Hi another video in the USB Power Supply series I rushed this uh prototype together for the electronics uh show hoping to get it uh, up and running and I've built the thing and uh, I'm about to power it up. so I thought we'd uh, go through some power up checks and uh, see what we get here and W I Noticed my first fail. Of course Murphy's going to get me every freaking time on this thing. my um LCD is supposed to mount on the top like this and of course it does.

Um, but I'm not going to use a connector. it's going to be the final Uh one will be solded directly on to the pin header there and it sits on there just fine and dandy. But and it's like it just clears, you know, or sits flush with the DC to DC converter there. But the whole idea is that I won't be using that socket connector and this LCD is actually going to sit a little bit lower.

and unfortunately the damn thing hits that DC to DC converter. W and I Checked that on the Cad drawing and sure enough, um, my outlines were slightly overhanging. there one was on the mechanical layer which I had turned off when I was, uh, mostly, uh, laying out the board and I just I just didn't notice I completely missed it. So yeah, fail eh.

trap for young players. There you go. That's what happens when you're rush and you don't check things properly or you don't do your proper 3D modeling and then your 3D design rule checking after for a prototype like this? No need to do that. but there you go.

I came a Guta that's the first. uh Murphy's mistake. The second, of course is you'll notice that there's no footprint. Uh, there's no chip down on that footprint down in there that is, uh, supposed to be the Um ft uh 230 XS uh USB uh USB to uart converter chip.

but unfortunately, it's not available anywhere even directly from their website until like October or something. Oh man, I swear I that they had stock when I designed this thing into the circuit and uh, no, I've come a Guta so I couldn't get that. um I couldn't find stock of it anywhere really. So yeah, I'm screwed.

Anyway, it's not a showstopper, but it's just means I won't be able to do the um serial stuff or I can, uh, use bridge in another chip if I have to to get it um, up and running? No big uh deal really. but yeah anyway. I'm sure there's bound to be more errors, so let's get to it. and the first thing I'm going to do is pair it from my uh bench power supply here.

set the current limit uh yeah, 130 milliamps that'll do uh to 5 Vols and uh, we'll um, hook this thing up and uh, see how much uh current it draws and uh, that will protect the circuit if uh, something goes horribly wrong. So let's do it and you'll notice of course I haven't hooked up a battery yet I just want to, uh, make sure that uh, it all powers up fine and dandy before I Uh, connect the battery up. So I'm going to just connect my power supply briefly across here and see what I read about 60 odd milliamps on the supply there. you can't see that's not on video, but 60 milliamps? That sounds about right? So yeah, I'm happy we can power this thing from a regular USB port now.

No problems and hey, Bingo the charge LED has come on. Um, I believe that is correct for no battery connected. So uh, anyway, that's good. Let's uh, measure some voltages.

First up. I'm going to check the voltage out from the DC Todc converter on the other side of the isolation and it should be around about Oh there it is. F about to say about 5.5 Vols CU There's no load effectively. Um, you know the circuit's drawing very little.

So uh, I'm pretty darn happy with that. 5.5 Vols And uh, where's our DC our, um, our 2.5 volt voltage regulator down here? Oh, wrong side of the cap should be 2 and 1/2 volts and it is. So our voltage regulator is working just fine so we don't have any uh shorts on uh, the board at all. So nothing, uh, grossly wrong there.

like loading down our power supply. Happy with that. That's always the first step is something shorted out? Nope. And next up, we want to measure the uh charge current of the battery I Want to check the battery charging circuitry which is around there.

So let's uh, measure. First of all, um, measure the Uh battery voltage Existing Battery: I've just got a stand like a a small 4.2 volt charging lithium ion sorry Lithium Polymer cell here and it's at 3.81 volt. So um, there's already some uh charge in that. so uh, when we hook up that battery and uh, plug in the power, it, uh, probably won't go into um, uh, the constant current charging mode.

it might, uh, jump, perhaps directly into the constant voltage charging mode and we'll be able to see that the voltage on the Uh battery whether or not it's uh, spot on 4.2 Vols Um, cuz this is like a plusus uh, 5% 5% or 75% chip or thereabouts. um, 4.2 or whether it's lower than that. So if it's lower than 4.2 it means it's in constant current charging mode and uh, if it's 4.2 it means it's switched over into constant voltage charging mode and then we'd expect the current to drop off. And I've done a tutorial on this.

um I'll have to link it in actually. so click here and you can see my lithium ion uh charging tutorial if you haven't seen that. So let's give this a go. I'll hook my meter up to Milliamps, it's beeping at me.

Go away. Stupid default. AC Current? Yeah. blow it out your bum.

All right now. let's uh, measure the uh current in this thing. So um, yeah, let's hook up the battery. Here we go.

We shouldn't have to disconnect, we'll just leave it. We'll just hot connect the battery on there and what do we get? Bang 20 milliamps and dropping? Beautiful. Pretty happy with that. It looks like because it's dropping.

It's not a constant current it I believe it is in, uh, constant, uh voltage charge mode. Uh, which is kind of what I expected. So I'd expect to measure the voltage on there at precisely 4.2 volts. All right? So I've left.

Uh, this meter here connected up measuring the battery current there and it's just dropped below 100 milliamps and it is dropping. So I'd expect the battery voltage if I actually measure it to be Tada 4.2 volts spot on. So it's definitely in, uh, switched over from constant current charging mode into constant voltage charging mode and uh, basically, uh, once this current gets low enough to the lower cutout uh, threshold programmed into the uh, microchip, uh charger chip there, then uh, it will. It will actually cut off and uh, that will be the end of charge and that Led should turn off.

But that could take um I don't know hours. So uh yeah, I'm not going to wait for that. but uh, that seems to be working. A treat.

All right. Next up, what I'm going to do is disconnect the Uh USB and uh, we should. It should stay off. It should stay.

uh Switched Off and uh, until we press the uh on button to switch it off. So hey, hang on hello hello, We got 16 milliamps that should be off. Let me disconnect the battery. Okay, so we're going to connect the battery and no, what? uh-oh that should definitely be off.

Let's have a look at the circuit. Now here's the charging part of the circuit and here's the battery and Q1: There should be switched off by R7 It should be held off, but it looks like it's somehow conducting. so it's switched on and it's conducting current out there. and uh, it can't be conducting back into um, the DC to DC converter and the chip itself because that uh, diode there D4 is.

it is going to be reverse bias. So really, um, so current draw is uh, coming from the rest of the circuit somehow. but there's something going on in here, which is Switched that transistor on Q1 there is on. um, almost certainly because there's nowhere else the current can flow unless there's um, something wrong with the uh charger uh chip here cuz it's supposed to have and when you remove the input power, it's supposed to have an off State leakage current of you know, like a microamp or half a micro amp or something.

something. tiny like that. So unless that Chip's faulty. Um, this transistor must be turned on.

or there could be a short on the board. um, causing something. So let's measure a few voltages. Okay, let's measure the output of D6 which is actually the uh Power Rail and uh, see if there it is I put a yeah 3.42 volts and on the other side of it, 3.7 Yep, so that transistor is definitely switched on because then it's feeding the voltage into that diode.

D6 And of course it's dropping. Uh, it's a shocky, you know. So it's dropping about3 volts or thereabouts. and uh, so that bugger of a transistor is switched on? Why? Well, I had a quick look under the microscope for some uh, shorts and stuff.

It all looked uh, fine. You know, unsold joints or anything like that. No, you know, incorrect. uh, component type.

It all just seem fine. So let's measure the Gate of that uh, transistor q1. Well, mosfit. Um, where is it? It's that one there .6 volts.

That's bizarre. that should drop to hang on. This is where I need my third hand have to hold my tongue at the right angle? push? the button should drop down to zero. It does.

There you go. So why is it at 0.6 odd volts? and oh, hang on I know I'm getting a very bad feeling. um 0.6 Vols is probably not a coincidence. it's a diode drop of a um bloody Npn transistor.

Oh, hang on, where's the circuit and and what do we have in our circuit here there? it is a 2394. we're getting a one Dio drop from there to ground and I have a horrible feeling that um I I've seen this before I reckon this um because I've done it before I've made this stupid mistake before I reckon the footprint of that 2N 3904 transistor might be back to front cuz I thought I I'll go through as the final step and check all the transistor. uh Footprints but I didn't I was rushed for time I wanted to get my board out I reckon the son of a is back to front, check the data sheet and there you go. What a mongrel! That was an easy fix though.

Absol Absolute classic mistake. I've done it many times before and I'm sure I'll continue to do it and it uh I've said it before and I'll say it again. One of the classic rules of electronic design is never assume anything. Never assume your Footprints correct, never assume something's going to work.

always double check it and I was in a hurry I didn't do it and yep I got caught out. Ah happens every time I think everyone soon or later in the industry is going to make that same mistake. You're going to get a footprint back to front and uh, here it is I've hooked it up and we're getting um, well, let's go down to microamps and uh, there you go. It's like not even you know I could get out the microcurrent of course.

but uh I you know I'm not characterizing the performance of this thing There You Go 0.15 microamps, reverse leakage from the battery and if I press the Uh button, let's go back to Milliamps there. and uh, if we press the button here, we go should switch on. There we go 15 milliamps and it should switch off when I release the button and sure enough, it does not a problem whatsoever that's working fine because it's not latching on CU There's no software in the microcontroller to uh, latch on the power, but that's working a treat. Ah beautiful man! I Now I got to flip all the other resistors as well.

What a pain and it was pretty easy. All I had to do here was uh, rotate the device around like that So normally uh, this pin would be up here of course cuz these are the two bottom pads on the footprint and that's the top pad there on the So 23. I just rotated it around and bingo. Problem fixed.

and depending on how you swap the pins, there's many ways to do it and some transistors can have different Uh pinouts depending on the Uh manufacturer and the Uh type, so just be careful. It's very easy to do depending on what you uh um, how you goof up the footprint. you uh, can either rotate it like that or you flip it upside down. You have to bend the pins down and then do it or rotate or a combination of both.

So there you go, pain in the butt. but uh, easy fix in the end. and I'm sure I'll find other things wrong with the board. It's not a big deal, this was just a rush.

um, a rush prototype board just to get something out there. I Fully expected uh to have a a second uh, revision of this board and uh, all these things as long as you go through, find them in the Prototype testing. Not a problem, you just fix it up in the next rev. There you go.

Um I Don't think I'll test this any further. CU there's not much more to test I Really need some uh software in this thing? Haven't done the software yet So I think I'll uh on home and start cut in some code so hope you like that. And if you do, please give it a big thumbs up. and if you want to discuss it, jump on over to the Eev blog.

Forum Catch you next time.