Hi just to follow up to the BM 786 uh, troubleshooting video I did where we came to the conclusion that it's most likely looks like, uh, the main processor here has, uh, failed because we can't program it uh from the header which is, uh, this one over here like we can't program it directly uh from the header even though the voltages around here look okay. but a few people have had, uh, some comments so I thought I'd have a look at that uh yes, I did actually uh, measure um, all the decoupling caps to make sure like none of them were shorted and uh, stuff like that. uh, the bypass caps. sorry I forget who mentioned this but somebody in the comments mentioned this resistor down here.

Now clearly, uh, this resistor and this cap here these have been hand soldered as have, uh, some other parts that we'll have a look at. Now, why have they hand soldered? uh Parts like this, Is it like a mod or something like that? No, almost. Uh, certainly. I Think it's just that the pick and place machine when they assembled this, these parts fell off because you know, Parts fall off the heads all the time and so the board goes from the pick and place machine through the Reflow oven and then it usually goes into an optical image inspection system which Compares it against a golden reference board and then it can pick up that uh, you know, components of fell, you know, fell off, they've been misplaced or they're tombstoned or whatever.

Um, so yeah, they're probably parts that have fallen off, and rather than you know they don't put it back in the pick and place machine and they're not going to scrap the board. So you know they have somebody do manual rework. And obviously someone at Prime is not very good at doing manual rework because this is pretty please poor. Now this is the LCD driver chip.

Maybe This resistor here is to do with the contrast also set in the contrast or something like that I Don't know. but yeah, that's the OCD driver so that really shouldn't have anything to do with it. But I did find something interesting in here. Look at that, it's 0.2 Ohms.

That doesn't seem to make sense. like it could be like a is part of a low pass filter for the power supply. but at 0.2 ohms, that does not sound right to me. So I got my reference board.

Let's have a look at that one that it is 9.4 K We can flip it around. Yeah, 9.4 K that's that's a good tip just to measure resistors in both directions. Just make sure you've got no Active Components in there. you still got other components and you'll notice that this one actually has the resistor in the middle whereas the faulty board has it has that resistor completely missing.

I Don't know what the issue is there. Joe Smith Who, uh, you should be familiar with his channel does multimeter destruction testing. He noticed you two over here. Like some people said, oh this, uh cap I think uh, def Pump who does repair videos as well? uh if you want.

if you like your repair videos, he does multi-part series repair videos. So check out Def Pumps. uh Channel and um, no, that is, uh, actually part of the tray so it's not like shorted out to that Veer it's supposed to be now Joe thought that this, uh, you two here had a blowhole in it. um so I don't know what U2 does there but uh, that is that looks like for all the world, not like a blowhole.

It's just flux. so it's all just flux I believe that's just flux residue. So yeah, if we clean that off I just wanted to show you that before I can I could actually clean that I'll just get the isopropyl I do have flux cleaning stuff, but oh yeah, there's also that black stuff in there as well the gunk in there. But anyway, I think you'll find that'll actually clean up nicely.

So yeah, there's no, uh, there's no blowhole in that anyway. that resistor there is completely suss. so I'm going to get that out of there even though it's got nothing to do with the microcontroller circuit. So the problem is is that we can't even identify the micro so you know I really need a better pair of tweezers I do have I lost my set I don't know where it is there we go.

Go on. Skeet that is zero. Ohms, No wonder we're measuring point is they actually so? so there's nothing wrong with it. But once again, this meter was working.

so obviously I don't Maybe they made a circuit change? Okay, so the other board I've got here has got a on the bare PCB 20 33rd week uh 2020 whereas this one's got the 12th week 21. So yeah, obviously they've made a that made them more recent change there I could open up at like a brand new stock unit or whatever, but it was obviously working so there's nothing wrong with it I So I'll just put that back I don't think that's an issue. Give that a bit of a clean too. Okay, this has actually locked up my programmer I can't cycle through I've got to actually repair it IDs the attempts to ID the chip stick it in the Via there so much easier than the PIN just the tip.

just slip the tip in I was a single shot. Capture that Boom. There we go, we got something. So ah, there we go.

That's nice isn't it? The uh National instruments actually captures outside the window. That's the zoom out feature. I've done a video on that looks like the national instruments. Does it see where it comes in handy? I'll zoomed in.

Oh, otherwise I'd have to repeat this whole process. No I don't look boom boom boom and you know you don't worry about overshoot like that. In fact, I'll show you the other ones because they're on all the time. So there you go.

That's third from the second from the top. so that one looks like data. Oh, it's an interesting level thing happening there. Like a little bus contention or something.

So obviously we've got a decay there. So the bus has been disconnected somehow and there's glitching over there. That's interesting, isn't it? Let me look at the other one. It's not a continuous clock, so they're those two pins.

anyway. I'm going back to the second one. Oh, there there we go. So we're getting two pulses, right? So there.

That's uh. all the data on the programming, uh, header. So it's all there. It's not being loaded down except that other bus.

but that's a bus. It's like a bus contention. Like it's a bus thing. It's not like the entire line is shorted or anything like that, so it doesn't seem to be a big deal.

A couple on the top. These ones here. these come out. Let's go around here.

They go around here. they go to here. Oh flippity. Oh, it's getting close to being oh okay.

they're 100 ohm resistors. Are they those two high speed signals? So that pin in that pin? So that's not an issue like you know and some people have said quote, you know a few people said I just Reflow the main chip. Yeah, I'm not going to do that now it's Larry Ah, what's that? Yeah, look for the reset pin. I Agree I adjusted my ATM switcher so you can now see these signals so there's those two main signals going in there I think I'll get out the good board just as a matter of course and see if that's the same on the good one.

Oh okay, so it only reads it once. Okay, so I have to I have to cycle through that on my program and the good thing is, um, the stand line programmer by the way, I don't have to have it hooked up the PC I've got the firmware actually programmed into it and then I can just hit the program I can just hook it up. hit the program button. Boom.

Very nice actually. There you go. So it's obviously like just continually for the other one. it's just continually cycling there that doesn't actually help us.

does it have a look? I'm not sure of the memory depth here. now we're reaching the reaching the limits and I don't have to set up complex triggers and capture and all that sort of. you know, like yeah, because then we'd just be getting into quirks of the programmer. and and the Chip And how it's programmed and you know how it's You know the ID is detected and all that sort of stuff.

So as I said, I don't have the Uh pin out for this because it's a like a sort of semi-custom uh device for Bryman, or it's at least a custom variant I believe. So for a reset pin, you'd probably be looking for like an RC power up or something like that. I Mean there was that switch thing. we saw that on the bottom of this switch.

Uh, before that it had that contact and that could be like a power on reset contact. but that's that's different because that's when you rotate the switch. This is with the switch in the off position that that's the data in and data and the other one's got to be the data out. I Just noticed something embarrassingly dumb, which nobody, absolutely nobody picked me up on everybody.

Oh, stick me up on this. Well, I don't think so. I'm sorry if somebody did. This is not the processor.

this is the multimeter chipset. so there was no point changing that. Crystal Makes absolutely no difference whatsoever. This is the main processor.

How is it the main processor? Because you've seen me. I've been probing these clocks around here. It just dawned on me this is the damn processor that the previous video for some reason I was fixating on that other chip which is the multimeter chipset. don't? Well, that's embarrassing.

So yeah. I Can't believe nobody picked me up on that. I Expected more from my audience. Anyway, that's water under the bridge.

Okay, since discovering that this is actually the processor here, Um, yeah. I've actually gone back to the data sheet and it turns out that the pin out seems to be correct because this is now a 64 pin chip. and that does match. Uh, the data sheet for not the four part number.

The one I've that my programmer reads out is a four digit part number, but the data Sheet's only three digits, so it's got an extra one tacked on the end. Um, but I can't find any information on that one. Now just as an aside and a trap for young players, one of the annoying things about this particular Uh Micro is that it's available. Not only is available in a Uh Lqfp like this, uh, low profile quad flat pack, but it's also also available in a 64 pin Qfn.

Now, the pin out for the Qfm in is slightly different to this one. So I do actually have the pin out for this one and it matches up. Um, this is the clock going in here and this is the data for the debug interface. So clock and data.

I've discovered that this one here is Uh VSS or ground and this one here or pin uh three here is actually uh, positive. So ground and positive rail here. But the Qfn pin out is like shifted one pin around so that pin 64 here is the ground and then this one's the clock. This one's the data and I've uh and and in this particular case, pin two is actually the reset pin.

So aha, we've found the re reset pin it buggers off under here. Does it go over to there? I don't know I can Buzz that out VCC goes through that zero Ohm resistor to there. So what I can do now powering it through the debug interface, we can measure the voltage. Rail there there you go.

So 3.5 volts we can look at Pin in two which I have verified is the reset pin. got to be careful in your probe here. don't assort anything out and I know it's an active low reset and there you go. We're actually getting 3.3 volts there.

Yeah, it's not been processed and not being pulled in a constant state of reset. but that doesn't mean that there's not like a fail capacitor. I'm going to actually follow that reset Trace In fact, I'll just probe make sure that is going to where I suspect it is. Yep, no I know it's that via going out there.

So I just whack a light under that. Geez. I need to turn that down a tad. There we go, that one.

So that's going out there. Aha, there's your reset cap going down the ground. Could that cap be failed? Hang on. I think I traced the wrong PIN So that's another tip.

Just be. uh, double check whether or not, um, the trace That you are tracing is, uh, the one you're tracing if that makes sense. Yeah. I goofed that.

It's this one here. I Yeah, I was way off as the reference I took those four pads as the reference, not those four pads. It's ah duh. Yep, you're probably screaming that at home, right? So this is the reset line here.

Aha, a Diode. So let's actually measure that. Yeah, Diode's okay. No workers turned around the other way.

Yeah, Okay, easy to test and rule out actives like that. Easy to. usually. um, diodes will test in circuit like that and then there's something going off over to here as well.

But once again, like our reset line isn't being held low so the process is not been actively reset, the data sheet definitely says active low. Yeah, it ain't that. Anyway, it might be chasing a red herring with the reset line, but you know you just want to make sure because if there's a cap that's pulling it low and that cap is, you know it's it's open then which is a failure mode for caps. They can fail short and open.

uh, multi-layer ceramic. uh, caps. and if it fails open, it doesn't get its power on reset. That could cause a problem.

Okay, I just discovered that when I I can switch the power off from the programmer when it goes into that programming mode. So what? I can do? So I can adjust the time base here. Make it short: 20 milliseconds, 50 milliseconds per division. Something like that.

I can single shot capture that. then Force the programmer. There we go. two reset pulses.

That's interesting. Let's compare that with a good one. One reset pulse isn't that interesting there. There is no reset line from the programmer.

Um, that's according to the pin out on the programmer so that's very interesting. but in any case, that is not. And I see, uh, like power on, pull up reset pin on the micro. This is the direct reset pin.

There you go. So yeah, there's there's something active. Hmm. there's not like a reset chip or anything on here.

I Don't think. Aha, that's interesting. The reset line is connected to there. The plot thickens according to the programmer.

Pin out that is actually VPP Okay, so VPP go is the reset line on the programmer. Okay, right. that makes sense. Um I Didn't think there was like an active reset circuit because like a lot of times you can get you know from TI and a whole bunch of other makers you can get like active reset circuits which cleanly give you a resup pulse when you power on as opposed to relying on a RC power on.

So obviously, um, yeah, it's it's getting the pulse. so that's probably why there's two pulses there is because that's the programmer just going did and it's trying to reset and then it doesn't get anything and then it resets again and it's just trying itself over and over as we've seen with the clock and data. Whereas the good unit when we hook the programmer up, it only it. it does its thing, reads the chip ID and then stops talking, right? So at this point I know that the processor is getting clock data and getting reset as well.

which is the programmers? VPP I Don't know if it actually does a VPP function and actually like pulls in VPP is uh, programming? uh Power Um, so it's You know, old school is like 12 volts, even 24 volts something like that, you know, but it pulses it to a higher voltage I Don't know if it actually does that I don't care. it's not reading out the chip data. So the the debug interface is getting everything. so the Chip's getting its uh Power and it's getting clock data and reset from the debug interface and it's not talking.

So once again, the only conclusion I can come to like I did in the first video is that that processor is bunk. Possibly I'm gonna have to find a donor unit and try and remove that and then um, solder on another one from a known donor unit that actually talks to the programmer. If I can't talk to the programmer, that's the whole fault. Yeah, obviously I'm not going to debug the multimeter in any other operational regard.

There's just no point when it doesn't go directly from the debug interface like that. and that's basically is bypassing all the power on any sort of like power on reset or anything like that. It's it's been controlled from the debug, um, interface. and it's just it's not doing it all right.

I have actually found a donor unit. Let's actually remove this. put some flux on here. All right.

let's see if I can remove this. Bingo! Didn't damage anything there. Nice. Do the same thing for the dead chip so we won't be able to do any more debugging on this.

I'm afraid I am pretty done on that. Sorry my tagano is not going to be able to view this. There we go. so that's the faulty one.

and no pads were harmed in the filming of this video. If you're wondering what flux I'm using, it's an Edson FL 911 seems to work for me. Your mileage may vary now. I've got to make sure I put the right chip in and I'm going to make sure I get pin one over here.

put that on. I might just uh Wick off some of the solder on those pins. see that Wick straight off there. That was nice.

Seems to be a couple of shorted out pins on there, but I'm not. go oh that one pin over there has been a bit bent and I'll Attack pin 64 over there decided not to Reflow this just do some drag soldering There we go. Just give those A little dab dab and they should be soldered nicely. It's not pretty.

Um, in fact there are a couple of pet pads over here which I don't think they go anywhere. they will lifted off. Let's have a look. Yeah, those pads don't go anywhere.

So they lift off uh very easily and just the glue on this um PCB material is not that good. So yeah, they lift it off. but the chip reads the chip reads I'm the programmer no workers and I was able to successfully uh program it with the latest version as well. But if I plug it in in the programmer mode I do get insertion error.

That means the processor is now working and the LCD driver is working something. Something's reset itself there. Well look at that. There you go.

So we're just insertion error there. something else is going on. So I'm going to put that transistor back in there because there was nothing wrong with that. Okay, I'm just going to do a quick and nasty Reflow of that old chip onto the donor board.

Sorry, you can't see this. Okay, so this board did work before. It's got those pins. Look absolutely correct.

Let's plug the programmer in. Boom! Yep, no chip. ID So there you go. and I can Buzz out those and definitely those pins are getting over there.

Yep, everything's everything's hunky-dory I can measure volts on there. the Mantis was in the White House using that to inspect and I can actually measure sorry I haven't got the on-screen multimeter 3.3 volts. Yeah, so that's confirmed I transferred the faulty chip onto the donor board which was powering up, but it sort of had like some other issues or something. I can't remember exactly what.

but uh yeah, that was from a junk. uh Ben moldy made up. but it did power on at work and you know it. programmed and all that.

uh, sort of jazz. So yeah, that's confirmed. Um, silicon failed. What? what? what? No idea why? you know.

If you've got any idea, leave any comments down below. Let's power it on. There it is. I was able to program that with the latest firmware.

uh 609. Unfortunately, there's an insertion error I don't know what the insertion area is so whether or not that was a fault that I I don't see how I could cause the that's really annoying. It does that in all ranges. So I think I've had that error before.

Yeah, I've had a 40 meter with that before and I had to replace it at least one of them. So I I don't think I ever got that meter back I Never investigated, but there it is. Um, it is repaired in that it now Powers up. So yeah, we definitely had a faulty processor in that thing.

but now it looks like there's something else wrong and well, that's it for this video. I I I'm done I didn't actually want to repair this thing I just wanted to troubleshoot down to find what the issue was and we definitely found a problem faulty processor and we uh, replaced it. We we deduced that in the first video, but the second video yeah, absolutely confirmed that and I put that chip back on the um back on the other board here and it buzzes through everything. buzzes through fine uh from the programming header and it just doesn't does exactly the same fault as before so there's definitely a dead silicon there.

so there you go anyway. I've got myself a donor board now, which is uh, kind of. Handy and as I mentioned in a live show recently where I actually uh got people to guess and somebody did actually a couple of people I think ended up guessing uh, the uh, goof up I made in that video now a lesser YouTuber of course would have like hid that because they'd be too embarrassed to show that they made a goof like that and well, you know this kind of thing. I I Can't tell you the mindset I was in when I um started thinking that was the DMM chipset it's obvious just from the topology as I explained in the live video, just the topology of the uh thing it was.

it was really obvious, but no one in the comments picked it up. I'm absolutely surprised. That's one of probably my first video ever where I've made a major goof like that major embarrassing goof and nobody's actually picked it up until I prodded some people hey, there's something wrong and they started to put their thinking cap on it. Yeah, um, there you go.

So I hope you enjoyed me leaving in the goose like that? It shows that these things happen. probably wouldn't have done that. Um, well, wouldn't it happen if I had a schematic and it wouldn't have happened if I probably wasn't shooting a video? shooting a video was like different level of stuff your mind gets all you know. anyway.

Anyway, whatever. Um, there it is. it was confirmed faulty Pro processor. So yeah, that insertion error I don't know I might leave that for another video if I've got the time and motivation.

So anyway, I hope you learned some valuable lessons from that video. If you did, please give it a big thumbs up. And as always, discuss down below, catch you next time.