you got to be nice and close to get to cause interference with each other so it's funny if you're ever on a boat and you have the marine radar going you you almost never see interference from other marine radars they're all on so a lot of things have to the moons really have to align to get to have the radars jam each other but that that being said yes they can interfere with each other but you have a few things working in your favor for one the frequencies have to line up very closely so if it's an fm cw radar like my radars and they're only interested in zero to 100 kilohertz away from the car and let's say a radar is chirping but it's it's it's chirp is 10 megahertz away from yours it's slope meaning its slope is 10 megahertz advanced or behind or retarded from yours you're not going to hear it so the probabilities of the trip lining up are pretty low with and then also the beam widths too so like cars you know are stacked up on top of you know traffic there i would think that their beam width alone would probably be limited you don't have to you have to look at in terms of the isotropic case where it's free space all around you can have some caveats in this but it has to be looked at for sure yeah because even

Dave Jones: if the odds are a million to one of it happening you gotta know you got a million you got a million cars on the road oh yeah you've got an accident every day you know i mean i think you'd have to have

Greg Charvat: some sort of uh the radar would have to be smart in that you would have to do you would have to receive have to do a receive pulse every now and then where it doesn't transmit and it just sweeps its own receiver and checks the spectrum and if it sees something else then it starts blanking that and these radars are going to be you know pinging they're going to be pulsing at such high rates that they can they can spend a pulse every now and then checking the spectrum i see no reason why the spectrum cannot be dynamically allocated and i can think of 10 ways to do it i mean it's it's it's very doable we just have to have the be willing to take a few risks and work out the bugs but it can be done

Dave Jones: man bloody radar everywhere i'm gonna have to start wearing tinfoil underwear yeah then we'll

Chris Gammell: know we'll definitely see you there yeah yeah exactly right so what about this uh this did you guys already mention the startup i know i've been dropping in and out of here but have you mentioned the startup yet yeah that you're working on where we're hiring and yes oh you started with that right

Dave Jones: what is your experience like though at a startup is this your first startup it's my first yeah and a bit but you're one of like half a dozen i'm having a look at the people here you've got like a dozen people working there already right so it's a well-funded it's very well-funded we have more money than time right there you go folks turn up with your oh you guys are in hand and you'll get

Greg Charvat: a water cash yeah we're going for the tricorder prize yeah that's great and uh we i we we talked

Dave Jones: about that previously is it but why why what's the price are you going for the prize money or are you going for the fame for the fame not the prize money right so how do you you know how does you how does your investor feel about that because usually startups are like you know we want a return on our investment screw the fame stuff we just want our money back our investors well the fame makes more

Chris Gammell: money yeah well not necessarily oh please if you win that the i mean tether spacex won the x prize

Dave Jones: didn't they didn't they win the spacex prize yeah yeah but they're doing okay freaking rockets in the space yeah that's right they were putting rockets not they were they get for that they got

Greg Charvat: a million bucks don't listen to they get for putting rocks a million bucks or something what was the prize uh i think it's 10 i think most experts can be enough for the rockets man no no no it's not no they spent 50 times yeah they burned 10 million of fuel just getting halfway up there yeah yeah

Dave Jones: whatever happened to the lunar x prize or whatever it was called that really yeah yeah yeah you can yeah you are the first person to land a rover on the moon okay okay a rover and transmit a high def video back wins the 10 million 30 million bucks or whatever it was that's interesting i don't know it seemed to just dissipate it was a big fan sounds like a microwave uh antenna maybe

Chris Gammell: maybe that's another thing you guys should do team up with spacex it seems doable kind of you know

Greg Charvat: it seems doable i mean the bar is not that high it just has to give you at least one picture and then

Dave Jones: it dies right yeah well exactly i did like one team's shot at it uh well their their theory behind it was that well we'll do instead of putting all our eggs in one basket and doing the one mission we'll do 20 missions that cost fifty thousand dollars each yeah that's one of them's bound to work

Greg Charvat: yeah right in each mission they dump like a thousand radio-controlled uh tonka trucks spewing out of their spacecraft above the lunar surface with those those free you know those free digital cameras you get the yeah like ten dollar digital camera from the drugstore each one has one of those duct tape to it that's like the million monkeys idea right you just uh eventually it's one of these these rovers these these these tonka trucks it's gonna it's gonna bounce off the moon and come back to that's how you retrieve the picture it doesn't transmit it or anything that's right

Dave Jones: that's how i would do it so so your experience as a startup is probably wonderful because you're actually getting paid yes as opposed to a lot of people i know doing their startup well they're just

Greg Charvat: you know starving and eating dog food yes yes our investor has no problem paying uh where he needs

Dave Jones: to get the best people in the world to join us that's the way to do it folks yep make sure someone else that's right yes

Chris Gammell: did you guys already talk about what it's about because that's the thing i didn't understand is

Dave Jones: what are you doing it's not about yeah we said it's not about yeah it's not about social media no

Greg Charvat: it's not web 2.0 you're actually building shit we're building yeah we're building a lot of novel how dare you yeah it's very risky it's risky well hardware is so awesome we've built three prototypes already uh since i started in uh november last november yeah and uh we're working on a tech demo and uh you know we it should be it's a wild ride we're actually executing very very well

Chris Gammell: so so so far so good but uh and we talked about how we don't know what it is i'm not i'm not like we're not allowed to say it's uh all hush-hush yeah unfortunately i wish i could maybe in the winter i

Greg Charvat: could talk more about it and reveal some more but i just can't right now it's all shades of you know

Dave Jones: it's all shades of valve here chris and i did uh i did post fun that he hadn't been poached by valve

Chris Gammell: yet so yeah yeah yeah well once he's done with this yeah but no i mean it's and you know that's an interesting case though because with startups and a lot of other things like even funding situations sometimes people get funded just based on the people right and i could see very easily if i was a you know an eligible hire person who wants to get hired for a startup that based on greg's work alone it would be something i would want to do just because he's doing it right i mean that's that's the kind of thing that you know that those those are good situations because they're exciting you know you have to take a little leap of faith but whatever yeah well that's good anyone out there

Greg Charvat: wants to join us who can uh... code fpgs and uh... design boards send me an email by all means and i'll

Dave Jones: look at every single email you send so and i will help the family totally do that too yes we will

Greg Charvat: out the valve actually we have the means outbid valve but no question about it but but yes are you

Chris Gammell: doing their log of it out there are lives are preference

Greg Charvat: yeah east coast every time

Chris Gammell: you picked the wrong one you picked the wrong horse i know seriously midwest gets screwed especially you know like with all the military stuff yeah anyways yeah

Dave Jones: i'm sure we have listeners out there who will no doubt don't send a resume folks and that's what we talked about don't send a resume send a board send send pictures of your board photos of you with your board yes yes

Greg Charvat: yeah do should i give my email address out now or do we put it in the show we'll put it in the show okay good yes yep email me for you know just please we could use some people and it's hard to find those people who are really good they're very tough to yeah i almost think that they're they're the underappreciated lot where you know they're kind of at some giant company you know designing boards or you know coding the fpgs and no one knows who they are because they're not they're not publishing papers and stuff it's easy to find phds because they're they're out on the scene they're publishing but uh right it's a it's a

Chris Gammell: career move to be self you know self-promoting right yeah it's uh whereas

Dave Jones: that's how everyone got hired at valve right because they're all bloggers they're all putting their work out there yeah yeah and and that's how they were found you got to put your work out

Greg Charvat: there these days exactly yeah but but for all those who haven't but have done cool stuff you know send me a note we want to see what you've done you know it's we find this exciting we're trying to build up a team to fab boards very very quickly yeah oh that is that is fun i can't wait i hope

Chris Gammell: to get involved so i can learn how to do this so so what's stopping you from i mean i'm sure it's a time thing now but i mean you could very easily just go by you know eagle or kai cat whatever else

Greg Charvat: right you could just that's a good question i you know what i i never fab i never build something just for the purpose of of building it i want to learn something from it and it it's part of some bigger things so for example the lat the ham radio that i built recently the six and two meter transceiver i they're all dead bug or through you know or perf board and i couldn't justify the time learn that

Dave Jones: for that that radio i just couldn't and a circuit board is just another means to the end of

Greg Charvat: producing something and learning yeah i could still get there without the yeah and also it's it's like some people like your first board being rf it would have been rf2 well yeah well it's like

Dave Jones: you know if people ask me why don't i make my own boards at home anymore well because it's just a means to an end of getting the board that i want why waste my time doing that i don't care anymore you know

Chris Gammell: i just but eventually that that argument you know like eventually you have to you know like like i always do that too right i say the same thing but then you know 50 boards in i'm like man i should have done that 50 boards ago it's the same thing i mean like yeah i i yeah you you don't

Dave Jones: want to spend your time right manufacturing a board you just want to design the board send the file out and bingo magically it comes back one day that's you know you don't want to waste time doing that and

Greg Charvat: like imagine that's the same for you great well oftentimes my career objectives have have been to prove a concept as soon as i've proven the concept then i move on to the next project so nothing's volume i'm you know everything i've done is one so you've never finished anything

Dave Jones: you've never taken it through the production you're you're you're one of these 90 engineers that's right like us ah yeah i'm finished playing with that next that's right well once it's done you know i got the the 23 tube

Greg Charvat: home theater okay i'm done that works move on to the next thing let's build this ham radio now let's build this through all radar all right it works what's the next thing yeah

Chris Gammell: that's interesting as you move into like dfm situations i'm sure you'll uh i'm sure you'll learn a lot at that point right that's like uh oh just design for manufacturing type stuff oh man

Greg Charvat: see i don't even know what that means that's how bad i am yeah oh sorry i'm going to have to learn that's why i need help well that's just a sorry it's an acronym that's why we need we need people to join our company to help us with these things i don't even know what dfm means see how much help we

Dave Jones: and then you'll find out by not having anyone in the company who knows what dfm means you'll find out that doing a hardware startup is probably not the best idea because it's a pain in the ass

Chris Gammell: getting something manufactured oh boy did you guys really talk about the the preamp you built the the

Greg Charvat: the buddy guy show thing no we haven't in fact it's no we have what was that i mean well it's sitting in front of me right now actually really yeah yeah it's uh how big was it well the uh preamp itself was pretty tiny it's uh looking at probably about um four four by three by an inch and a half tall plus two tubes sticking out of the top and a transformer for the high voltage supply

Dave Jones: and the knobs and things now it's obligatory that the tubes have to stick out the top so you can see the glow oh yeah yeah absolutely it's obligatory yes yeah there's absolutely no point building a valve

Greg Charvat: bit of gear unless you can see the tubes i don't see the point personally yeah we got to have them out there wank factor folks there has to be a wank factor right and and you may as well put the transformer out there too you know because that looks cool and then and then i thought for for aesthetics i put the knobs right next to the tubes right between them even you know so the gain and the pre and post gain are right there so it looks like it kind of looks like um an old school prototype of some kind but it's got a good finished look to it a nice military look to it so what was the story behind actually getting caught with this thing well you know so i built this thing because my my friend of mine was a dj at the the michigan states radio station the impact which is they have a web radio it's actually it's a great radio station if you're into independent music and uh i think it was on wednesdays they have this program and they still have it's called accidental blues and they play all these blues you know uh you know songs and whatnot and this guy this guy his name is ryan noise he's a really cool guy and he knew the blues so well he used to trade he'd trade bootlegs with people and this is back in like when you would post your list on your website your web 1.0 oh yeah geo city's website yes and then you get an email saying i'll trade you this cd for a copy of this cd and you

Chris Gammell: would do that's how things that's how it happened they weren't like the high fidelity tapes even like the uh the real the real no we were past that at this point but not much further past or mini disc

Greg Charvat: mini disc was another one that was well in fact we use a meatus recorder for that for this uh for the show that's what we used oh okay yeah so we were well within the mini disc uh era and so did you have this high volt i mean you did like a boost converter or what i uh no it was real simple it was just a um 555 timer the ubiquitous 555 and it fed into a uh a one watt audio amplifier ic which fed the secondary of an old military um audio output transformer so it fed into the 8 ohm side and then out the 3300 ohm side came you know a couple hundred volts yeah and it's perfect and like you and you crank up the frequency so it's really high so you don't hear the hum when you're recording and it's easy to filter it with uh small capacitors because if you had to use the big you know electrolytes there's no room you couldn't cram in the preamp yeah um and uh yeah so we built that it actually was on the front cover of audio express magazine um so we built this thing i built it because my buddy who is the dj wanted to record this buddy guy show in chicago at buddy guys legends bar and so we we had this you know a bunch of 21 22 22 year old college guys you know road trip to record the show and uh you know lots of things seem like good ideas when you're that age and uh we get in the car and we're going down to chicago four hour drive we're like yeah it's gonna be great you know and i show everyone the preamp and everyone's just really fired up about the preamp and my friend's like this is gonna be awesome this sounds so good i i got this uh lapel microphone it's like a really high-end one i brought it from the station it's like a 500 mic i'm like yeah great this is perfect then here's the here it's got this nice little thin lapel wire you know you know that goes between your clothes it's really like thin then it's got this you know this big plug on the back or whatever the three-prong mic plug whatever you call those oh xlr yeah xlr and then and then it's got to plug into this 20 foot roll of half inch cable to go to this other weird adapter that's six inches long that's a transformer in it to plug into the tube preamp because i had high impedance 50k input they wouldn't talk to the xlr that was our first problem all right so we're like okay whatever and and so we so then i'm like okay but whatever we do guys we have to leave room for the eight double a batteries that hang off on a separate battery pack off the back of this damn thing okay fine you know so we put this whole thing together it ends up being you know about uh two bread boxes in size of cable and batteries and crap you know walking looking like a transformer i know well good thing it was before it was before september 11th they would have thought we were terrorists or something you know with all this all strapped yeah yeah just all strapped in there trying to be covert so so my buddy's like we got to do this we got to do this i'm going to throw in a backpack we'll put some dirty underwear on top of it and some clothes and it's just a bar they're not going to check our backpack so we go in there and then they didn't care and so we went in with the backpack and uh we set up underneath a speaker a pa speaker and we showed up like three hours early before buddy guy went on so we decided that uh you know we're 21 22 23 year old males from a big 10 university we decided to have a few libations before the show and uh many rounds later my friend ryan forgets how to turn on the preamp i don't know so when we go to set the levels because you couldn't leave it on the whole time three hours before the show we'd kill the batteries because we got tubes there right you can't leave those film it's lit they'll just suck the batteries down so we left it off and uh he couldn't turn on and then one of the bouncers saw him and grabbed him and the backpack and yanked my friend away to a back room where they interrogated him about recording the show because i don't even know why they even care but they did and they kicked him out onto the street chicago and in january freezing cold out there you know like 10 below zero jeez and so we're standing so we're standing there and my buddies and i who are left and we're like what should we do you know our friend ryan you know you know brought us down here for this but we have his pitcher of beer with us he gave it to us when he got yanked out and we have his 25 he gave us to go to the bar to pick up more drinks so i said you know what i say we stay here and we drink his beer and we watch the show because that's what ryan would have wanted that's what he would have wanted yes and we'll spend his money too at the bar because that's what ryan would have wanted and that's what we did he got hypothermia whatever he was freezing out there oh yeah yeah it was rough oh it was rough we had to come back the next morning to pick up the the the recording rig they didn't break it or anything give it back to us no no they didn't and we went up to the door to ask for it back they're like which one's yours because two other guys got busted for doing that that night they were nearly as nice and no and so we told them it was the tube rig and they're like oh you guys are the tube rig wow that was cool you know so anyway that was fun oh that's a great story that's what happened with

Dave Jones: that thing now you're a ham aren't you yeah yeah and you've seen this video i sent you before the show oh yeah those guys we'll post this link this link in of a guy called julie kentwell he's an australian here and uh i you agree do you agree with his uh quote from this video where he says uh there are two types of ham shacks those that are an absolute pigsty and those where nothing happens

Greg Charvat: i think that's fairly accurate and there are those with appliance radios where they're really clean

Dave Jones: they're really clean and they've bought their two thousand dollar yazoo transmitter and that's all

Greg Charvat: they have and you know automatic antenna tuner they don't even have to turn any knobs anymore right and it's hooked up to their computer so they use a mouse to tune the radio right which i just i don't even gosh what's the point if you're not going to turn any knobs and then there's the

Dave Jones: there's the next step where sending moss with their mouse you know with their mouse you and it doesn't

Greg Charvat: oh they don't even send it they just type it they type it and then the computer picks it up and translates it for you know there's no telegraph key anymore yeah i don't get it that's not me

Chris Gammell: uh then there's the worst the worst type of ham operator where they get their uh technician license and then don't do anything with it that would be me

Greg Charvat: well i haven't been too active either i see my thing is it's always been like this for me and ham radio i like to i like to build the radios and try them out to prove myself they work and then i

Dave Jones: almost never use them again yep actually communicating with people what's the point of that you know

Greg Charvat: i use it for a week and i'm like that was a big achievement we got this thing to work we got the design works everything's fine okay move on to the next project sweet oh boy

Dave Jones: how amp hours up it is sadly i have one last thing i want to answer a question go for it okay someone

Greg Charvat: wanted to know about you know someone said that i saw greg's website and some pics of his mother's senior design project orbit adc yes yes my mother is an electrical engineer and she's the one that

Chris Gammell: got me into electrical engineering wow there you go yeah so yeah where where did she go to school and

Greg Charvat: like what where did she did she work where does she work she went to school at lawrence tech um and uh which is a tech is an engineering school outside of detroit it's very hands-on and it's a co-op school and uh she's now working at uh ford motor company still there she worked at ford and gm with bounce back for a few times and she's at ford right now to this day uh she's now in purchasing she's got

Dave Jones: an mba and she's doing supplier quality so i know yeah oh you can just hear the collective groan of the 5 000 amp hour listeners yeah but then she knows all the tricks of the engineers that's even worse a

Chris Gammell: purchasing agent that used to be an engineer is a very dangerous person it is it is and it keeps

Greg Charvat: them going over she does a great job over there and keeps them going made it through all the cuts and everything so that's sort of oh yeah got me into it uh it's my mother so do you guys and you said she got you into ham as well yeah yeah well what happened was she's a hopper as well she's a ham as well i remember i was i was into cb radio when i was 10 i started getting the cb radios and i got all my buddies every kid on the block had a cb after i was done with them and we would talk to each other this was in 1990 when there's no cell phones or anything we yeah we talked to each other that was still the cb craze was 70s oh it was it was no one was on the cb except us that's right that's great

Chris Gammell: and if you're not near a highway you have the band yourself oh man it was great we talked to each

Greg Charvat: other you know secretly at night and we'd have like five you know five of us ten-year-olds on there you know it was so much fun that's awesome yeah yeah and everyone everyone had a cb in their basement from the 70s so it's easy to get people right and i knew how to tune up the antennas and get the mothballs out of them but anyway my mom said you know i was doing ham radio or cb radio for a few years so i was like you know right around age 13 she goes i was buying another cb radio and she goes i don't want you to buy waste money on another cb radio you're not learning you know from that as much as you could from amateur radio i don't want you wasting money on another cb radio let's get into amateur radio and my friends can't talk yeah they can't talk to but but i got all my buddies in a ham radio too so i brought them along oh that's good yeah yeah that's awesome but her thing was it was expensive it was getting expensive that cb radio hobby let's do ham radio

Dave Jones: build your own i'm so jealous growing up i didn't know a single person in my entire universe that was into electronics oh yeah until i was like yeah until i was like 15 16 and started to study this stuff i i knew absolutely no one that's it oh it starts early you gotta start early you know it

Chris Gammell: helps do you do any like outreach stuff at mit for like uh doing the same kind of thing because so i

Greg Charvat: i officially i'm not um i'm sort of peripherally involved because i'm down in connecticut now focusing on the startup so whenever they want help i go up there but and help them but uh there's a high school outreach that that was a program that was done this this this past summer where the high school kids built the coffee can radar sweet oh wow and it was great it was like a huge success they had a blast because they actually got to do something cool i mean that radar is a lot of fun it it really works like you can measure cars and people running down the street all kinds of cool stuff so it's a lot

Chris Gammell: of fun yeah that's awesome that could be another reason to you know kid it up more right so oh yeah like we said if any listeners are out there yeah get kidding yes seriously and let me know so i can

Greg Charvat: tell other people i get emails about you know two three times a week asking for a kit of that and i you know i can't help them it doesn't exist so yeah wow all right well thank you very much greg for joining us oh thanks for having me guys yeah thank you for having me it was a lot of fun and where can we catch you oh you know what go to my blog mr vacuum tube that's where you mr vacuum tube mr vacuum tube and on twitter as well right that's right mr vacuum tube is also on twitter and uh you'll see every uh fun engineering youtube video i like it goes up on twitter for some reason awesome oh yeah i'm subscribed but mr vacuum tube the good stuff's on there you'll see that i put a lot of stuff up on mr vacuum tube yeah well thanks for having me guys it's it's i love talking yeah i could go on for hours and hours

Chris Gammell: i think we need to have a roundup of all the uh rf people we've had on at some point and uh just have it out and need to able to sit back and listen yeah yeah yeah yeah talk about our trips to the

Greg Charvat: date and ham bench and stuff and the guy with the antenna's head no you go to that a lot oh yeah my entire lab is equipped from the ham bench and everything nice yeah it was during maker fair

Chris Gammell: this year i hope to make it next year but it was during maker fair same weekend this year i would go

Greg Charvat: to dayton over maker fair if you're trying to staff up a lab or in terms of you know equip a lab rather i would go there right right man you can get anything at dayton anything it's true i mean i once saw a

Chris Gammell: missile at dayton that's like dave on a missile at that uh at that junk store yeah really yeah how

Dave Jones: much was it yeah oh it wasn't actually a missile it was just a nose cone you know okay this was a

Greg Charvat: missile it was a sparrow air to air missile oh wow yeah really yes is that guy wearing a tinfoil hat

Chris Gammell: i don't know i couldn't imagine i got news in case they come for me my only question was how do you

Greg Charvat: take that freaking thing home do you strap to the top of your car and drive down the freeway with

Dave Jones: it oh boy god i'd love to see that tsa rubber glove time folks oh man yeah it would be really bad then

Greg Charvat: you should have a good excuse you'll be making the six o'clock news all right well thanks again greg we really appreciate it thanks for having me guys see you mate all right see ya

Speaker ?: you you