- 3D printer upgrades
- Waiting on parts
- Printed the pump brackets
- Bending Tubes and mounting brackets tonight
- Frequency counter woes
- Found the led driver. It is not a pic like was suggested
- Translated the document
- Page 7 has a schematic that looks like it will help
- Rotary switching
- Doing it “right” this time
- Purchasing the switch before designing the board
- Part numbers
- STmicro website?
- Dave Jones and documentation
Rapid Fire Opinion
- Wanted: Practical App Note Suggestions
- Chris Gammel posed the question on the The Amp Hour subreddit
Special thanks to whixr over at Tymkrs for the intro and outro!
About The Hosts
Parker Dillmann is MacroFab's Co-Founder, and Lead ECE with backgrounds in Embedded System Design, and Digital Signal Processing. He got his start in 2005 by hacking Nintendo consoles into portable gaming units. He also runs the blog, longhornengineer.com, where he posts his personal projects, technical guides, and appnotes about board layout design and components. Parker graduated with a BS in Electrical and Computer Engineering from the University of Texas.
Stephen Kraig began his electronics career by building musical oriented circuits in 2003. Stephen is an avid guitar player and, in his down time, manufactures audio electronics including guitar amplifiers, pedals, and pro audio gear. Stephen graduated with a BS in Electrical Engineering from Texas A&M University.
Host 2 00:10
Hello and welcome to the macro fat engineering podcast. We are your hosts, Parker, Dolman. And Steven Greg. This is episode 218. So last week, Steven bought a really like cheap frequency generator.
Host 1 00:26
Sorry, not a frequency generator. I totally wrote our notes wrong. It's a frequency meter counter counter counter. That's it. It's it counts the way I totally wrote my notes.
Host 2 00:37
Yeah, no, I looked at that. I'm like, wait, wait, wait. That's not right.
Host 1 00:41
I knew it. But like, yeah, my head was on autopilot. Yeah.
Host 2 00:46
And you were going to look into hacking it for adding the ability to record measurements. Yeah.
Host 1 00:52
So I actually will Parker gave me homework to actually go do that. He said, I expect by next podcast, this one, that I'll have something and I do. I have something. Like not a finished thing. But I have something. So I correct. Okay, so actually a couple of given awards for partial credit. I'm totally reaching for the stars right here. I expect a sticker at the end. So I did, I did crack the counter open. And it is as much of a piece of shit as we expected. Now, don't get me wrong. It's $100. Amazon wonder it's what we can expect? Yeah, well, so I have something before before I get into it, I actually have something that's interesting about it. Remember, I was complaining about it didn't settle very nicely, like it had the values jumped around a little bit. And, you know, I'm splitting hairs because the values jumped around, like in the 10, the 1000s decimal place on, you know, like, I'm reading 100.056 And the six digit is jumping around. It's flipping for six or seven every so often. Yeah, well, maybe a little bit more than that. But you know, I of course, I hate it when it does that. And I was complaining about that. So I cracked open this thing. And it has like an individual Well, it has three PCBs in it has the main PCB that is obviously doing all the actual counting. It has a secondary PCB that has all the Display Digits on it, and then a third PCB that the BNC connectors are directly soldered to, and it's like the input, I guess preamp or whatever, it does all the input handling, because there's a variable attenuation and things like that. So it's handling all the input stuff. So that board it was supposed to have or be completely surrounded by metal shielding. I opened it up and it had a metal shield on the top side of the board and the bottom side of the board, but only like the very corners of one side of the cans were soldered and it was like clamp shield open. I guarantee you was doing zero shielding whatsoever.
Host 2 03:15
That's a unit that someone tested failed, and someone opened it up to fix it.
Host 1 03:22
No like the pads weren't even soldered it what it was was whoever was assembling it just did an awful job. Didn't solder any of the points basically just tacked it on and shipped it you know, like guarantee it was not tested
Host 2 03:35
China factories have Friday for 30 units. Yeah, this sounds like 13
Host 1 03:41
Yeah, so I will Okay, cuz I had to disassemble everything in order to get to the what I was really looking for, which was the LED driver or the display driver on the front board. So the design of this device was clearly meant to never be taken apart because like the stackup of how all the boards and everything kind of fit together was like soldered in place like the boards were put in then everything was soldered, you really can't get it apart. I had to do some destructive things to get it apart. And I end like when I was taking it apart like the button caps that actually you know fit on everything they're just held in place the it's just like an interference fit with the switches and things like that. So if one of those falls inside the chassis, you have to like de solder at all just to like fix the buttons on the front. It's this is a one time assembly device. So what I was doing was foolish, but But Okay, so back to that preamp or the input board or whatever you want to call it a conditioning board. I re soldered and receded all of the shielding and it's significantly better now like it settles nicer. And previously if I didn't have anything connected into the BNC It would sit there reading 60 hertz, go figure. So now it just it's rock solid at zero, because it's actually shielded, which is nice. So, you know, all of what I was going for with reading the information and averaging it is a little bit moot now that I actually have like proper shielding in there. But I still want to, you know, investigate it a little bit further so so on that front display board on the backside of it is a is an IC that socket it in. And from its socket ID, yeah, no, it's actually socket ID. This is not the SMD version. Not entirely sure what we'll know. I guarantee you it was because for some reason, the dip version was cheaper. That's got to be the only reason every design decision in this thing was because it was cheaper. Yeah, that's
Host 2 05:49
The thing is its own. I bet you if you took every take that chip off, was there an SMD pad underneath it? I haven't pulled it off yet. Okay, maybe I'm not sure there's the cuz I'm looking at the part right now you haven't. What's the part number?
Host 1 06:01
So the part for this is z l g 7289. Bs. And from what I had researched before cracking open my unit people were saying that this is was a PIC microcontroller? This is not at all a PIC microcontroller. This is like a cache used to have a term for this burger like ancient Chinese mystery. ICS.
Host 2 06:24
Oh. It was like ancient Chinese. Micro. What ancient Chinese Mike? Circuit secret. I can't remember that was like way back in episode like 10.
Host 1 06:37
Like the very first day I started it at macro fab. Like you showed me some kind of list where you like you are compiling ancient Chinese data sheets or something like these, like super secret. Indiana Jones like going deep in the cave to find the data sheet for these. This is totally one of those parts, though. Oh, yeah. This is absolutely one of those parts. So this is that part where if you search the part number on on Google, you don't you get a whole bunch of gray gray market websites. And just some like random stuff.
Host 2 07:07
It's on LCSC? Well, it
Host 1 07:11
Is yeah. And that's actually where I was able to find the datasheet. For this. Unfortunately, the entire data sheet is in Chinese.
Host 2 07:20
Is I'm looking at right now. Well, but
Host 1 07:22
I found that I found a website that that you can just upload PDFs and they'll translate it and kick it back out. So I totally did that. And I have a pseudo English version of the datasheet now. And so from last week, Parker was guessing that it is an SPI based chip that just drives all the displays. And you're absolutely right on that. So the main processor, which is down on the main board, just shoots a bunch of SPI up to this chip. And then this chip handles everything on the front board driving all the displays. So this ice is also a keyboard scanner, too. So it can do in in outs, but they're just using it as a display. Interesting. Yeah, I'm looking at that the schematic right now. Yeah. And so on page seven of the datasheet, which we will, I guess we can have the datasheet or post a link to it or something like that. On page seven, there's a fairly nice little schematic that's just like, Okay, great. So like, here's all my ins and my outs and things like that.
Host 2 08:25
I just remembered what I call the ancient Chinese semiconductors.
Host 1 08:28
That's it. Yeah, yep, that's it. I went as soon as I got this data sheet. I was like, yes, that's, that's on that list for I'm actually looking for that list, and I cannot find it. Oh, Dan, I would love the like, because I know you and I looked at that list a handful of times,
Host 2 08:45
Because that list was I bet you it's it's live in somewhere in my Google Drive in the bowels of hell. Back in 2014. Right. When
Host 1 08:53
We came up with the idea for the podcast, like we were talking about, like, oh, we need to have a segment where we talk about ancient Chinese semiconductors.
Host 2 09:02
Yeah, we did. We talked about that in the podcast. We Yeah, but
Host 1 09:05
Like way early on. So yes, I was kind of excited to see this because it was just like, oh, this is old school, right. So I haven't dug too deep into the datasheet. The thing that sucks about this datasheet is it's like it's kind of like a, you know, the first page is just like a picture, basically, of things. The second page is electrical data and like the absolute maximums. Page seven is a schematic and then all the rest are just registers. And that's it and not a whole lot of good information on like, oh, this does this. It's more just like register maps. So it's kind of the crappiest way of having to reverse engineer because you just have to look at a register map and say like, oh, well, what's that doing? So if I see that, here's what that's doing. You know, it doesn't seem to have or at least just like on a first call. answer doesn't seem to have like a, if you want to write this here is this, you know,
Host 2 10:05
I think it's, I've done a lot of, like, low level C work. This is pretty standard.
Host 1 10:13
Yeah, I mean, I've done a bit too. It's just like, there's not like a, there's not a clear portion in the datasheet. That looks like to write a number. Here's what you do. You know, that's what I'm not seeing.
Host 2 10:27
You want like a, they do have that? It's just an assembly?
Host 1 10:32
Well, yeah, okay, so back at the at the end, the very end
Host 2 10:35
Is actually assembly, which I'm gonna guess is probably. That looks like maybe. What was it? What's that core at? 51? Yeah, yeah, probably at 51. Assembly.
Host 1 10:55
But like, all of the comments on it are translated. So some of them make sense, like, delay 10 microseconds, that totally makes sense. But then like, some of the other ones are like, by Z O G, 7289. A receiving a byte of data, the previous high, like, banks, now I have to actually read the code to figure out what you're talking about.
Host 2 11:17
I think the most important thing is actually the on page 60. You got a timing diagram. Yep. And with that, you can easily just reverse engineer the protocol. You don't need anything else, you know that. Like, Chip, select goes low. And Woz. Data is valid on clock.
Host 1 11:36
Hi. Looks like yeah, clock. Hi.
Host 2 11:39
Yeah, and then you and then the little bit of information above it, it looks like it's doing an eight bit the 16 bit bursts.
Host 1 11:47
Yeah, right. And I can I can hook up my logic analyzer. And basically look at what's displayed on the front, and then figure out what that actually was. Yeah. So that's what I would do that I mean, that that is the next step. It's just not like, I it's a little bit of reverse engineering, as opposed to just like looking at this chart. Oh, yeah, figuring it out. So luckily, it doesn't look too ridiculous, it just a few hours of work. So the idea is to just take something as simple as an Arduino and connect it to those lines, the SPI lines, and then just sniff off what I'm seeing there. And as soon as, as soon as I can sniff it off, and like accurately figure out what the data is intended to do, then everything after that is really simple. You know, just interpret and modify, translate into whatever I want it to be, which is just averaging basically, this is a lot of work for just averaging values,
Host 2 12:51
But it's a lot more stuff down now you can do, you can record over a long period of time, right? So you can see drift and that kind of
Host 1 12:58
Stuff, ya know, there's a bit more behind it, that makes this a little bit more powerful. And actually, if it's just an Arduino and four lines, this could be a pretty cool little thing to just post as a hack for these things to put online in case anyone else wants to do it. So I bet you
Host 2 13:17
A lot of people would like to do that. Especially if you can basically say for sub $100, you can get a frequency counter that has. Yeah, that has that data recording. Yeah. So yeah, I'm looking, I'm looking forward to that. Because basically, when you solve that, I'm probably going to do it. Oh, yeah,
Host 1 13:37
Yeah. So the best part about it is like all the garbage about having to like break it all apart. And you know, all the all the desoldering, you don't have to do any of that. If you just pull the lid off that ice is available from the backside of the board. I haven't even figured out what pins are there. But one half of the IC is really easy to get to the other half is a little bit more difficult, but not like impossible or anything. I wonder,
Host 2 14:05
Is there a way you could pop the chip out and then sandwich a PCB in between it?
Host 1 14:09
Yeah, there's tons of vertical space. So you could totally do that. Because that's
Host 2 14:14
A lot of modifications for old old equipment that are dip socketed parts is you pull the old chip out, and then you put a PCB that's gotten pins in it into the old socket, and then plug your chip back in. And that way you can sniff all the signals
Host 1 14:29
Correctly. Actually, that would that's a great idea. Like if I get this all working, I'll totally design up a little daughterboard for that. That. Yeah, you could just have four wires soldered to that. daughterboard and then plug those right into your Arduino. Yep. Yeah, that's a great idea. totally do that.
Host 2 14:47
So last week, I was talking about my 3d printer. And unfortunately, you know, I ordered all the parts still waiting on him. Oh, I
Host 1 14:57
Think you're gonna see like something caught on fire or something like
Host 2 14:59
No No, no, not yet. So yeah, I'm still waiting on parts. It's kind of partially disassembled. I'm still trying to figure out a solution for the cable management. Mainly because I got like I had that loom, I have some split loom, and that the that two by 10 wide connector, it's like a ribbon cable just doesn't work. And so I'm trying to think if I can just replace that with I talked about last week like I round cable, a two by 10 Round keep, like, make a custom one or something. Yeah. I'd look into that more. Probably do that tonight.
Host 1 15:37
You know, one of the things we were talking about last week was lubrication and grease for it. Yeah, a
Host 2 15:43
Lot of people were commenting on that. Yeah, I
Host 1 15:45
Was noticing. I read that earlier today. In the in the Slack channel, there was a lot of good ideas behind that. What do you end up going with?
Host 2 15:52
So I went with, because people like sewing machine oil was really good, or lithium grease is what your suggestion was. And a couple actually other people who do CNC work, were suggesting that super lube 51 004 is what a lot of people use as well. And then someone mentioned that they use what is it hopped number nine? No, that's what I ended up using. Oh, really. They use break free, which was gotten me on hops. Number nine. So break free is like a specialized gun lubricant. Yeah, but it's a really good light oil that's penetrating fluid as well. It's actually really nice for because it just gets in everywhere. So it's really good for small roll pins and stuff where you need a when you have a lot of movement over small surface areas that are really hard to get lubricant into, like getting a lithium grease in there, or any kind of grease will be really hard. Yeah, break free is really good because it's kind of a penetrating fluid. And it'll just like soak everything in it in it. Yeah. And so I'm like, one of hops number nine, which is like the generic gun. That's like Walmart gun oil. Yeah, it's like the everyone's got everyone that's a that owns guns has because it's like, it's like comes in your starter set for cleaning guns. Yeah, because it's good at cleaning. And it leaves a protective oil finished. It works fine for just general lubrication for your hand guns and rifles. So I use that and it works. Nice now longevity wise, no idea. But it cleaned all the old stuff off and left a nice smooth finish and left a little thin oil on it. And it seems to work great.
Host 1 17:44
Just out of curiosity, where you didn't make any difference in like sound.
Host 2 17:50
It is quieter is cool. Yeah. Mainly because it was I was starting to see. So you won't see it in this print. But I can see it. But you can see has a little bit of a step here.
Host 1 18:03
Oh, yeah, I'm seeing that like at the bottom like half millimeter of one of his
Host 2 18:08
Prints. Yeah, it's actually it's actually shifted over. And that's because it missed a step. And when usually, since I have never changed the settings on this printer, it's always running the same speed, it's well under the speed rating. Basically, the friction is building up and it missed a step. That's interesting. So cleaned it up. And that went away. Because it would happen at random spots in a print basis. Basically this the printer would make a really weird noise and you'd look at it. And it would look fine. When you pull your printout and it's got this weird step and step in the Z.
Host 1 18:46
Host 2 18:48
Annoying. Yeah. That's all fixed now. So that work now what I advocate this to anyone know what, number nine? I haven't looked online if anyone's ever used it for long term success, but seems to be working great. Now. It could be eating my bearings alive. And I had no
Host 1 19:06
Idea. It's super corrosive. But I mean, it's a gun oil. It's not I mean, it's not it's not corrosive. Yeah, so
Host 2 19:12
But it could like be trashing it and and causing more wear, it just sounds. Usually when a bearing surface sounds quieter. Generally that means it's better.
Host 1 19:23
I mean, you had an inflection on your voice like you're asking a question. Yes. I don't know where electrical engineers. Yeah, we don't really don't deal in the goofy stuff.
Host 2 19:35
Goofy stuff. So I'll, I'll post back in a year. If hops number nine is the way to go. That could be the like the winning formula because it's a cleaner and oil built in. It cleaned all the old, varnished oil off. So I'm pretty happy with it so far. We'll see. Nice
Host 1 19:59
Oh, nut on that m one grand that I bought a year ago or two years ago now. The like you originally it wasn't. It wasn't. It wasn't oiled. It was greased. So I actually ended up buying what is a mobile one just like the red automotive grease, because that stuff is like pretty close to the original grease. The M one was with yo was, you know, lubed up with that stuff. It's pretty great for two surfaces moving against each other, which is exactly like the M one has plenty of moving parts like that. So well that's
Host 2 20:36
Actually some people's thoughts is the use a heavy grease. So it sticks around on your printer on the slide surface. It really depends.
Host 1 20:46
Well, that's why that's why the lithium grease came to mind right away. Yeah. And like the CNC that I that I run it work. Like whenever I agree that that stuff is pretty viscous. That mean that stuff is like earwax.
Host 2 21:00
So I think he's just different schools. The hopslam are noticing is working great. Yeah, we'll see. It's a $450 printer. So if I ever gets trashed, it won't. It's not the end of the world. Right? Right, right. Unless I'm printing like COVID 19, ventilators, then it would be the end of the world. I guess I shouldn't joke about that. But yeah,
Host 1 21:26
I hope everyone's safe. Me too. I'm actually staying home today because I am not feeling fantastic. So Oh, yeah, exactly. I've had like that persistent headache.
Host 2 21:39
Is your wife just throwing beer and food down the stairs there into your basement?
Host 1 21:45
Designed a little beer elevator that comes down
Host 2 21:48
To just the size of a TV tray? Yes down.
Host 1 21:53
That's a good idea.
Host 2 21:57
No, it shouldn't be your fridge, you shouldn't be have your fridge should be on a freight elevator.
Host 1 22:03
Or just the bottom of the fridge because comes down into the base goes on.
Host 2 22:07
Now, I can't remember what movie it is. And I've always wanted to build this and my mind didn't exist it like a thing. But it was a 70s or 80s movie. And it was a frigerator that was round. And you in the person press the button and it pops out of the counter. So it's completely round. It looks like a spike are one of those spinning spice racks. Yeah, except it's a frigerator it's bigger than a spice rack too. And it pops up and you can like grab stuff off of it. And then you press the button and it goes back down into the the counter. Now you lose like all your cooling capacity when it pops out. And all the air you know goes everywhere to the rooms not very efficient. Well, no refrigerators aren't either you opened up and your feet get cold. All the air, all the cold air just fell out of your frigerator
Host 1 23:00
There that seems like something that would have been in 2001 A Space Odyssey you know, could be
Host 2 23:05
I don't remember what movie it is. And I haven't found it since. So with the power of the 6000 plus people that listen to this podcast. Someone knows. Yeah. So let me know on Twitter or slack.
Host 1 23:21
You know, back when I was in college, a buddy that I knew actually designed a beer launcher that was that was connected to his fridge like it had like a like a pneumatic autoloader Yeah, it had like he would fill his his fridge it had a magazine like a beer magazine. And, and he had these little discs that were like different colors and stuff like that. And you could hold up the disk and the fridge would see it and it would launch a beer at the desk. He spent a ton of time making this beer laundry just so that like it could he wouldn't have to get up from his chair. He could sit on the couch and just lift this disc up and it would shoot a beer at school. Yeah, it was pretty cool. Except, I mean, the beer was like super foamy when it got to
Host 2 24:13
The vinyl wrap your couch then? Yeah.
Host 1 24:17
Well, so remember that. The macro amp that I designed beginning of last year, or Yes. Well, yeah, I mean, not just last year. I mean, I did five years ago or whatever. But I've finally built it. Yep, it's the one with the new tube. Right, right. One of the things that that didn't go well in that design was some of the rotary switches that were in there. I have Rotary, three rotary switches that were that controlled the volume but also like the balance between left and right. And everything in the amp worked well, except the rotary switches were wired incorrectly or the footprint was incorrect, because I didn't actually have those footsteps. Yes. And I guess then I guess wrong, such. So I actually have a project coming up here soon. That's sort of like a mini quick little thing that that I'm going through. And I'm using those switches. And I'm being a good boy this time around. And I actually earlier today bought two of those switches in different varieties, and I'm not touching anything until I get them in hand. And I'm actually going to, like, measure them. And because I was thinking about it, like I could make generic footprints where like, you could put it in any direction, and it would work. But I don't want to do that, like I want it to be like, just right from the get Yes. So like, I'm taking a, I'm doing it right from from, you know, from the beginning here. So whenever that comes in, I'll actually take a couple of pictures of it. Because every datasheet seems to contradict itself for this particular type of rotary switch. And this is the cheapest rotary switch out there. actually really liked them, they feel nice, and they last a long time in their beefy, they're just enormous. So your, what your project has to be able to handle the size, you know, which whatever, I don't care, this project doesn't really particularly matter, I'm actually making a I'm making little design, like sort of development boards for some projects I'm working on with a buddy of mine, where I want to be able to switch in a variety of different capacitors. And these are perfect for that one of them is a what is it is a it's a two pole six position. And the other one is a one pole 12 position. And 12 Different cap values is more than enough for what I'm looking for. And these things are like, a little bit over a buck, which is great for that. So but just I don't want to like make these nice boards and go through all the trouble and then figure out that the footprint was upside down or something stupid like that. Yeah, we're like, we're in the wrong place. That made me so mad when I got those boards. So you know, I if I could, I would do that. For every component on
Host 2 27:07
Wrap the hook the part numbers up on our,
Host 3 27:11
You know, the thing about these switches that are annoying is like a bunch of people sell them SparkFun sells them. I bought these from tubes and more. And Mauser stocks and stuff. And they have different part numbers at every place. They're there, that kind of component. You know, they're an ancient Chinese rotary switch, you know, okay, so have you ever searched for rotary switches before? Have you ever had a need for that?
Host 2 27:36
Kind of, not as intensively as you'd like. So for the wagon years, like the mirror switches I built? Those are actually rotary encoders. I'm not using the rotary encoder parts. Yeah. But there they have encoded so like, for a project? Not really, I have used them. So my answer would be no.
Host 1 28:00
Well, okay, so we've we've talked about selecting components a bunch of times, and like, yes, you know, searching for connectors sucks. Searching for capacitors is okay, searching for resistors is usually pretty easy. Searching for rotary switches. really sucks. If you asked me just because like, they're hard to find. There's not a lot of stock, in my opinion, they go from like, $1 to $20. instantly with no reasoning in between. Yeah. And and they don't seem to have a lot of rhyme and reason between like, a lot of other electromechanical components, like they fit within paradigms that work between manufacturers, Rotary switches play by their own damn rules, like all that, like, there's not one that plays nice by everyone else's rules. So I try
Host 2 28:49
To find like, what's the what's the footprint? I know borns makes a lot of them. But I know a lot of other manufacturers who it's that potentiometer footprint that's got like, two lugs, that straight it's a vertical potentiometer at soldered onto the board. Yeah, a lot of a lot of stomp boxes and stuff use them. But it's like that same footprint that's got three, all the terminals are all in the same spot. Everything's the same size. They have the same kind of links, and a lot of manufacturers build those kinds of potentiometers too, right? No one builds a rotary switch like that.
Host 1 29:24
Exactly. Okay, bi technology, alpha. There's there's a ton of other people that make that exact pot. And what's nice about it is like not only not only is it all standardized, it's standardized around nice numbers, like 10 millimeters, like that's a really nice number. Try to find a rotary switch that fits in 10 millimeters. It doesn't exist, you know, and probably because people don't have lots of needs for rotary switches, and everyone's needs are going to be pretty different. So that's a very custom part. I get that but it's also like If you have a situation where a rotary switch would work, well, it'd be like, Ah, man, come on, like, I really wish I could just get a rotary switch for this situation without having to go get something custom. Like, as soon as you have to go get some custom, then it's just, it's such a pain. You know, it's funny, you'd be surprised how many designs I've done and even stuff at work, where we're like, okay, so how can we electrically switch these signals to simulate a rotary switch? And by using an encoder and a processor, as opposed to finding a rotary switch? Like we've done that before? Because it's, that's easier. And it ends up being more expensive. Now, whatever. So I, you know, I'm doing it right. This time. I'm getting the the rotary switch before I designed the product.
Host 2 30:52
So, a couple of weeks ago, I was talking about the Bruery Yeah, it will probably because of you know, COVID-19 so many
Host 1 31:01
Excuses. That's all I'm here is excuses. Excuses. Parker brewery. 2021.
Host 2 31:07
You won't be ready for April 4, which was my anniversary the first time I brewed Oh, right. We talked years and years ago. Yeah. But I did get the brackets for the pumps printed. Yeah, so beefy as hell. They're really beefy. They're like 50% infill three wall. So it's like 1.2 millimeter wall thickness, total polycarbonate, like the sun roll supernova. And these will still be on this planet. That's how beefy they are. So I got them all printed, and I got the, the quarter 23 quarter inch long bolts pressed into them. And so tonight, I am going to bolt a pump to it and bolt it underneath onto the countertop and actually, maybe bend the first tube tonight. We'll see. Hopefully, I've been wanting to actually, especially like for this work from home stuff. It's like man, when we asked him to work from home and have a homebrew when you're done, Lou. Yeah, it'd be nice. Man. I want a beer.
Host 1 32:15
I want to homebrew now. Yeah,
Host 2 32:17
I want to homebrew are so bad. I've wanted to homebrew ever since the July last year when I was up at your place. Yeah, we drink your brown ale. I'm like, oh, yeah, homebrew.
Host 1 32:30
That, you know what's funny, so I ran out of co2. But I never really emptied that keg. So it's still sitting over there. Probably super flat. I mean, it's in a fridge or it's cold, like,
Host 2 32:44
Well, is it still under pressure? Probably. Yeah. So I'm gonna have co2 in it. Yeah, it's just that you I doubt to push that out. I bet you that's pretty good. Still,
Host 1 32:55
I don't know. Oh, that's that's mold old the air
Host 2 33:00
Out? Couldn't be it depends on. It's not a hoppy beer.
Host 1 33:03
So no, no, this one was a super hoppy beer. I put I put a lot of hops in this. It had eight hops in the in the boil. Eight ounces of hops a lot and then round three ounces in. In dry hopping, so it's plenty hoppy.
Host 2 33:19
That's a lot for brown ale. Yeah. Yeah, it's probably not good anymore.
Host 1 33:23
Yeah. Probably tasted.
Host 2 33:25
That's your homework for this week.
Host 1 33:27
Taste. Taste that roast. Maybe I'll maybe I'll post a picture on.
Host 2 33:33
Who knows? That can be really good. It's like when we open up that last bottle of that stout we made Yeah. At Steven Gleaners, which was the old I think we talked to him a couple times on the pike we actually talked last week you employee on the on macro fab. And we went out to his lease not Lisa farm. Yeah. Why and out in the middle nowhere Texas. And we bought our last bottle of this beer reproofs even I was like I don't think was the first period booboo is the first time we ever entered a competition. Yeah. And like we did really well for not lazy we got our beer was disqualified for being the wrong style, because we used Australian hops instead of American hops.
Host 1 34:18
Right? When it was supposed to be an American beer is supposed to
Host 2 34:23
Be an American style. But so we made an Australian style. Yeah, okay. And it was a really good beer. And we had one bottle left that was at that point two years old, if not more. Yeah, I think it was two years at that point. And, man, it was like, you know, like old cheese. Gets like old cheddar gets like grit to it. That's awesome. I
Host 1 34:46
Was about to say, I don't want my beer to taste like old No,
Host 2 34:51
It didn't taste like cheese. So old cheese like a 10 year cheddar. Gets us like, rich to it.
Host 1 34:58
This cheddar cannot be fired.
Host 2 35:02
So, oh really old aged cheese gets his like, really cool texture grittiness to it. It almost was like that to me. Like it was. It was so good we really need to we really need to prove that beer again. The problem is though is I? I've been putting together this brewery rig. My new mash tun cannot hold that much. Great. So we have to you I'm gonna have to go old school man. Yeah, I'm going to keep it just for the mash in that one beer. Oh, yeah. So imagine that and then put it back into the new brew kettle. Although like, man, because I can only hold 15 pounds of grain in my new this was like 32 pounds is 3030. Yeah, 32 pounds. And I know I can hold 18 pounds in my new mass rake. Yeah. Which is 18 pounds is a lot for a beer
Host 1 35:59
It yet what 18 pounds worth of volume ends up being like, a massive amount of grain and then you get it super wet. And it's huge.
Host 2 36:10
Yeah. So like 15 pounds of grain is for let's say, if you're doing a five gallon beer is that's a really big beer. And in for a 10 gallon batch is still a pretty big beer. But then, yeah, I don't know if we ever told the story of how we brewed that beer.
Host 1 36:26
I think we probably told it like four or five times our listeners are like, Oh, this but I love that story. Yeah.
Host 2 36:36
Yeah. Well, let let the if we haven't told that story before someone let us know. And we'll tell the story. Yeah. It's, you know, okay, so we really need to brew it again sometime. But we have to start at like seven in the morning. Not Yeah, cuz we started that beer at like, after work. On a Friday. I got home at like four in the morning. Yeah, it was brutal, brutal. I was like, awesome, Steven, because like we had to like, because we needed like eight weeks to ferment, like conditioning fermenters. And like, we're like, it is eight weeks away before the competition and it's ready to brew tonight. So we went out, got all the ingredients and got the my place at like 6:30pm on a Friday and started brewing. It was like one o'clock in the morning. We are in ice. Cool down the ward. He's like, I like corner store. And so we were like Stephen tore off in his truck. When get ice. Stephen said he was the only soul alive in that whole area. Because I let that point I lived in like the bad part of Houston. Oh, this
Host 1 37:44
Guy gave me the weirdest looks because I run into the convenience store at one in the morning. And I grab like 10 bags of ice. And I'm like sweating. It's guessing what the hell is this kid?
Host 2 37:58
Yeah. And at that point, I looked at bad part Houston. Yeah, so it was really?
Host 1 38:03
Yeah, so it's like the he's doing some shady stuff here. Yeah.
Host 2 38:08
But like, he was like, 100 pounds. sighs
Host 1 38:11
I gotta keep this organ alive. Yeah.
Host 2 38:16
Good times. Yeah, we had the when when you're back in Houston, we'll have to brew it again.
Host 1 38:20
Yeah. Well, let's wait till this virus goes away. And then maybe we'll figure something out. Yeah, I was actually supposed to come in April sometime. But yeah, that's probably not happening
Host 2 38:31
Now. That's not gonna happen. Yeah. So
Host 1 38:35
Oh, well, sometime. So I got something I want to talk about real quick. That is, I don't know, like, I don't want this to be a complaint, I want this to be more of just like, I want to see if your situation is the same or if other people's are. So I deal a lot with St. Micro. And, um, you know, I wouldn't necessarily call myself a fanboy. But you know, pretty much all of our micros at work or St. I do a bunch of se stuff on my own here. So like, I'm, I'm in bed with St. Let's put it that way. I like them a lot. But their website is super freaking annoying when it and I've noticed this is kind of a trend with the people who provide services and applications for microcontrollers and things like that. Why do I always have to log in to my account with all the stuff just to download your, you know, utility program or whatnot. And why is that whole situation annoying as hell, you know? What I mean by that is like, you never remember your password. You never you never remember your password. And it's like, Well, okay, so you go to like, st link v2 or whatever and the page looks, frankly kind of crappy. And then you get down to the Bottom where it's like download this button and you click that button and then like another pop up shows up. And then it's like, please log in and then like you can't remember, and then you got to go somewhere else. And one of the reasons why I'm complaining about this is I've had to install this on my computer. But I do manufacturing with with a handful of other people, I have to install it on on operators, computers, I've had to do that a handful of times at at macro fab. When it comes down to getting like, whatever your their tools are. Why do I want to like, what does it matter? Why do I have to log in just to get your thing? And then why do I have to prove my email address? What like, if you want me to use your thing? Why do I have to go through all these extra steps? Like what are you gathering? That's the part that I'm so that a salesperson can call you. But I've never had like it. But I've never had any of that. And I've never known anyone who's had any of that. Like, I mean, maybe maybe we're not big enough, you know, fries for them to care about, but I don't know, it just seems
Host 2 41:03
Their algorithm is determined. You're not buying 100,000 chips.
Host 1 41:07
Yeah, that's that's exactly. But But what this kind of brings up is, I remember a Dave Jones video and Evie blog video from like way, way back, where he was talking about documentation. And Dave, Dave was was saying like, one of the hallmarks of a good engineer is having good documentation. And I totally, absolutely agree with that in every possible way, like, engineers can get so caught up and so locked up in their, in their, like engineer minds that they forget to be clear. And they forget to like understand that someone else has to sometimes interpret what they're, like spewing out. And this is this is less of me complaining about engineers, and more of me trying to encourage, like, try to think of your audience and try to think of the person who's ingesting the information that you're spewing out. And you know, I know nothing about St. Micro and their website and who designed it or whatnot, but like, to me, like this kind of user experience and interface and things and like trying to figure out what is it that I need to download? And where is it like, because if you try to go, I was trying to get their DFU downloader the other day. And like, they don't have it in a good spot. And I found it through Google as opposed to going to their website. And it's not called like, this is our DFU programmer. It's called like STM dash, O three dash, P O G or something, you know, it has a name, like, that's a great example of it not making sense. And a great example of like documentation not being clear. And so like, you know, once again, I want to make it clear that I'm not just trying to like crap on SD Micro, I'm just trying to say like, this is a great example where it's like, clarity makes a lot of sense. And it's been my experience that the micro controller guys are super unclear. I don't know, what is what are your thoughts on that?
Host 2 43:07
No, I totally agree. It's, I tend to be you know, from working with me for so long is I tend to be the person that's like, like, I want to get the MVP out as fast as possible. Yeah, I don't care about the documentation right away. Documentation is very important. But only like two or three reps down the road? Because it's how I view it. Is this all of your documentation, good documentation takes a long time to do correctly. Yes, I maintain the knowledge base at Mac fab. So I like documenting stuff is what I do. But when you do a MVP, and you're changing stuff so rapidly, it's one of those, you got to think is it worth that wait to release this? So documentation is done? Or do you just release it? Because you're gonna change it in two weeks or less? You at least that's how I view it. So like, I'm like, Damn torpedoes. Without the documentation diversity, then once we get a somewhat mature kind of thing going, then yeah, document the hell out of it. Because then it's not changing all the time.
Host 1 44:16
Yeah. I mean, I can see that. And I think that that works well. In when, when your your project size is is small to medium and your engineering count is small. Yeah, I think in an ideal situation, documentation is walking hand you know, step and step with, with design like totally either done by the same person or they're done very closely with whoever is actually making
Host 2 44:47
It in an ideal world. ideal world here. documentation should drive the design.
Host 1 44:54
Yeah, well, right. Yeah. Well, I mean, they one one is in front of the other, but only by a slight The amount?
Host 2 45:01
Yes, in the ideal world, I've never, I've never seen that actually work. Like I've never seen that actually implemented ever,
Host 1 45:10
You know? So I worked at a place that we were ISO controlled. So it's sort of the opposite of what you were saying where it's like, oh, well, you kind of like, forget about the documentation until rev two or three at this place, you couldn't rev until you had the documentation. So it was the exact opposite where the documentation controlled everything. And maybe that was a little bit overly hyper controlled. I mean, don't get me wrong, ISO 9001. I freaking love it. And it's great. But if you're trying to be quick and nimble, it's a roadblock
Host 2 45:47
In so there's, there's a good medium somewhere in between. And I just viewed as that's my job is trying to be the quick and nimble right on the team. And then I have another engineer who's the opposite way. He's the ISO 9000, kind of engineer our documentation all the time. And so I try to fight the other way. Because you reach the middle point, which is where the good medium is that yeah, yeah. So that's how I view it is, I gotta fight this one way, because this other person is fighting the other way. And then you because that's what's an engineer's job to compromise. Educated compromises, right? Yes. And so you have to have the other viewpoint when, when your other teammates that way, and Troy, if you're listening, that is you, by the way, not complaining about it. That's just how how it works. So,
Host 1 46:42
You know, I fit way into the documentation side of things like I, I would rather documentation be flawless, then send something out, that's not been perfect. You know. In fact, I would argue to say that I actually enjoy the documentation side of engineering more than the design side, like I like making all of those documents and, like, I like putting together the packages that tell people how to do things, and how to do it well. And it just the thing is, I've worked on the other side of manufacturing enough, like in other words, reading other people's documentation, I've worked on that side enough to just say, like, I'm pulling my hair out, like to all engineers out there, just remember, you're not the only person who has to read your documentation documentation. In fact, you're the probably the least amount of people who are reading your documentation. So take the extra two seconds to be clear. And the biggest thing is, don't assume anything, don't assume that someone reading is inside your head. Like, engineering documentation is one of the few places in the world where you can just be like, dry and cold and tell the entire story. Like word for word, and people are okay with that. This is not creative writing here, this is like, assume I know nothing. And your documentation has to tell me everything. I think the hallmark of good documentation is that you can hand it to anyone who is slightly technical, and they can get the entire picture from what you just handed them. They don't have any questions left for you. That's sort of the big thing. And, you know, sort of back to that SD Micro thing. Like, I don't know, St. micros part numbers. So why should I have to go search for your part number to find your DFU programmer? What like your documentation should say like, this is our DFU programmer. Here you go. Like, that's a great example of just like, make it clear to me, because, yeah, sure, everyone at your company knows your part number schemes. I'm a guy on Google. I don't know your part numbers screen, you know, like, don't assume that I do. Fun stuff.
Host 2 48:54
Yeah. So we got one RFO today is this actually goes kind of in hand with what we were just talking about. Actually, I didn't know that St. Micro conversation would go that far. Yeah. Go like that. But yeah, on the reiterate on that is, is there's a happy medium between the two points, especially in product development. And that's why I go the other way, is, because I have, you know, Troy on my team is super rigorous. And that is why I why I wanted him on our team because he's so good at that stuff. But, you know, you got to get product out there so people can actually look at it and give you feedback. Yeah, it's,
Host 1 49:40
There's a happy medium between doing things, you know, to the tee and time.
Host 2 49:46
Yes. And it's finding that happy medium is actually the really hard part. Yeah. Because that's, that's what makes them breaks the product is the happy medium because if you have you rushed the gates and you have nothing and then someone gets the the plastic brick in their hand and like how do I turn it on? Well,
Host 1 50:08
You know, and like I was saying earlier about being on the other side of manufacturing, there's plenty of times I've received documentation where I read through all the documentation that the client sent, and I was seriously like, Wait, is this? Is this everything? How do I make this? Like, you haven't told me how to make this? Like, there's not all the information here. And you literally thought this was all the information you needed to give to me? Like, how much between the lines? Am I supposed to read?
Host 2 50:36
All of it? Oh, you're supposed to get your Tarek card deck. Yeah. And start. You have to do that for all
Host 1 50:44
Your customers reading reading the lines on my palm? Yes, no on their palm their palm, right. But they didn't send a picture or documentation or drawing of their palm. I just No, no,
Host 2 50:53
No, you do it over Google Drive now or Google Hangouts? Because you can't read their poems six feet away. Right. Right. Yeah. All right. One RFO. Okay. Yeah. relates to that. And it is from the Chris gamble posted this question on the is it the V amp hours subreddit?
Host 1 51:15
If you're very literal, yes, we're Divi amp our subreddit,
Host 2 51:20
English language weird. And this is wanted practical app node suggestions. And I thought this would be really cool topic to talk about real quick is, if you had a, he's looking for suggestions on what you would like to see and add notes. And I guess so Stephen, since you do a lot with the analog world, what would you like to see? And AP notes for ICS that you generally work with? Huh?
Host 1 51:51
Yeah, that's, that's good. Okay, so
Host 2 51:52
Like he has got he's got stuff like, When do parasitics matter? And your circuit? So if you have an let's say, you have an op amp? When When do you get parasitics of your your return loop actually matter for that op amp? Because most time, if you're like, in the middle of all of its specifications, probably doesn't matter. But you're running on the bleeding edge or something. Yeah, that might matter.
Host 1 52:16
So okay, I think I what I would do is take a kind of a note out of Maxim's book, because like, we've talked about this multiple time maximum has awesome, awesome, awesome data sheets with AP notes. And what MX my favorite, they do two things that are great, in my opinion, they show a simple example. That is like, here's a op amp in a inverting configuration, just like textbook stuff. And then they show an example that is almost nearly complete or fully complete, like a thing, not like a sub circuit, not like a block, not like a thing that does like, here's a multiplier or something like that, they'll show like, here's a multiplier that does this kind of sensing with this sensor. And here's the equations behind it, like they showed the full system. So I think both of those things are really good to have an app note. Now, I you know, one thing I would also kind of temper that with, on the on the other side of things away from analog, on the digital side of things, I What I'd love to see is a full digital system. It doesn't have to be super complex, but like, show me your chip with all of its peripheral components, and show me your chip connected to a processor or blah, blah, blah, like doesn't have to be super complex. But show me all of that. In one app note that gotta love when I see that.
Host 2 53:44
I like exercises, I deal with the digital realm a lot. And with programming, I like to see like almost like block diagram, programming is a good way like, especially if you have a initialization sequence for an IC, like it's an I squared C chip, and you have to send it these series of commands to set stuff up. Like this actually goes back towards when you were looking at that Chinese datasheet is sure it just gives you all the registers, which is fine. But you have to know what order do you need at them? And what's initialization when you first boot that chip up? Yeah, that kind of stuff. Clarity and stuff. Yeah, what kind of configuration Do you have? And so having a, a good, like, what's called pseudo code level documentation on the on the initialization code, so that if you took this thing to any microcontroller and plugged it in, and then wrote the equivalent of that pseudo code in whatever language Python it could be like micro Python, it could be C could be C++, whatever, or assembly. You would get something working. That's what I like to see. And in my app notes
Host 1 55:00
Yeah, I absolutely agree on that. You know, another thing that I think is, is cool is like, when you see, like take a processor datasheet, where you see a processor, but you see it in a in a, like a known good condition, one where you see like, even even the power circuitry delivered to it. What I like about that is, first of all, it shows you like, this is this is a circuit that will work if you build it, but it also gives you ideas on like, Oh, I haven't thought about, you know, this power, good ice or something like that. And here's an example that even though my intent is to look at the processor, I'm seeing these other peripherals that you guys have tested, and it's good. It gives more creative ideas in that sense. I love seeing it that way.
Host 2 55:48
Yep. You see that a lot in like USB stuff where like, you'll see oh, they did they're there ESD protection this way, or they're filtering the shield this way. Why is that? And usually their app notes will explain why they did that way. Or you can email them. Usually they have a field application engineer that can tell you why they designed it that way.
Host 1 56:10
Yeah, usually, you know, actually, I think a good example is what's it called? FTDI. The FT 230 X, you know, there's multiple modes that it has in terms of like power over USB, or like self power or something like that. They have an example of each one of those. And, and that's kind of nice to just see like real quick, like, oh, that's what they're getting at. You can always explain it in a paragraph. But a paragraph in an image means like you've complete the
Host 2 56:37
World. Yeah. Yep. Cool. We should get Chris Gaiman back on. It's been a while since he's been on it
Host 1 56:43
Has been it's been probably over a year, hasn't it? Yeah, I
Host 2 56:46
Think so. Ya know, for sure. I'll hit him up on on Slack. Cool. So on Twitter. So that was the Mac fab engineering podcast where your host Parker Dolan and Steven Gregg. Later everyone take it easy.
Transcribed by https://otter.ai