Hi, it's repair time. This space heater thingamabob has, uh, failed. so we're going to take a look at it. It's an Arlec Uh brand, which is one of our large, um, rebadger, uh brands here.

You know it's just a bunnings, um, hardware store, uh kind of thing. And yeah, it's just a passive thing. It's not um, air cooled or anything. It's just got an element in it and uh, you know, turn it on.

Hot air comes out here, it's got a thermostat and bob's your uncle. um, and it, just it doesn't heat at all. But the actual uh, digital display, uh thing on the uh top of it. that actually, uh, works.

But yeah, it seems like the element doesn't turn on. so let's crack it open. And by the way, the uh, rolling feet on these things. Uh, so how you doing it's just got a couple of self tappers into just like the thin metal backing on this thing and ah goodness, it's hopeless.

Anyway, uh, we do have the one controller in here with the uh, digital display. As I said, that works. it seems to do, uh, the business. You just set a uh set point, uh, temperature on the thing and it's supposed to switch on and off.

Uh, the element. That's basically it. You know, cool air in the bottom, warm air out the top when it's on. So these are all self-tappers and ta-da we're in.

There's the element and uh, geez, that's all. Look at how? oh I don't know. you can see how thin that is. but yeah, that's pretty how you doing look at that you can see the self tapper holes.

but you know these things are cheap, right? They're built down to a price. But anyway, there you have it. So as you'd expect, uh, we just got one big element here with uh, some heatsink, uh, fins on it, that? um, and that's it. Looks like we have a uh, temperature cut out there.

You can see that there. They're fairly common. We've seen those before. Um, that's just a uh, over temperature cut out switch.

It's a thermo mechanical type switch so no worries. There's got some nice insulation on that that would be the internal temperature sensor would it? There's not much that can go wrong as a controller which simply, um measures the uh temperature. It might be another temperature sensor in the top, but maybe not because you know it's got some holes here. So that's um, and this is like the inlet temperature, but it is very close to the element.

So yeah, I wouldn't have placed the temperature element there. But anyway, let's have a look at the controller and there we have it. There's not much to it. Big black thing in the back there, that's a relay.

Um, so yeah, that's uh, part of the chain on this thing. Um, so it's basically gonna put a relay. It's going to put the element switching through the relay directly across. uh, the mains input here.

So you know either the element is foley, the relays foldy, or the control circuitry um which is down here. Here's all the intelligent stuff that just you know or whatever. that's just a little uh, micro jobby and um yeah, it basically just switches mains across the elements so there's not a huge amount that can go wrong here. So yeah the first thing I would uh check is uh well a that the elements not uh, busted that we got resistance on the uh element and then uh we would check to see if like the relay is switching voltage across the element.

Obviously if the element's good and you switch your voltage across it, ohm's law, it's going to heat up. Uh, I just noticed something. What the heck are you wedging wires under up under a metal shield for oh like that is. That is a cardinal sin.

You're going to cut through those um, silicon sleeved wires. That's unbelievable. We could say you can see the cut marks in there. Wow wow, that is unbelievable.

They just like it. It hasn't penetrated, but oh goodness what what? That is. Just a ridiculous safety hazard. Was that like just a like it got caught in it was caught in there during.

uh, assembly of this thing. Looks like it's riveted on there. It's pop riveted. That one up there Screwed.

But yeah, do they just keep them under there for cable management? Look, they've done the right thing here. Um, they've designed it properly here. Like they've thought about this. So obviously it's a production mistake on this thing.

Um, because look, they've sleeved it. They've put a, uh, take, like, bent this over so it's not a sharp edge for this single wire to pass through, but this thing jammed under there. Unbelievable. and why they've got two wires going into the red end of the element? I don't know.

because the actual um, they go up to the same point on here, so I don't know. Maybe like there's two physically two separate elements in there. I? yeah, whatever. That makes no difference.

And by the way, I can't see any extra temperature sensor up on here. It's just a micro controller. uh, dual seven segment display, some bi-color Leds and some switches. and Bob and a crystal.

Um, and bob's your uncle. So yeah, it looks like the temperature sensor down in the bottom is the one that measures the ambient. Okay, let's try and get in here to the element. the heat here.

This is the dual red one and the blue one. So we're bypassing the relay here and Ta-da no worries. So that rules out the element and the uh, thermal cut off as well, which will be normally closed. It'll just open if the temperature uh gets too hot.

So whoa, Whoa. What was that measuring 300 there? That was probably my probing. Yep, and if you do 240 volts squared on uh, 28. Ohms, there you go.

2 kilowatts. I believe this is like a 2 kilowatt heater that's under the maximum 2400 watts here. so you know you take out some of the wires as well and no wackers. Um, yeah.

Element and uh, temperature cut over temperature cut out are fine. So that leaves a rather dubious looking Yong Ning relay in there so well. A we should be able to hear that. um, switch on.

but um, no. if it's switched on like there's just like mains in here. Okay, comes to one contact of the relay. So this is red wire coming over here and it literally just contacts over to here and then it goes down.

And this is the neutral, the blue wires, the other, the neutral on the other side. and Bob's your uncle, right? So it should just go straight across the mains. So we need to somehow, uh, probe the relay. But anyway, there's a couple of, uh, a couple of Zener diodes down here.

I'd be, uh, just checking those out. Um, you don't need to check the Zener voltage, just check the reverse uh bias on them and you know, make sure they're okay. There's another resistor in here. Is that just a uh, dropper resistor for the Zeners there? The zeners, but uh, the all the digital circuitry.

um, powers fine. It seems to work. It's just like it doesn't come on, so it's either something to do with the temperature sensor or the relay. Of course, there's a couple of ways that a relay can actually fail.

It could be, you know, dodgy contacts. on the switching side. it could be the switching coil in there. could be open for example.

So you get in there and measure that. I have to get the board out to be able to probe that. or it could be the driving circuitry that usually you'd have like a switch in that transistor, but I didn't see one on the Pcb down the bottom here. So and I don't see one on the top board unless there's a surface mount Joby on the bottom or something.

Aha, that did look a bit too simple, didn't it? Um, so there you go. This is nice. We've got some got a diode bridge happening here, so that's giving us that's the output here. So we've got Ac coming in over here.

That's uh, is that direct? That wouldn't be direct mains would it? Anyway, some nice routed isolation slots around there that's really nice, so that's uh, that's protecting the relay. Anyway, first thing I'm going to do is measure the relay coil. These these contacts look really good. So you know, look for dry joints.

There's something oh good. Oh, they're really. they're shiny as wow. It's almost too shiny for lead free isn't it? Anyway, yeah, you'd be looking for when you've got a power contacts like this.

they can heat up. so you'd be looking for dry joints and stuff like that. but uh, no and on the relays as well. But that looks pretty good.

Ah, what happens. Here's our mains input. This goes over to here. This between here and here is that big power resistor we saw on the top and then that goes over to power the diode bridge over here and then that gives us some.

So that's a big dropper resistor there for powering the secondary side, which is all the electronics over here, which goes over to the controller circuit that we saw. But I believe like all of this works because as I said, the controller does actually work. It seems to go through the motions. It's just like the element doesn't turn on.

That's the only problem. So which pin is what on this relay? Well, obviously. uh, the low voltage uh side is going to be the driving coil. So this is the low voltage side over here and the big power trace here dead giveaway with uh, they've just left off the solder mask so that they can get, um, the solder plate on there to get some extra current carrying capacity.

So obviously, um, you know these are the switching contacts here and we've got another one over here. They're only. it's a single pole double throw there. So they're so it's a single pole double throw and they're only and they're just joining up the one side there.

So obviously this contact down here which is a low voltage side here is the coil and that one there is the coil. Well, this is actually rather funny. This uh, yongneng. their website is literally Chinese Relay.com so it's like it's the Chinese relay.

Unbelievable. Anyway, here's the data sheet for it. Uh, we should get about a 1600 ohm coil because it's a 24 volt relay jobby. So let's gotta pierce through.

make sure you've got sharp probes and pierce through any if there's any gunk or conformal coding. Oh, there you go. Almost bang on there you go. 1600.

So yeah, the coils intact. Okay, so no worries. Driving transistor down in there, you can see that. Q1, save a visual inspection.

make sure that's not blowing. Um, so that would. Obviously it's obviously by the layout there. Um, there's a flyback, uh, protection diode there.

and um, so yeah, that's our driver transistor that could be buggered. but our relay is good, but we don't know about our contacts like internally, like the relay could be shot. Unfortunately, at this point you have to actually power the thing up and then attempt to, you know, operate the controls, get it to a set point temperature, and then actually probe this at the same time. You wouldn't try and do this live.

I'd go in there and like solder some wires off so that then I could attach safely, attach probes and things like that. It's a bit annoying, but you know I was hoping like to find something before. I had to do all this right. So I've measured like the power resistor, the diodes here.

The flyback diode is fine. Uh, our uh, driver transistor? Uh, I've measured the Uh sensor temperature sensor here. I get 11k. meh.

Whatever. it's not, uh, broken so you know that's all good. I I think at this stage I'm just going to, uh, plug it back in and just, uh, confirm that it's uh, you know, still faulty. Oh seriously, these legs are just completely rubbish.

Unbelievable. Come on. taking the piss. Oh, that's oh no.

I thought that was E2. it's upside down. Silly me. that's actually Uh 23 there on the display.

I thought, oh, we've got an error. So it's measuring. that's actually the ambient temperature here. so it's actually measuring that correctly.

And then we can adjust the set point like it's uh 27. If I go like if I go under 23, relay should be off. There you go. so the heater thing's off.

I'll listen carefully for the relay. Okay, I'll go over the temperature so I should switch it on and that Led should come on. Yeah, definitely no relay. Click there.

So yeah, back to the Pcb here. This is a 2a so that's a Pnp. This is positive up here for the relay. so they're switching that with a Pnp and this one over here is a 1a which is an Npn.

So they're actually using like a two-stage drive there. So we're going to have base emitter. so let's measure the diode drop on the base emitter is that good. 0.68 Yep, no worries.

Um, so yeah, like that doesn't mean the transistor is good, but at least the uh, like the base emitters there so that driver transistor seems okay. Probably the easiest next step is to, um, just short out the drive transistor here and like just permanently turn. just put a little jumper link across the collector and emitter and just basically turn on the relay and see if the element comes on, see if it heats up. Okay, bypassed.

let's switch it on. Didn't hear a relay click. It's like anything warm. this is earthed.

I can touch it now. it should warm up fairly quick. I mean 2400 watts. I reckon the Uh relay's gone.

ski. Ah, there you go. I just wanted to make sure that there's enough voltage there and there's not that's directly across the diode bridge. Aha.

13.8 volts. This is a 24 volt rated relay, so of course the relay is not going to activate if you don't have enough drive voltage. Oh, there you go. So it is the power supply like I was fooled by.

like it's like fully working, but there's obviously uh, two rails in here. Um, 13 volts is of course is more than enough to uh, then drop down to the lower voltage to power the micro controller. whatever that is. 5 volt job is 3.3 volts.

Doesn't matter, but it's not enough to operate the relay. So we're actually dealing with a power supply problem. So yeah, that's directly out of the uh, Diode bridge. So switch it to Ac.

We've got uh. 0.2 volts. so it's not like the unit. There's some huge ripple on there.

There's just like there's just not enough volts. 24 volt relay? 13 volts. Actually, I rechecked the Pcb. The relay wasn't actually directly across the Diode bridge, so now I'm actually measuring across the relay, the voltage across the relay coil, and 9.7 volts.

Yeah, no wonder the relays turn not turning on. 24 volt Jobby: Hard to see, but I'm measuring that cap in circuit. and it's 200 microfarads with under an ohm. Um, so that cap seems okay in circuit.

And the Diode bridge Jobby: we're getting 29 at 0.15 Ohms 29 Mike, that is. So yeah, it's supposed to be a 33. Yeah, good enough for Australia. Well, I didn't have a Uh Diabridge in my kit readily available.

might have one, you know, somewhere in the parts drawers. but anyway, um yeah, I'll just rip uh, this one off an old board and give that a go because you know it could be something internally dodgy with the Uh Diode bridge. So give. worth a shot.

And there's just like there's not much else in circuit. Everything measures fine, the dropper resistor fine, the dropper cap, and uh, the dropper bleeder resistor across the cap. That all measures fine. so I don't know.

Diode Bridge? Okay, let's try that again. Yeah. 14 volts? Nope. Okay, this is the Ac side of the Diode Bridge.

What are we getting? Yeah, 15 volts Ac? Well, that explains it. Um, while you're only getting 14 volts, uh, Dc on the other side? Well, here's what happens when you assume I measured this in circuit. and it's the X-class cap which they're using as a divider along with the 47 Ohm series resistor to go into the Diode bridge. And as I said, there's like a uh bleeder resistor across this as well that measured uh, fine.

That was uh, 750k And I was reading 103 Nanofarads So I thought, okay, it's a hundred end cap. That sounds reasonable, but look, it's not. focus your bastard. Look, it's not a hundred n it's 2, 2, 4.

there it is right there. Um, so there you have it. Wow, I did not suspect an X-class cap. I don't think I've like, other than you know, Reefer Madness.

Um, like a modern X-class cap. I I have not seen a failure like that. Um, there's no bulges. Um, no, what's that down there? You know nothing doing right.

A 220 nanofarad cap measures 100 n and yep, that'll do it because it's it's part of the voltage. uh, divider. So yeah, I like wow, what a sneaky bugger. look at that what brand is that dodgy ass? Unbelievable.

Um, 100 Nanofarads supposed to be 220n? And when you're using a capacitive voltage divider like this? a resistor capacitive voltage divider? Yeah, that's going to matter. So it's starving our diode bridge of voltage and we won't get enough voltage to drive our relay. Go figure thou shall measure voltages. And of course Murphy says I don't have a 220 in.

Uh, probably put the kettle on X2 cap in my kit. So yeah, gonna have to. uh, salvage it Like you've got to get like the exact value here. And this is where you get out your old boards, scrap boards and start looking.

No, of course not now. 0.47 of course. Oh geez, we we're coming to Gutsah! Ah, win a winner chicken dinner. Keep your old dead Pc power supplies because look what we have here Bodged across there.

two two, four x two class even looks like the same pitch. Look at that. Ah, Bobby Dazzler snipped it out and it's got a bleeder resistor across the bottom of it. I'll have to snip that out too.

That is one nicely salvaged poly. Put the kettle on. 220 Microfarad X2 class mains, cap, even the correct pitch, same size, physical size and everything. I better make sure I put the right one back in.

Oh come on. seriously. Bloody Murphy 2019 Nano Farads for a 2, 2, 4. Oh no, I'm going to J Car.

I'll be back. So I'm here at J Carr and they still have the old school storage tubs, but wouldn't you know it? wouldn't You know it? the one I need. They do have a 220 N X-class cap but it is this large jobbie where like totally different pin pitch but I can actually bend that back in. I want the pin pitch I want.

Oh Kamagatsa never got them 104. they do have the uh correct pitch um these suntan uh brand ones but these aren't the X2 uh class. They aren't properly self-healing and you know they don't have all the Uh ratings and all the requisite symbols on there. And trust me for this sort of application.

Yeah, you want that so you don't want to be just putting in a generic uh, probably put the kettle on uh one in there. So yeah, we'll have to. uh, just bendy bendy. So luckily Jk is just a kui from here.

but uh yeah, they didn't have the correct pin pitch there on their website. It looked like it was. but um, yeah, when I got there, Nope. um but yeah, we should be able to uh, lead form that and I think we've got enough space on the board, don't we? There you go.

It's a bit of a squeeze in there, but uh, you know, height wise it's no bigger than the uh relay. There's no other like, uh, you know, clearance, uh, issues or anything like that. So yeah, she'll be right. All right, let's see if we get a winner this time.

Oh, I think I heard a relay click and yep, yep, we have an element getting warm Ski, no worries. And yes, I can touch this because this is uh, Maine's earth. Yep. winner winner chicken dinner if I lower this temperature.

Yeah. relay clicks off. Just heard it click, No workers clicked on. You can't hear that, but trust me, I can and we'll measure the bridge output voltage.

Give it a bell. Yeah, there we go. 24 volts. That's more like it.

Your relay's gonna switch on now. All right. So what we had going on here was uh, 240 volt Ac in. We had a big uh power series dropper resistor 47.

Ohm. I measured that early on and it was correct and we had a 220 nano farad X-class cap here. I explained next class in a second. We had the bleeder resistor across here and I measured this early on.

but this is where I come a guts or well. one of the places that I come a guts or in this one is that. Um, I measured it and it was a hundred. I measured a hundred n and I thought you know that sounds like you know that's like bang on a typical value that you might find in a dropper like this.

And and then uh. there was uh two, sorry one and a half mega cross here. these are 750 k each. These are just bleeders to bleed the charge off there when you remove the power.

so it's not dangerous and it goes straight into a Diode bridge like this. Now of course, I didn't follow my own rule of troubleshooting. thou shall measure voltages and if I measured the actual Uh voltage out the rail voltage out of the Diode bridge up front and I would have noticed that it was like half of what we needed to drive our 24 volt relay. So yeah, I could have found this.

I could have saved you know 10, 15 minutes or something of troubleshooting by actually measuring this voltage first. But then you know Murphy says that this would have been fine, it might have been something else. So you know Murphy's going to get you either way. But anyway, the way this works is this is A it's not a full capacity of our divider, but it requires um, this capacitor to be a certain value to give you a certain capacitive reactance or an Ac resistance.

And of course the formula is one over two. Pi Fc F is the mains frequency which is 50 Hertz here in Australia now that 60 Hertz rubbish and the capacitive value. So that dodgy cap in there being a hundred nano farads although who knows what it is at voltage, right that sort of measures low voltage on our cap meter. but you know who knows what right, It's it's dodgy as right.

It's dodgy brothers, so that is going to lead to the smaller the capacitance value. That's going to lead to a higher series resistance here. and that's going to starve your diode bridge. And that's why we're getting a reduced voltage out here.

So yeah, um, you need the correct value in there. So in this sort of mains application, you need what's called an X2 class capacitor and they call it an X-class capacitor because it goes across the mains. If it was a Y-class capacitor, then it'd be going down like this. So they call that a Y.

Well, so they call that a Y class capacitor because you can have a capacitor go into. you know, both of, uh, the mains inputs to ground for noise suppression. So that would be a Y-class capacitor of an X-class capacitor goes across the mains. and you need these specific ones with all the requisite uh, you know there's underwriters, laboratory, and all sorts of stuff.

Um, these are self-healing capacitors. Um, so they you know they fail safe and they've got self-healing dielectric and and stuff. So um, yeah, these things you know you don't want to just use a simple non-x class. um, high voltage capacitor on there.

even if it's rated for the voltage you don't want it. You want a specifically an X-class rated cap. So X and Y. So we use an X-class in this application because it's effectively across the mains.

whether it's being used as a divider like this, or whether it's being used directly across the mains as some filtering or suppression doesn't matter. You need the X class, But I'm glad we went down the path we did, because, um, you know it shows troubleshooting steps. I mean, it could have easily been the element. I mean, you know that's obviously the first thing you measure to make sure your elements are not busted.

It could have been the thermal cutout switch up here. It could have been a dry joint on the Pcb. From continued you know, heat cycling and stuff like that, it could have been the thermistor sensor down here not turning on. It could have been the Uh drive transistor not turning on.

It could have been like the Diode bridge faulty. like we actually, uh, physically, uh, replaced it and it wasn't that. And we just happened to measure like because I measured this sucker in circuit and if I measured like nothing right, if it was like open, then uh, like I would have gone. Oh well.

Okay, the cap's gone. but it measured a hundred nanofarads. Which happens to be precisely the value that you know, I pray a pr the most popular value for these classic capacitors. Um, so I thought that was fine and I just didn't double check it with the actual value on there.

I should have double checked. and if I did, I would have. Well, I eventually did and I realized that I didn't do it at first. So there you go.

I hope you found that and interesting. uh, repair. We went down the rabbit hole. It didn't take too long.

Uh, we could have found it a bit earlier. Maybe if I measured that uh, dia bridge. but yeah, like a diode. Could it like the coil? Of course.

Uh, it could have been faulty. It could have been the contacts. you know, dodgy contacts in the real house something. but I couldn't hear it switching on.

So you know usually you can hear these big mechanical relays switch on. I wasn't hearing that so I knew it wasn't getting the drive required and it could have been the drive circuit could have been. the coil could have been, uh, anything, uh, like that. turns out to be a dodgy cap starving the thing of voltage in it.

and it had enough voltage of course to power all the uh screen. Because as I said, there's like a two layers of voltage regulation in there. There's one, the 24 volts which is required for the relay, and then there's going to be like another five volts, uh, somewhere or something for driving the micro and the Uh display and stuff. And if the 24 volts is low like it was, it doesn't affect all the digital stuff.

That all still works fine. But and and the transistor was, You know it was trying to switch on the relay, but there just wasn't enough drive voltage enough. Uh, drive coil current to, uh, activate the relay. So anyway, that was a rather it.

Not often we get one that interesting. Could have been boring, like a broken coil or a dodgy solder joint or something. But anyway, I hope you found it interesting. If you did, please give it a big thumbs up.

As always, discuss down below. catch you next time.