They may be known for being electrical engineers but on this episode, Parker and Stephen dig into the more mechanical aspects of their current projects
How low can the power consumption of the Cat Feeder Unreminder go? Parker and Stephen discuss leakage current on this episode of the podcast!
Stephen gives the MEP an introduction on Flex and Rigid-Flex PCB assemblies while Parker looks at an automotive Analog Devices application note.
Parker is an Electrical Engineer with backgrounds in Embedded System Design and Digital Signal Processing. He got his start in 2005 by hacking Nintendo consoles into portable gaming units. The following year he designed and produced an Atari 2600 video mod to allow the Atari to display a crisp, RF fuzz free picture on newer TVs. Over a thousand Atari video mods where produced by Parker from 2006 to 2011 and the mod is still made by other enthusiasts in the Atari community.
In 2006, Parker enrolled at The University of Texas at Austin as a Petroleum Engineer. After realizing electronics was his passion he switched majors in 2007 to Electrical and Computer Engineering. Following his previous background in making the Atari 2600 video mod, Parker decided to take more board layout classes and circuit design classes. Other areas of study include robotics, microcontroller theory and design, FPGA development with VHDL and Verilog, and image and signal processing with DSPs. In 2010, Parker won a Ti sponsored Launchpad programming and design contest that was held by the IEEE CS chapter at the University. Parker graduated with a BS in Electrical and Computer Engineering in the Spring of 2012.
In the Summer of 2012, Parker was hired on as an Electrical Engineer at Dynamic Perception to design and prototype new electronic products. Here, Parker learned about full product development cycles and honed his board layout skills. Seeing the difficulties in managing operations and FCC/CE compliance testing, Parker thought there had to be a better way for small electronic companies to get their product out in customer's hands.
Parker also runs the blog, longhornengineer.com, where he posts his personal projects, technical guides, and appnotes about board layout design and components.
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.
Special thanks to whixr over at Tymkrs for the intro and outro!
Welcome to the Mac fab engineering podcast. We're your hosts Parker, Dolman
and Steven Craig.
This is episode 242. So I didn't get to set up the streaming stuff yet. I played around with it earlier today. Didn't get it working mainly because YouTube's got some funky stuff that for the streaming to work. If we use Twitch, I could just click a button to make it work because I already had that set up, but we're going to try YouTube. So yeah, over the weekend, I had a disaster almost a disaster, I guess. Oh, my garage. My I have a four bank battery charger slash maintainer. And it it took a dump. Which wouldn't be that big of a deal. Except that took a dump and also took all four batteries with it. That's expensive. Yeah, cuz one of those batteries is an odyssey AGM battery. And those are $400 apiece.
I shouldn't laugh I'm sorry. That really sucks. Yeah,
basically, it was like $800 worth of batteries and $100. Charger. So I did contact the company and they didn't tell me about the took the batteries with it, but they are shipping me a new units. I don't know what I'm gonna do, because I already ordered a better battery monitor charger units. Because I don't trust this old brand. Because of course when they said, Oh, yeah, we're shipping a new one. I'm like, do I need shipped the old one back? And they said no. And so I had take it apart, right?
Of course. Yeah, no, we're all we're all over here. Like, what went wrong with it?
Because I didn't take it apart. And it's a four bank. And it's interesting. It's got just four PCBs in it. So each one is individual units. So I'm also wondering, like, how did it fail and take everything with, like all the batteries? Because actually, only one bank is bad. The other three banks seem to work but did draw the power on off? I don't know.
So step back and take give us a little description when you say it took a dump and it took the batteries with it. Like what do you mean by that?
So what I mean by is on on Friday, basically I looked over at it and one of the LEDs because it just basically I have all my batteries on it just to keep them all charged up. And so I just looked at it and one of the LEDs was off. So I'm like, that's weird. So I'm gonna put my hand on the machine and the machine was burning up. And then I unplugged it, and then checked all the batteries and all the batteries were like almost dead. So something bad happened inside the unit, and then decided to basically pull all the power out of the car batteries and discharge it into a heat heater inside of it. Right. So I guess I'm kind of lucky that one didn't catch on fire. And so who knows actually how long it's been dead for it could have been dead for weeks and just slowly pulling power off off the units off the batteries. So I'm like, okay, something must have catastrophically gone wrong in this thing, right? So I opened it up. And this is the bank that's dem shown Steven. Nothing looks bad on it. Like all the parts are, are good. It just doesn't output any voltage.
It has one of those yellow tape transformers on it because it's clearly some kind of switch or flyback thing.
Yeah, it's a it's a it's a PWM controller. Sobia switcher powers. Yeah,
there's a there's a flyback and there's clearly a like an opto one of those four leg opto isolator guys on it. Yeah, right
on there. Yep. And so, the first thing I looked at was oh, it's kind of glass fuse on it. Is that good? It's so good. So the fuse didn't save anything. So something on this board failed that caused all the current from all the batteries that drain into it. And I know none of the components look bad everything the components themselves look good. Now the construction of this board is like it's terrible. It's first of all that appliance style PCB. We're so only has a one sided PCB. Yeah. For you know, super cheapness. Yeah, that's pinching every penny. Yep. Every single penny out of it. And it's got flux all over the board. Basically, whenever they did the through hole, they just didn't clean anything after doing old through hole, like putting all the wires on it and
yeah, but it's probably no clean flux, right? Yeah, it just looks nasty though. Like yeah, but they don't expect anyone to open it up other than Parker's
true. So I couldn't find anything really wrong with it. Now I'm getting a whole new one but I'm probably not going to use it. I'm probably just going to like celery salad or something? Well, okay,
actually, this has got me thinking now, it has four of these units on there. Yeah. And it looks like they're isolated output units. How would it dump all four batteries? I don't know. That doesn't make any sense.
Maybe they maybe actually more failed and failed through like the 120 volt side.
Yeah, yeah, maybe? I don't know, that doesn't make any sense. Well, and maybe they're, I mean, I wouldn't make sense that they're actually isolated that if it took all the batteries out,
yeah, I don't know. But I bought a newer or a different, not newer, I bought a new charger. And it is it's got some special modes on this charger that it can recover bad batteries. And it seems to be recovering them fine. What we'll know in a couple more days. But it's it's charged up like my my expensive battery it got it's up to like 75% now, and it seems to be holding charge and the charger doesn't complain about it. So it seems to be fine. We'll see. We'll know once it's fully charged. And I put a key in the wagon and it doesn't crank over. Right, because it's got like an internal short or something. But actually, this new charger will be able to detect that which is kind of cool. This seems to be kind of like a dumb in quotes charger, compared in comparison. But I did look at some of the parts that are on it. And so for the switcher part, which is like the upper half of this board, which is like the 120 volt side. It uses a SF 1565 SG, which is a current mode PWM controller. And you're right, it's connected up to Whoo, jump the board. It's connected up to the pipe, the yellow tape transformer, the Opto. And it uses a FQP F 12 and 60 C FET. And channel FET for controlling the switcher. And then I really wanted to know was like How was it managing the batteries, right? Like the use of dedicated IC or whatever, it actually has a Holtec MCU, which I've never even heard of before. And that is an HT four, six r 004. And so I was looking at it. And I'm like, looking at the specifications and like why would you choose this? Well, the one of the first lines in the datasheet is cost effective. And so I looked it up that this microcontrollers, like 22 cents in quantity. So that would be why so it's probably running some simple battery charging routine that this company probably didn't develop. It's probably whoever made the module board that plugs into it. So yeah, it's kind of a bummer. It was kind of like I was kind of angry all weekend at at this battery charger.
Are the batteries just there's no chance of saving them.
Well no, the new charger i got actually has special modes to kind of like help repair batteries as a fix it like diesel cating them and it seems to be doing a good job. It's like it's actually charging up these dead batteries. Okay,
so you didn't you're not actually out 800 bucks,
not yet. Not yet. Well, not usually, like AGM style batteries are usually really good slash okay with being destroyed like normal, flooded lead acid batteries do not like being fully discharged. Like it suffocates them. So like sulfur will actually accumulate on the lead plates. So for being part of the in the acid in there, so but AGM batteries and serves different construction and all that good stuff, they, they tend to be able to drop a lot lower voltage before you start hurting them. And it was right on the edge there. But it seems to be recovering nicely. We'll know basically, once it's fully charged, and then I put a load on it and a CV can hold its voltage at that load. So, hopefully next week, I say it's okay.
Everything's good. Yeah. Do you have a good weekend? Stephen?
Ah, we can Well, no, actually, I had a terrible weekend. Just even
my, my wife got the itch to buy a new dishwasher. And so we purchased well, I shouldn't say got the edge. We've been planning on getting a dishwasher for a long while and and so we we ended up pulling the trigger and I'm like, I can install a dishwasher. It's It's whatever you know, so a job that should be 30 minutes turns into like eight hours of work. Oh, no kitchen. It was It was awful. Yeah. And whoever whoever built my kitchen, they didn't. They didn't measure from the wood floor in my kitchen to the countertop they measured from the sub floor which is like an extra three Three quarters of an inch beneath the wood. So like I get everything hooked up and I'm ready to slide the dishwasher. It doesn't fit. No. Doesn't fit because it wasn't measured. Right. So
I guess it wasn't tall enough.
Yeah, it wasn't, it wasn't tall enough if you include the wood floor. Oh, no. And that didn't matter with my old dishwasher because it had a lip on it. That worked fine for that. So regardless, it was just like, it was like eight hours and four letter words. Trying to get this out to fix that.
The dishwasher is now three quarters of an inch shorter.
The dish the dishwasher is is in there. I have to take it out. And I have I had to force it in. Let's just put it that way. Angrily forcing you
ever want to hear that about appliances? No, no, no. It's really funny. Like the last week has been an appliance talk in our Slack channel.
Yeah, yeah. has been. Our Slack channel is really like active now. Which is awesome. I love seeing everyone's working from home. Yeah. Oh, well, for sure. So I don't know, like I got the dishwasher. And it's working. I got everything hooked up. And you know, of course, like, my house is 20 something years old. I just had Romex hanging out the wall. And that's how it hooked up to it. But new dishwashers don't have that. So I had to install an electrical box, and just basically route the Romex to an outlet and then plug the new dishwasher. And it was just like, thing after thing after thing of that. You know, I wasn't even thinking about talking about that. But of course you're like, how's your weekend? Well, it sucked.
So I have I have a partially working dishwasher right now. Like the door has trouble opening because it interferes with the the kitchen counter. So yeah, I what I have to do is I have to basically chunk out a portion of my wood floor that won't be visible. It's just they put the wood floor into the cavity. That's weird, because usually they don't do that. No, usually they don't. So it seems like an afterthought. But yeah,
I actually installed my parents his new dishwasher couple of weekends ago down in Galveston, Texas. And yeah, it was the same thing. You pull the old one out and get the lifted up out, you know, a half an inch to get over the floor, right and then slide it out. And then when you put a new one back in, it goes in drops down.
Exactly. And that's how that's how it should be. So right now, if you looked at like a side view of my dishwasher, it's a little bit cockeyed, and it's a little bit unlevel. Like it's leaning backwards into the cabinet. Because because it's the front lip is sitting on top of an extra half an inch or three quarters of an inch of wood floor. Which that's not how it supposed to be. Well,
you can fix that with us all.
Yeah, I'm gonna have to do something like like that. And it's only like, totally like a two foot section of wood that I'm gonna have to just, you know, to up out of the floor.
You know, I actually got a good idea. I don't know if you've thought about how you're going to do it yet. But I would take a circular
angrily that's how I'm going to do
a truly Yeah, no, I would take a circular assault and only make it like how thick your wood is. And then that would be in the US that is as IV score line, like on a PCV.
Yeah, that that won't work because I won't be able to get a circular saw across the whole face of it. Like, there's no way I'd be able to do that. It's almost more like take a Dremel and Dremel a line across it. angle grinder. I would well the worst part would be like if you slipped on an angle grinder then you'd get this lovely circular scratch across the rest of the wood floor in your kitchen.
It was covered up with another piece of plywood. Plywood Yeah, that's what my no use the plywood as your guide. Yeah, yeah. And just still like
I'm kind of tempted to just call somebody and just be like you do it like I was gonna do this. I but I mean, like it works. We've been running dishes since Saturday, you know it functions. It's just the door interferes a little bit and it's just like, on top of that, like, of course, of course you go to Home Depot, and they're like universal dishwasher. You know, connection kit. You bring it home. Well mine's not universal, like mine is the one that is not universal. Like you lied to me Home Depot. So I had to go and like cobbled together random parts to get because I had to go from the back of a dishwasher has garden hose connection. It has a male garden hose. So you need a female garden hose connection that goes down to a compression fitting to which copper pipe can go into. Well most dishwashers use three eighths copper pipe so you can find tons of dishwasher stuff and three eighths Well whoever installed my crap did half inch which is fine. I don't care, but that's a lot harder to find that so I had to do like adapters and all this other crap to make it work. Yeah,
so your dishwasher can put some water volume out.
Yeah, yeah, it sure as hell can. Right. And like so yeah, it was it was one of those things where it's just like, I went, I went to go install it. It's not three eighths, it's half inch, go buy stuff for it. Oh, it's Romex and, and there's no way to just install direct Romex. And I didn't want to do something get out like cut the power cable it just like, you know, your old old dishwasher was connected, hardwired. It was hardwired? Yeah. Hardwired straight into the woods. Yeah, yeah, exactly. It was hard. Well, there was a little like, panel that you'd take off, and then you'd wire nut it in. But new ones have
just plug in with Nema 15. Right, right, exactly.
So I had to take that Romex route it somewhere else under the kitchen sink and put in a small little J box. And so you know, ever every little thing ends up being like a chore. And man, I woke up Sunday, and the whole left side of my body just ate because I spent half the day with like one arm behind the disposal trying to work, you know? Yeah, that was
awful. Oh, yeah. Cool.
So yeah, that was that was my weekend, in case anyone cares? And I bet you a lot of people are listening to like, Yeah, been there. And then you take a drink of your beer, right? Yes.
So So okay, in more exciting news. exciting for me, at least a few weeks ago, I talked about getting custom magnetics manufactured. And just before recording this podcast, I went to the mail, and lo and behold, there's a big box of transformers sitting there. So I got all my Transformers in and they look nice, I'm happy with the with the way they look, the there's a few things that I didn't mention last time, I was talking about it that is worth mentioning now. Like, actually, the manufacturer of these transformers put an American flag on the on the Transformers like Made in America. But if you read like a little bit further, all the cores of these were wound in Mexico. And and they get he gets them sent to I don't know, Illinois or wherever he is. And then he he puts them the laminations in and epoxies or pots them.
What's that term Made in USA from foreign materials?
Right, right. Yeah, we put it in a box and shipped it to you in America. Yes. So that's just something to keep in mind. If you are like really, really gung ho about getting 100% Made in America, you got to, you got to make sure you, you know that they're winding and assembling and doing everything in America. I don't mind these being done in Mexico. But it kind of goes along with that talk about just like, if you want to know everything that's going on with your transformers, start a conversation with the with the manufacturer, and make sure you get everything done, right. So actually, after I got the Transformers on order, I went to Fusion 360. And I started modeling the Transformers because I want to, I wanted to make sure that they fit my chassis. And I want to make sure I have all my mounting holes, right. And what's interesting is I used the drawing that they gave me for these transformers. And I started running into a bunch of discrepancies as soon as I started modeling it because they said what cores they're using cores are standardized. So if you have like an EI 100 core, you can go out and find like the exact dimensions of Ei 100 cores, so I started doing that. And then I started using their anvil drawings. And I ended up with like, quite different dimensions and what they're drawn to said, So I emailed my transformer guy, and I was like, Hey, I modeled this, and things are seeming weird. My initial intention, or my initial thought is that I did something wrong, but check me on this. And he's like, oh, yeah, no, all those dimensions are wrong. Here's, here's the new ones. I'm like, Come on, man, you sent me a drawing. And it's like, and on top of that, it's kind of funny, because like, instead of updating the drawing and sending me like one with proper dimensions, I just got a bunch of pictures with like, a ruler on transformers. It's like, Yeah, you should use this as your mounting holes. So I'm not trying to say anything bad about this transformer manufacturer, because so far, it's been great, but it's also been like, I wish that was better. That aspect of it, but the price I paid for it is like I didn't pay the kind of prices you would get for them to go and make new drawings for, for one custom off run kind of thing. So keep that in mind. There is one thing that happened that is potentially a blessing in disguise for me. So I had originally asked for some specific taps on this transformer that ended up being very similar to a current product offering that this this company offered. So we worked out a deal where it's just like a little make some slight changes to it. And I get pricing That's basically that model. And it's just slight changes to things. Well, I had asked for a 12 volt tap on it. And I haven't checked anything yet. So I'm not I'm not saying anything's wrong, but on the side of the transformer, it says that I got a 12.6 fold tap,
they just reuse the sticker.
Maybe you never know it. Well, no, I can tell you the Not really because they don't have 12.6 on their other transformers. That's the one thing that was unique because I asked for a 12 volt tap. So the thing is, this company typically does guitar amp transformers. And then the guitar amp world, you never really asked for a 12 volt tap, you always ask for 12.6. In fact, if you say 12 volt most of the time you mean 12.6 is I see it being like a confusing thing for them. Is that
like in the car rolled a 12 volt 12 volt batteries and your system on your car is actually like 13.6 volts. Yeah, the alternator?
I think I think it actually stems from the fact that it's just takes longer to say 12.6 When you basically always mean 12.6 volts when you say 12, but I added volts. And I went back and was like, oh, did I screw it up. And I went back to my original request that I sent in. And no, I wrote 12 on there. And the thing is, I'm not, I'm not going to fight anyone on it just yet. Because it might actually be 12. And this might be a typo written on the side, I haven't measured it. At the same time, I can actually stand in my circuit to have a little bit more voltage, I'm kind of on the cusp of being on the low end, with my current values in my in my circuit. In fact, I tested, I tested my circuit at like low end of mains voltage. And it was it was below what I was comfortable with. So I was going to adjust some values, some resistor values in my in my circuit to compensate. But having to input six might actually just work out fine. And I don't have to change my resistor values. So like I said, it might just be a blessing here. The one thing though, is in that situation, going forward, I would need to be explicit. It's like no, I really do mean 12.6. Now, even though I said 12 Previously, so if I if I ever do another order with this guy, I'm gonna have to be super explicit about that. But I don't think that's a problem right now. So after this podcast, I am going to go and plug them in and find out all the mysteries that are inside this transformer.
See if I can let the smoke out.
Transformers that big. You don't really want to let the smoke out.
I've done it before. It's not fun. Very first transformer I bought that was like a big boy transformer like this very first one, it was over 200 bucks. And I was in. I was a freshman in college. And I wired it up wrong. And just that was it. It was like 200 bucks primary on the secondary. I shorted somehow I still don't even know. But I shorted the heater tap which was 10 amps. It was it was 6.3 volts 10 amps. And yeah, the smoke came out real fast on that. You know, what was what was really nice is the transformer company I ordered from it was a custom transformer. And they were super nice to me. And they're like, Hey, we know you bought one of these because you're doing a project and they're like, we'll sell you another one at half off. And I was like, Oh, you guys kick ass. So
during that company's name,
oh, Netflix, oh, Netflix, oh, Netflix. Yeah, they they do like they were they were like cream of the crop for the particular app I was making back then. And they were also extremely expensive. Like, at that time, I couldn't find a more expensive transformer.
So it was what do you use 12.6 for what's that point six volts for
most, most small vacuum tubes have do dual tripods in them, and you can run them at 6.3 volts, or you can run and that that's the heater or you can run them in series and put 12.6 on it. And you get lower, you double the voltage you have the current demand on it and current is what couples into other parts of the circuitry. So you can you can lower your entire current demand and get lower noise coupling. However, I'm not doing using it for that at all. So it's just kind of a an old crossover from old to bland 6.3 and 12.6.
Yeah, that's that's a weird 6.3 volts is kind of a weird number two is yeah, it's probably someone the first time someone ever made one like Oh yeah, 6.3 volts works great.
Yeah, and that became the standard actually. So the the first number in most standard tubes. You By now, the first number is its voltage 12x. Seven is 12.6 volt, tube, six V six is a 6.3 volt tube, you know, so you can easily find it out that way.
Learn something new every day or a day. So the cat feeder and reminder. I actually ordered parts for it. Now finally, it's taking me taking away
your cat, it's gotta be getting fat.
A little bit actually. I've been keeping her outside more often so that the heats kind of sheds weight off. It actually hasn't been that hot recently. So she also likes being outside. So you know, the people were saying your cat should be inside and like my cat likes to be outside all the time, is that when it gets fed, then it wants to be inside. So yeah, I ordered the a breakout board for the Hm 10 941 which is the solar harvesting microcontroller. I found that a breakout board a person makes on tindy not Tinder and that'd be really weird to find. Find a breakout board on Tinder.
It's a match made in heaven for Parker right there.
Yep. You know, I'm not gonna give him go down that road. So I picked some solar supercaps solar supercaps there, the SCC s 30, B 10, six, P B R or P RB not PBR the beer. And these are 10. Fer adds at 2.7 volt DC li crap 10. And us two of these in parallel, not parallel series. So you get five fair ads, right? Yeah, yeah. But then you get, what? 5.4 volts of voltage. And then, so I decided, Okay, with this, is this enough power? Do I have enough power to make my device work? So the first thing I did was I went okay, what is my current draw of my timer, which is the LTC 2956. And it draws eight. No 9.8 micro amps total. And then I was like, Okay, what about the LEDs? So the LEDs I found are the system I'm going to do draws about 40 micro amps, an LED. So what I'm going to do is I'm going to have one LED, you know, I haven't like a status LED that blinks the bass, he's like, okay, the device is working, right. And then, when it's time to feed the cat, the cat feed indicator turns on and the other one doesn't blink anymore. Kind of the same
way. These are LEDs that are illuminated at 40. micro amps. Yeah, you post them. Yeah, but when you post them, they got to be way higher than 40 microns. Yeah. But
the average is 40. Oh, okay. Okay.
I'm sorry. I thought you were saying it was 40 micro amps when they're emitting light. I was like, I need to know what that led. No, no, that's the that's the average. Got it. Got it. Okay.
Okay. So my average power draw for the cat feed. Cat feed or on reminder is 40.8 micro amps pretty low, it's probably gonna be a little higher than that. But that's probably a good start to start. So that basically rolls out to be five fer adds times. So 4.5 volts is what the A 10 Nine for one will charge up to the supercaps and then has a lower threshold a 3.6 volts, basically under 3.6 volts, it turns off the outputs because it can't run its LBOs anymore. So that's point nine volts of swing that you get to play with. So five times that divided by how much current you draw. And so I am doing 60.8 micro amps, basically like one LED and one blinking LED at like half a half duty cycle. And then the timer and so that turns out to be seven, not seven 774,000 seconds, which is about 20 and a half hours, which is plenty of time to work because you know half the day is light and half these dark.
So so if it was like an overcast day, you'd still probably be okay, let's look at it.
Yeah, and actually I found the solar panel I'm going to try is the am 1816 ca which has a 84 Micro ramp peak at peak power 84 micro amps and that actually is peak power. tested at 200 Lux, which is like, dark indoors. So anything like if you turn the light on, it's going to be like, powering the crap out of that solar panel. So it's a solar panel designed for low light. So I'm basically gonna cobbled together something on my, my bench and then monitor it and seeing how well it works. I might actually be able to get away with a smaller solar panel.
You should you should try. You just build a dark box and see if it runs for 20 and a half hours. See if you can Oh,
yeah. And just having the LED poking out the box. So you can watch it?
Yeah, yeah, whatever we're up. Because that would be really fun to see if how close he actually got to that number. Yeah. If you're if your current estimations are anywhere near Right,
yeah. But I'm, I'm mostly interested in the solar panel. If it's enough, I think it's plenty to be able to power this because it's, it's over double at peak power, the current output of this panel at the low end of like, indoor lighting. So that should be good. Very cool. And actually, I was thinking like, do I even need if I'm going solar? Do I even need a, a indicator? To let me know that? Hey, are the devices alive? Because now it's always powered up? Right? Because it's got solar,
right? It's still nice to know that it's doing something right. A heartbeat somewhere?
Yeah, somewhere? Yeah. Yeah, make sure your cat gets fed.
That's usually not a problem, they usually let you know when they're hurting heartbeat is a lot stronger than your little LED. Alright, cool. So in kind of in parallel with my new high voltage circuitry, I've been developing a test connector that I can use in my products. And it's sort of if you know, what tag Connect is, it's based off of tag connect technology, effectively with with a connector that can clip into a PCB and use pogo pins to basically have a free in quotes footprint on your PCB that you can connect into. And it doesn't, you don't have to populate anything on your main board, you're talking about a populate with a connector. Yeah, you don't have to populate a connector on there. Basically, you just have exposed pads on a board and a connector that interfaces with those exposed pads with pogo pins. And it somehow has a captivation mechanism, some kind of mechanical mechanism that hangs onto the board via some kind of like holes or something like that. And that's what I've been developing. And I finally got the boards in last week that I made a while ago, I actually ordered these boards like, a long time ago, but I've been in a bunch of goofy talks with the PCB manufacturer about this. In fact, let's let's let me go a little bit into that real quick. If you're going to make mechanical items out of Fr four and order them from PCB vendors, be prepared to just have discussions about the fact that like you're fine that this is purely a mechanical thing, because it's frankly, it's confusing for for PCB manufacturers. I cannot tell you how many times I had to re upload my files to the My PCB manufacturer because they just couldn't understand that I wanted just fr for like just fr for cutting to a change that with holes in it with holes in it. Yeah. So originally I uploaded my files. I double checked everything in my Gerber viewer. Everything was fine. Well, I got I got a message back being like there's no gerber files in in this thing. I was like, I looked at it and my Gerber viewer, yes, there's gerber files in there. Well, they were complaining because they didn't get a copper layer. And every PCB has a copper layer, right? Well, I didn't have any copper on my board. I just wanted fr four. So it's like fine. Okay, so what I did was I put a copper fill on the top board. Well, that wasn't good enough, because I don't remember the reason but like, it just had copper on it. Now, but like for some reason, they complained about it. And I think actually no, I remember what it was. It didn't have a hole in the board. And every PCB, for some reason has a hole in the board. So I was like fine, okay, so I put a via just randomly in the board. And I was like, great. Now I've satisfied the requirement. Now there's a hole in the board. Well, they came back and they were like, Excuse me, sir. You had originally put one or as your your layer count. And we now see that your, your board has two layers, because I had a visa in it. So it had copper on the backside of the board. I'm just like, oh my god, are you kidding me? So I was like, yes, change my layer count to two. And eventually they were like, Okay, now this is acceptable. And, but they would it was funny because they wouldn't manufacture one of the other boards because they said it was too small, even though like I put a bounding box and I made it bigger than their minimums, they just clearly didn't want to make it. So I was like, Screw it, I'll just machine my own piece out of gara light or something like that back at home. But regardless, just like I said, be prepared to jump through some hoops because they want to make PCBs, and they want to make things that look like PCBs, if you give them something that doesn't look like a PCB, and it gets confusing for them. It might be a headache. And for me, it was like, it was like a week and a half of going back and forward about like, a yes, I've re uploaded boards. And yes, I've done this and done that. And I just eventually had to like cheat the system. You know, regardless, I've got I've got a little board here. And I'm showing this to Parker here. But we'll we'll post some pictures of it. Basically, I have a board with a bunch of pogo pins going through it. And the pogo pins are made by mill Max, and get ready because here comes the part number it's Oh 955 dash zero Dash 15 Dash 20 Dash 71 Dash 14 Dash 11 Dash zero in case you wanted to know that. And I'm sure most of you guys don't care. But mill Max makes a whole bunch of really fantastic pogo pins, and they make them in like every flavor you could possibly want. So if if you ever need to do like a testbed or something like that, check out mill Max, most of them are available at both DigiKey and Mauser. These particular ones are pogo pins on one end. And then on the other end, they're solder cups. So
these also have shoulders. Yeah, yeah. So that's one thing is, don't pick pogo pins that don't have shoulders on them. Right. So what the shoulder enables you to do is to set the height of the pogo pin with the surface level of your PCB. Yeah, if you don't have the shoulders, you can adjust the height to whatever, which is fine if you only have a couple. But if you have like what Stephens holding up, which is like 1616 of them, it would be it's heartbreaking. Nightmare. I said heartbreaking to set that your technician will hate you.
Oh yeah, yeah, and it won't be right ever. So yeah, the with these ones, basically, I have a little board that I pass 16 of these pogo pins through and they that shoulder sits on the pads, and then I can just solder them in place. And since they have a solder cup on the backside, I can just solder a wire directly to them. And then that's going to go off to my little test box with that has all the circuitry from my high voltage, low current testing in it. So the mechanical aspect of this is that I have two other PCBs that fit at a 90 degree to my little test connector board here. And it actually has a little nub on one of the boards that interfaces with a hole on my test footprint. So you cannot insert it backwards and blow everything up and waste 100 bucks worth of delicious parts in your testbed. So I've tested this out basically beat tested things. And so far, this is actually working pretty good. I'm happy with how this all turned out. There's, there's a handful of changes I'm going to make because I'm going to order another version of this. Basically, I want to refine this until like, I'm happy and it'll just work across the board and just make a kind of a standard and what I'm doing now. But so in the second revision, well first of all the footprint that I made, that this this connector interfaces to, I made the pads that each Pogo pin touches the diameter of the pogo pin just to test that. And so if my connector is is angled or off at all, it doesn't make contact. So I think I need to make the landing pads two to three times the size of the actual diameter of the pogo pin. And that means any angular offset will just it'll still work. It'll still work. Yep.
Or you can have you can put what like how tag Kinect has it is they have three nubs so the so the part or the the connector can't rotate.
That's actually I wrote that down in my notes I'm going to on the top side of the footprint I have one one hole in the PCB that a nub goes through and on the bottom side, I'm going to put two so you still can't put it in backwards and then it'll prevent rotation. So I think I think all of those put together will just make like a row bus system that connects every time I also I set the height of all of my pogo pins such that at nominal, when they're pressed into the board, they're half the travel of the pole correct, I think I'm going to put it a little bit further because I want it to press even further down in, I think these things have a travel of like 40 or 50 thousandths of an inch, I think I'm gonna make it such that you have to do like 60 70% of that whole travel, because usually I spent half of it. Yeah, and that's what I did, it just doesn't feel perfect, I think a little bit more would be better. I want to I want to, and that'll take up any slop of if there is any difference in height between them. Because right now I think 25 thousandths of an inch 25 thousandths of an inch is is fine, but I think I just feel a little bit better if it was a little further. And then I have to make, I haven't made the pole design and the pole is the fingers that grip into the PCB that actually hold it in place. Like if you seen the side of a tag connection connector, they have these fingers that fit in and kind of grip the board. And I designed them up, but I haven't really figured out exactly how I want to do it. That's kind of a hard problem to get past. That's the more mechanical side. And so like I'm not good at abortion, but I but I have I have holes in the side of my test board that that the fingers are going to grip into. And I've got the poles designed, the only thing that I don't have right now is how would their spring loaded, because I want them such that if you let go of the poles, once you have the PCB in place, it'll actually like swing in and grip into the board. So
I have a dumb, dumb question, please. Why not just use to two by five tag connects?
Oh, high voltage is
okay. Is you need more isolation?
I need I need plenty of isolation. Yeah,
that totally makes sense now, because I'm like, why are we designing this thing that already kind of exists? But now that totally makes sense. Because now you need high voltage and isolation.
I need Yeah, high voltage isolation. There are 16 pins on this. Only one of them is ground. Four of them are low voltage, all the rest are upwards of 500 volts. So totally makes sense. Yeah. Yep. Yeah, why go through all this? Like headache of making a special thing? Right?
Well, sometimes it's fun to just try to design something like that. But I was just like, Okay, what's what we're getting at with this?
Right? Right. The you know, the other thing also is, I'm going to add a little bit of extra board on this, because this board, I think it's like an inch long my hole connector. And it's going to have 16 wires coming off of it. They're going to be pretty small gauge like 22 gauge. But I think I'm going to add some PCB material such that I can bundle all the wires together and then make a strain relief. And basically, add on I might use adhesive, I might do zip tie, I'm not entirely sure. But make something that holds the bundle of all
16 I wouldn't use as a big tube of heat shrink over it. Yeah. And
then and then grip on to the board.
Because I would use heat shrink that's got like isolation rating of like 600 volts or
higher. Well, all the wires are going to have to have that right. Yeah. So but yeah, this is this is pretty cool for a first go round. And like I said, I beat tested all of it. So it's all functioning. It's just, it's a little flimsy at the moment. So I'm gonna do another another round of prototyping on it. But so far, I'm pretty happy with it. Okay.
Sounds good. Yeah. Is that gonna wrap up our podcast, I'm actually really looking forward to seeing how that works out, especially in the isolation side of the voltage.
You know, I checked all of my clearances, and I have more than enough creepage clearance for each one of these bins to be, you know, if you had two adjacent pins to be 505 zeroes, yes. You know, and they should not short. Well, and whenever, whenever I actually connect all 16 wires to this, each solder cup is going to have heat shrink over each one. And then okay, and I'm going to put adhesive and basically just like in glue it all together in there. You tune them. Yeah. Because I hope to make a few of these and then just be done with it for ever. Yeah,
the what's gonna be interested in seeing how you know what would work really well? Is 3d printing those clips
or see and seeing them at work. You could do that. 3d printing would get all the dimensions just right. I wouldn't have to just deal with what material I had. Yeah, you You want to 3d print me some clips?
Yeah, I can do that. Actually, if you want to throw me what you want to model?
Yeah, I actually have, I think I have a modeled up somewhere in fusion. So I'll look at it again. Because like I said, I haven't worked out the actual. So there's two issues here, or there's two methods. And I kind of don't like both of them. So one of the things is, I could have the, the pole arms that actually grip the board, I could have those so they're not spring loaded whatsoever, I could have it where you put the connector in, and then you open the poles and grips the board. And it could rely on the force of the of all 16 pogo pins to press against it. And that would probably work. Or I could have the poles be actually be spring loaded themselves. So they want to grip the board. And that seems better. To me, but I just don't, I haven't envisioned exactly how that works. Like where the springs actually fit, you know,
I would use, especially since if you can make a 3d printed material I would do. Because I'm, I'd be print, I'd be printing it out of polycarbonate. So I would use the ability of polycarbonate to flex a bit and make it a compliant hinge. Okay. And so when you would have I could draw it, it would be really hard to explain on the podcast, I'll draw it up. And just, I'll just draw up on napkin and send you a picture and then we can put that in podcast notes, but it'd basically be like, it snaps in. And then you squeeze it and then like, you know, it's it moves the pins on the bottom side and you have like a little compliant engine there be really easy to make.
Yeah, I mean, we could basically just copy Tech Connect, right? Yeah,
I mean, that's what you're doing. Well, but okay, so Tech
Connect is big. You look at the fingers. They face outward. Yeah, that's how this mind face inward.
I think inward would be fine. I'm sure I'll show you the drawing. I'm thinking of got it.
Okay. Yeah, whatever works. Yeah, I mean, if that works, that's fine with me.
I'll just like I'll draw up something, send it over to you and you model it and I'll see if I can print it or not.
The thing is, they're not going to be big. Like these are these are all pretty small things.
Should be fine. Cool. We'll see. Sounds good. So that was the Mac Feb engineering podcast. We're your host Parker Dolman
in Steven Gregg. Later one take it easy
They may be known for being electrical engineers but on this episode, Parker and Stephen dig into the more mechanical aspects of their current projects
Stephen gives the MEP an introduction on Flex and Rigid-Flex PCB assemblies while Parker looks at an automotive Analog Devices application note.
How low can the power consumption of the Cat Feeder Unreminder go? Parker and Stephen discuss leakage current on this episode of the podcast!