It's not the 1M resistor dropping the voltage or limiting current - it's the blue X2 capacitor. This kind of circuit is called a capacitive dropper. It works based on the dynamic impedance of the capacitor to AC. The value of the capacitor (probably some hundreds nF) dictates the output voltage and maximum current. The 1M resistor is just a bleeder for the cap's stored charge so you don't get zapped touching the mains input terminals even after unplugging. Funny you should mention Big Clive as he frequently covers these kind of circuits.
Also... you should never use this type of circuit in anything with exposed terminals, as the 'dropped' voltage is with respect to mains (240 RMS, 338 DC). So there could be 338v at one side of the relay coil and 314v at the other, for example, but the coil only 'sees' 24v.
I would think if you bridge the R12 with a solder blob or 0ohm resistor, you'd get the 4h "mode" (30, 60, 120, 240 min). My guess is it literally halves the clock. You could add a small toggle switch to R12, and choose between 2h and 4h mode too, all the versatility! I was on holiday in Cyprus last year and the landlord had put the 4h units on all the Aircon units!! Annoying!
Surely it could have been improved! I suppose just installing the capacitor a few mm from the PCB would have increased its life considerably. Maybe. Maybe not that brand :)
Cool fix. My thoughts: 1. Capacitive droppers (already explained by HwAoRrDk) are generally okay (for low power stuff), but they're not very efficient and don't provide galvanic isolation, so everything has to be in a double-isolated enclosure. 2. Another failure modes of those power supplies is the dropper capacitor itself. These X/Y rated caps are the "self-healing" type. If there's an over-voltage on the mains (e.g. nearby lightning strike, or surges from the transformer station switching lines) it can cause arcing inside the cap, creating a short. But these caps are designed so that the short burns itself out and the cap keeps functioning. But obviously, the burned area is no longer available for storing an electric charge, so the overall capacitance drops slightly. If that happens often enough, eventually the capacitance will drop to a point where the cap can no longer provide enough current for the circuit. I had this problem with a remote controlled ceiling fan. The control board used a cap dropper, and every time I turned the fan on, it would reset. Turned out the dropper cap had gone from 100nF to about 10nF and had to be replaced. I know this wasn't the issue here but I think it's worth keeping in mind. 2. With just 50 or 60 Hz mains, you can often measure the ripple using your multi-meter in AC mode. 3. This thing is powered all the time, right? And it probably isn't super efficient. You might want to measure the quiescent power draw, and cross-reference with your electricity bill to see if it's worth keeping around. That fan I mentioned above? It turned out it cost me 50€ (~42£) per year, just for literally hanging around. A better quality unit with proper low standby power might pay for itself within a few years.
Thanks, that helps a lot! It's nice to learn new things every day! I have a switch to isolate that circuit so now I am switching it off when not being used. The downside might be - I guess - that the switch might fail sooner. Good point about measuring the quiescent current - energy is not cheap these days. Thanks for taking the time to leave this message!
Amazing job mate. Hope this video encourage people to repair appliances, as maintenance companies try to scalp you on every occasion. Also about the value of not producing more e-waste that could work again. BTW, I just saved my boiler main board two months ago, as the maintenance company kindly offered a "reposition" from 350 quid while I spent 10 in components (a cap blown and killed "the" TOP244 AC/DC converter). Please, you all, try to fix your home stuff !!! Tony set the example here 😀
For me it's mostly a "let's reduce waste" exercise. I appreciate that repairing these items cannot be done for a living (it's worth £20, I cannot charge someone £5 to spend some time on it!) but we cannot dispose of everything all the time! Well done with your main board!
@@Tony359_2 Of course we can't. Nigh on every electronic device I own was repaired or bought 2nd hand. My alarm clock was used. Works fine for me. My laptop is 12 years old and doing find with some upgrades; also brought used. My microwave is 25 years old, even my laptop charger has a terminal block in the middle and a 3D printed cover for it. I'm not suggesting people go out and do the same thing but it's nice to know you're getting more life out of the devices than was intended!
Over the last few years I have had a few cases (water heater, Yamaha amplifier and several socket power meters) where a film capacitor from a transformerless power supply lost capacitance and was unable to supply enough current to the system. Often, the decrease in the capacitor's capacity causes the next element of the circuit - the resistor - to overheat. It is possible that this is not the case with this particular device, but I wonder if the capacitor has its original capacity. Quote of the Day: "Metallized polypropylene film capacitors are known to be one of the most common causes of failure in electronic systems"
@@Tony359_2 I first read about transformerless power supply in the late 90s. There was even made a special IC for it. But if something can be done cheaper, it will be done that way. 😁
@@Tony359_2 you should not think of it as a transformer. The capacitive dropper supply is more like a resistor divider, just using the capacitor as series device instead of a resistor. It is also useful to think of it as a current source rather than voltage. As an example: 330nF has about 10k impedance at 50 Hz. If you connect it to a rectifier and 24V zener you will see approximately 200V AC across the capacitor which then results in 20mA AC current through the cap. This current must be sunk by the circuit. Here your zener is probably red hot because it is undersized for burning around 0.5W continuously while the relay is not on. Typical junk grade electronic design IMO. The zener would need larger package (SMC would be my choice) pretty much solid planes connected to not run stupidly hot.
I would also add that in my view capacitor failures are not the big problem today. With LED lighting integrated in fixtures there are lots of instances where poor cooling or general thermal design (like this device) lead to burn out much before the LEDs get worn out.
The neutral appears to be going through the film cap, so it could be using a capacitive dropper to lower the voltage across the actual circuit, film caps can go bad too and lose capacity.
if you make legs to the diode and made it flying, it will run more cold. Also you can change the diode for a stronger one, probably not smd but it will be capable to run more power through it. The capacitor can be moved using coated wires, since it is 24v and low current, it will prevent degradation by high temps.
I'm now just switching it off when not in use! :D Particularly in summer time it stays in "standby" for months... Had I known it was self destructing inside...
Great video, I like the dual camera set-up. As others have said, it's a capacitive dropper - a cheap way to drop a large AC voltage at a very small amount of current. Inductors are expensive. Keep learning and taking us along for the ride, subscribed
It's surprising to see how few components are actually needed to make a circuit like this function. Also if you were to try and repair one of these at home, you'd now have a good idea of what to replace.
Hi Tony. Interesting circuit today. There's a lot to be said for drawing out a schematic like Clive does. It's a great learning / teaching ade. So you've had a Win with this one.👍😎
I have recently had exactly the same problem with mine. The only difference is that I also needed to replace the zener diode. I had previously replaced the X2 capacitor a few years ago, as its capacitance had dropped significantly.
I have no idea if this is totally normal, but my dolce gusto coffee maker also seems to have a simillar circuit for powering the "computer" inside it, i've seen that the last time i had to fix it because its 110V and someone plugged it in 220v (we have both voltages around here where i live in the same electrical installation).
Hey Tony I liked this one. Always better to keep something away from the landfill. It is certainly satisfying to know that many things can be repaired quite ‘easily’. congrats!
Those are capacitive droppers. Those are not isolated and basically mains network see same current constantly but phase shifted (bad power factor). This is why they usually use 24 or higher volt relays to keep max needed current down. These are usually used mains devices where is low volt control circuit but case is working as insulation. Like room thermostat for floor heating or dimmer for lights.
The X2 capacitor will fail some day, it can take many years. This type of capacitor is not designed to carry current. I talked to a capacitor manufacturer and he said X2 capacitors are only designed to last 8 months to get them thru EMI testing. I see a lot of engineers use X2 capacitors to make it easier to get thru UL and because they just don't know. Across the line capacitors are designed to fail safely by clearing any fault. Eventually there is nothing left to conduct. Resistors in series also fail internally from surges opening up resistive paths. This is very common with carbon film resistors and have replaced a lot of them in capacitive dropping circuits. They will never look burnt, just open.
Thank you - only 8 months?? Is it 8 months under harsh conditions or 8 months and that's all? This capacitive dropper (sorry I didn't know about them when I prepared the video!) has been running 3 years 24/7. I was lucky then :)
Do you have a thermal camera? As this will show any component that is under thermal stress, say underated or a fault condition, it would be handy in this sort of repair or finding poor undrated components on mainly Chineese equipment. Thanks for the well narrated video.
Did not see the video to the end. Yes, the B & Q ect will buy products that work and usually meet the new (post EU) standrds, but they do not take products apart to check them relying on the original company or importers, for quality. That is why you get product recall notices in such stores and sometimes in the daily newspapers. Be very careful when buying on eBay as products are comming in with non-standard plugs which don't have a fuse this is very dangerous also some of the switch mode power supplies have been banned from use since Sptember 2023 but are still being sold on eBay. These are laptop power supplies that could give mains voltage at the low voltage DC output, thus destroying the laptop or seriously damaging someone.
I refuse to buy power supplies on Ebay/Aliexpress. Too dangerous. This is not BQ BTW :) But yes, of course they just make sure the product is following regulations so they're covered.
When replacing capacitors that have died due to high thermals over time, it's useful to leave them long-leaded and space them away from the PCB as far as reasonable. I'm not surprised the zeners are toasty all the time, even when the switch isn't thrown, classic capacitive-dropper zener topology really.
Yes, I am learning this now :) There is no space for extra legs for the cap unfortunately. There is a plastic cover which is just on top of it. I guess it could have been designed so the zener didn't end up under the capacitor. But then the device would last many years :)
How much power is this wasting when the towel radiator is off. With that zener at 95+ degrees C all the time, it must be wasting several watts. What a crappy design. I would modify it so when the relay switches off, the whole thing switches off. You could have another push button to turn it on.
To modify it, you cut the trace that goes from live in to the input of the film capacitor and run a wire from the output of the relay to the input of the film cap. Then you add a momentary switch between live in and live out to turn it on. Once the relay is energised, it keeps its self on. When the relay turns off, the whole timer turns off completely. You save money on your electric bill and the zener and capacitor last longer.
It can't waste several watts when we're talking about half a watt diode here to begin with ;) Just because something tiny is hot doesn't mean a huge amount of energy is wasted. Capacitor droppers are used in almost all low power double isolated devices, you'll probably be surprised how many things you own have that same "crappy" design inside them.
I don't think so. I suppose if something goes wrong, there is not much to burn in there, traces are small. Does the big 1Mohm resistor act as a current limiter?
A DMM definitely is too slow to properly troubleshoot this sort of problem. There is obviously something happening in the circuit that happens many times per second (causing the light to flicker and the relay to keep cycling on and off), and a simple meter will not show you anything that fast. You really need an oscilloscope if you want to have any real idea what's going on in this circuit at those speeds. Basic oscilloscopes really do not cost that much these days, so if you don't have one I'd really recommend getting at least a cheap one. There are lots and lots of things you just can't see or understand a lot of the time if all you have is a multimeter. Also, your explanation of what was happening I think is not quite correct. Just because there's another zener there does not necessarily mean the IC is "fine". Also, the relay wouldn't actually care that much about ripple in the supply voltage (it's just a coil, its own (quite high) internal inductance will actually counteract most ripple or other small variations in the power it sees anyway). What was almost certainly happening is actually that the capacitor was just good enough to supply power to the whole setup as long as the relay was not engaged; however, the moment that the relay was turned on, the capacitor could not compensate for the abrupt change in load current, which caused the power supply voltage to suddenly drop, and that caused the IC to "brown out" (undervoltage). When that happened, it turned off the relay, at which point the supply voltage bounced back up to normal. Then the IC wakes up and says "oh, wait, I'm supposed to be turning on the relay now", turns on the relay agaiin, and the whole cycle repeats. A zener does not act "just like a capacitor" in terms of smoothing. One important difference in this case is that a zener will only pull *down* the voltage if it tries to go above its limit, but it will do absolutely nothing at all if the voltage goes *too low.* And that is almost certainly what is actually happening in this case. (Oh, and yes, it is very common and normal for even reputable electronics to contain chips with removed identifiers and such. IMHO, it is actually *more* common to find this with many reputable higher-end devices, because the manufacturers making them actually spent their own R&D time and money developing a working circuit design, and they don't want some knock-off shop just stealing their whole design and churning out tons of fake copies for cheap, so they do this to try to make it harder for people to clone their hard work. The really cheap manufacturers don't care as much if somebody else clones their design, because it wasn't actually their design to begin with anyway, so they are actually the ones who often don't bother doing things like scraping off the numbers. Filed-off identifiers is actually often a *good* indicator regarding circuit quality, because it means somebody has actually spent their own time and effort to design a good circuit and they have pride in what they produced and don't want other people stealing their work.) Oh, I do also agree that putting a capacitor right next to a zener that is constantly drawing a significant amount of current (and producing a fair bit of heat) at all times was not a very smart design decision, though, and almost certainly is the reason why this device failed so quickly. If there is room in the case, I might actually recommend soldering some slightly longer wires on that capacitor and moving it a bit further away from that zener, so the new one is not subjected to the same kind of heat stresses that the old one was.
Thanks for your comment. I do have an oscilloscope! I didn't use it on this occasion. Yes, I appreciate mine was a crude analysis - with some inaccuracies. What I mean about the zener on the IC was that it is used to regulate the voltage from 24V DC to 5DC. Unless there is a MASSIVE ripple on the 24V line, that should not affect the IC which is regulated by the zener 19V below the expected voltage. I assumed that the capacitor still had a little capacitance left and could at least cover for those 5V required by the IC. Maybe I was wrong. Thanks for watching!
I have a feeling Chinese electronic components quality in 2020's is like the quality of Chinese cars in the 1990's. Specially capacitors inside cheap electronics. So far, and after watching so many repair videos, the no.1 failed component is a Chinese made capacitor. :( Please don't get me wrong, Chinese factories build component to a price. If you want a Chinese capacitor with quality similar or close to a Rubicon one, you can find, but not in the cheap electronics that floods Amazon and other markets.
a good point I never considered: good capacitors are Japanese. That said, I suspect that a low-grade Nichicon, rated 85C, running in an oven for 3 years might have had a similar outcome :)
@@Tony359_2 You have a point and I suspect a PCB using jap caps like Nichicon will have a better design where the capacitor will not be running all the time but only when activate the boost function. Anyhow, great video.
Maybe. But I assume it's the worst of the worst quality, which means something like 1000 hours at 100C. If it's been running at 60C constantly, 24/7, it won't take much to die!
@@Tony359_2 Okay. I experience more and more of one thing in the country where I live. The quality control does not seem to exist and more and more shops sell AliExpress cheap type of electronics.
Tldr: zener cooked main smoothing cap end of the story Total classic of failure You could wire the cap with wires away from it to save it from the heat Even a good caps are rated maybe at max to like 10k hours under 70-90c
Well, the zener is at 95C, the cap is on the other side. Let's say 50/60C all the time? If the cap is "crap" and rated 1000h at 85C, then it won't last long :)
The IC is probably a microcontroller, there's not much to say except that the analysis was very didactic. In Chinese devices, you should always be wary of electrolytic capacitors. The average lifespan is about 5 years, but when they were manufactured in the USA, Japan, European countries, etc., they lasted decades. As for resistors, I worked for a year assembling boards in the 80s. Up to 4 colors (digit, digit, multiplier, tolerance) I knew the value as soon as I saw them, for more color bands it was a mess.
I don't work with resistors very often so I always find it challenging - particularly because sometimes it's easy to read them the other way round! :) Capacitors yes, I guess it's easy to make a crap one and then pretend they're good on the datasheet.
Most producers would tell You, that this is perfect design when device fail right after warranty period. This looks like aliexpress quality, but I saw a lot better power supply's there. If I'm not mistaken, currently EU demands, that device in standby need to draw less than 0.1W, so I guess that this constantly being hot would not fit in that description. Something shady, in my opinion, is going on here.
@@Tony359_2not necessarily. In the UK the warranty is 6 years despite what they try to tell you. You do have to prove that it's a design fault or manufacturing defect that was present when purchased though.
I probably don't want to know! :D The device is "safe". It has double insulation and if something goes wrong I really don't think anything too bad can happen. As usual, a £20 item sells better than a £30 which is designed to last 3 times longer.
Yes, I've just educated myself about that, I was not aware! How do you learn things? :) How would have learnt about this type of capacitive PSUs if I hadn't tinkered with this PCB and made this video?
That's the whole point of this channel... So Tony can learn and show his findings to us... Stop gatekeeping. When people like you do things like that, no one learns, knowledge is lost and everyone ends up less enthusiastic and worse off as a result. From watching this video, I learnt plenty. Especially from the comments but yours wasn't useful at all unfortunately :(
It's not the 1M resistor dropping the voltage or limiting current - it's the blue X2 capacitor. This kind of circuit is called a capacitive dropper. It works based on the dynamic impedance of the capacitor to AC. The value of the capacitor (probably some hundreds nF) dictates the output voltage and maximum current. The 1M resistor is just a bleeder for the cap's stored charge so you don't get zapped touching the mains input terminals even after unplugging. Funny you should mention Big Clive as he frequently covers these kind of circuits.
Thanks - I was totally not aware of this type of circuit. I'll look into that and learn more!
Capacitive dropper is Clive's name for these things. Electronically speaking it's a capactive reactance circuit.
Also... you should never use this type of circuit in anything with exposed terminals, as the 'dropped' voltage is with respect to mains (240 RMS, 338 DC). So there could be 338v at one side of the relay coil and 314v at the other, for example, but the coil only 'sees' 24v.
The "NO/YES" fully reads "R12 NO 2H" vs. "R12 YES 4H", so presumably you can adjust the timer with that resistor.
ahh! Yes there are two models, one goes to 2h and one to 4h. Interesting, the 4h one is more expensive LOL!
I would think if you bridge the R12 with a solder blob or 0ohm resistor, you'd get the 4h "mode" (30, 60, 120, 240 min).
My guess is it literally halves the clock.
You could add a small toggle switch to R12, and choose between 2h and 4h mode too, all the versatility!
I was on holiday in Cyprus last year and the landlord had put the 4h units on all the Aircon units!! Annoying!
Nice to see it came down to the usual 'crapacitor' & not the toasty looking Zener!
Surely it could have been improved! I suppose just installing the capacitor a few mm from the PCB would have increased its life considerably. Maybe. Maybe not that brand :)
Cool fix. My thoughts:
1. Capacitive droppers (already explained by HwAoRrDk) are generally okay (for low power stuff), but they're not very efficient and don't provide galvanic isolation, so everything has to be in a double-isolated enclosure.
2. Another failure modes of those power supplies is the dropper capacitor itself. These X/Y rated caps are the "self-healing" type. If there's an over-voltage on the mains (e.g. nearby lightning strike, or surges from the transformer station switching lines) it can cause arcing inside the cap, creating a short. But these caps are designed so that the short burns itself out and the cap keeps functioning. But obviously, the burned area is no longer available for storing an electric charge, so the overall capacitance drops slightly. If that happens often enough, eventually the capacitance will drop to a point where the cap can no longer provide enough current for the circuit. I had this problem with a remote controlled ceiling fan. The control board used a cap dropper, and every time I turned the fan on, it would reset. Turned out the dropper cap had gone from 100nF to about 10nF and had to be replaced. I know this wasn't the issue here but I think it's worth keeping in mind.
2. With just 50 or 60 Hz mains, you can often measure the ripple using your multi-meter in AC mode.
3. This thing is powered all the time, right? And it probably isn't super efficient. You might want to measure the quiescent power draw, and cross-reference with your electricity bill to see if it's worth keeping around. That fan I mentioned above? It turned out it cost me 50€ (~42£) per year, just for literally hanging around. A better quality unit with proper low standby power might pay for itself within a few years.
Thanks, that helps a lot! It's nice to learn new things every day! I have a switch to isolate that circuit so now I am switching it off when not being used. The downside might be - I guess - that the switch might fail sooner. Good point about measuring the quiescent current - energy is not cheap these days.
Thanks for taking the time to leave this message!
My sympathy to the relay contacts under this fault condition..
LOL!
Amazing job mate. Hope this video encourage people to repair appliances, as maintenance companies try to scalp you on every occasion. Also about the value of not producing more e-waste that could work again. BTW, I just saved my boiler main board two months ago, as the maintenance company kindly offered a "reposition" from 350 quid while I spent 10 in components (a cap blown and killed "the" TOP244 AC/DC converter). Please, you all, try to fix your home stuff !!! Tony set the example here 😀
For me it's mostly a "let's reduce waste" exercise. I appreciate that repairing these items cannot be done for a living (it's worth £20, I cannot charge someone £5 to spend some time on it!) but we cannot dispose of everything all the time! Well done with your main board!
@@Tony359_2 Of course we can't. Nigh on every electronic device I own was repaired or bought 2nd hand. My alarm clock was used. Works fine for me. My laptop is 12 years old and doing find with some upgrades; also brought used. My microwave is 25 years old, even my laptop charger has a terminal block in the middle and a 3D printed cover for it.
I'm not suggesting people go out and do the same thing but it's nice to know you're getting more life out of the devices than was intended!
Over the last few years I have had a few cases (water heater, Yamaha amplifier and several socket power meters) where a film capacitor from a transformerless power supply lost capacitance and was unable to supply enough current to the system. Often, the decrease in the capacitor's capacity causes the next element of the circuit - the resistor - to overheat. It is possible that this is not the case with this particular device, but I wonder if the capacitor has its original capacity.
Quote of the Day:
"Metallized polypropylene film capacitors are known to be one of the most common causes of failure in electronic systems"
I need to see how that works, I wasn't aware that a capacitor could act as a transformer!
@@Tony359_2 I first read about transformerless power supply in the late 90s. There was even made a special IC for it. But if something can be done cheaper, it will be done that way. 😁
@@Tony359_2 you should not think of it as a transformer. The capacitive dropper supply is more like a resistor divider, just using the capacitor as series device instead of a resistor. It is also useful to think of it as a current source rather than voltage.
As an example: 330nF has about 10k impedance at 50 Hz. If you connect it to a rectifier and 24V zener you will see approximately 200V AC across the capacitor which then results in 20mA AC current through the cap.
This current must be sunk by the circuit. Here your zener is probably red hot because it is undersized for burning around 0.5W continuously while the relay is not on. Typical junk grade electronic design IMO. The zener would need larger package (SMC would be my choice) pretty much solid planes connected to not run stupidly hot.
I would also add that in my view capacitor failures are not the big problem today. With LED lighting integrated in fixtures there are lots of instances where poor cooling or general thermal design (like this device) lead to burn out much before the LEDs get worn out.
thanks for helping me understanding how a capacitance dropper works! Fascinating!
The neutral appears to be going through the film cap, so it could be using a capacitive dropper to lower the voltage across the actual circuit, film caps can go bad too and lose capacity.
I didn’t know about capacitive droppers, I’m learning from comments!
In the UK when a relay does what you describe we say it’s chattering.
ahah cool!
if you make legs to the diode and made it flying, it will run more cold. Also you can change the diode for a stronger one, probably not smd but it will be capable to run more power through it. The capacitor can be moved using coated wires, since it is 24v and low current, it will prevent degradation by high temps.
I'm now just switching it off when not in use! :D Particularly in summer time it stays in "standby" for months... Had I known it was self destructing inside...
I would use a bzd27cxxx.
I always upgrade my rigs when I have spare time 🤣 there always some engineering flaws to solve and give them more years of life
Thanks for these tips. We could potentially get 10 years out of this cheaply built rubbish as opposed to the 1.01 the retailer hopes for.
you'd be surprised it comes with 3y warranty! LOL! It failed after 3.1 years :)
Great video, I like the dual camera set-up. As others have said, it's a capacitive dropper - a cheap way to drop a large AC voltage at a very small amount of current. Inductors are expensive. Keep learning and taking us along for the ride, subscribed
Thanks and welcome!
It's surprising to see how few components are actually needed to make a circuit like this function. Also if you were to try and repair one of these at home, you'd now have a good idea of what to replace.
Yes, this is "dangerous" but has a double insulation so it's allowed apparently.
Hi Tony. Interesting circuit today.
There's a lot to be said for drawing out a schematic like Clive does. It's a great learning / teaching ade.
So you've had a Win with this one.👍😎
BigClive is the Master. I'm not even the Padawan, I'm JarJar Binks :)
I have recently had exactly the same problem with mine. The only difference is that I also needed to replace the zener diode. I had previously replaced the X2 capacitor a few years ago, as its capacitance had dropped significantly.
More design flaws, amazing! :D
I have no idea if this is totally normal, but my dolce gusto coffee maker also seems to have a simillar circuit for powering the "computer" inside it, i've seen that the last time i had to fix it because its 110V and someone plugged it in 220v (we have both voltages around here where i live in the same electrical installation).
Hey Tony
I liked this one. Always better to keep something away from the landfill.
It is certainly satisfying to know that many things can be repaired quite ‘easily’.
congrats!
thanks. It's also annoying to see that price is more important than longevity when designing something.
I think, it's atransformer less power supply which contains x capacitor , resistor and bridge rectifier used only in low power
Those are capacitive droppers. Those are not isolated and basically mains network see same current constantly but phase shifted (bad power factor).
This is why they usually use 24 or higher volt relays to keep max needed current down.
These are usually used mains devices where is low volt control circuit but case is working as insulation. Like room thermostat for floor heating or dimmer for lights.
Good rule of thumb: it's usually the electrolytic capacitor. They are a major point of failure.
yes but it would have been boring to just swap it and call it a day! :) Thanks for watching!
Thumbs up for doing sketchy testing methods. 😂
The X2 capacitor will fail some day, it can take many years. This type of capacitor is not designed to carry current. I talked to a capacitor manufacturer and he said X2 capacitors are only designed to last 8 months to get them thru EMI testing. I see a lot of engineers use X2 capacitors to make it easier to get thru UL and because they just don't know. Across the line capacitors are designed to fail safely by clearing any fault. Eventually there is nothing left to conduct. Resistors in series also fail internally from surges opening up resistive paths. This is very common with carbon film resistors and have replaced a lot of them in capacitive dropping circuits. They will never look burnt, just open.
Thank you - only 8 months?? Is it 8 months under harsh conditions or 8 months and that's all? This capacitive dropper (sorry I didn't know about them when I prepared the video!) has been running 3 years 24/7. I was lucky then :)
Do you have a thermal camera?
As this will show any component that is under thermal stress, say underated or a fault condition, it would be handy in this sort of repair or finding poor undrated components on mainly Chineese equipment.
Thanks for the well narrated video.
Yes, keep watching till the end :)
Did not see the video to the end.
Yes, the B & Q ect will buy products that work and usually meet the new (post EU) standrds, but they do not take products apart to check them relying on the original company or importers, for quality.
That is why you get product recall notices in such stores and sometimes in the daily newspapers.
Be very careful when buying on eBay as products are comming in with non-standard plugs which don't have a fuse this is very dangerous also some of the switch mode power supplies have been banned from use since Sptember 2023 but are still being sold on eBay.
These are laptop power supplies that could give mains voltage at the low voltage DC output, thus destroying the laptop or seriously damaging someone.
I refuse to buy power supplies on Ebay/Aliexpress. Too dangerous. This is not BQ BTW :) But yes, of course they just make sure the product is following regulations so they're covered.
When replacing capacitors that have died due to high thermals over time, it's useful to leave them long-leaded and space them away from the PCB as far as reasonable. I'm not surprised the zeners are toasty all the time, even when the switch isn't thrown, classic capacitive-dropper zener topology really.
Yes, I am learning this now :)
There is no space for extra legs for the cap unfortunately. There is a plastic cover which is just on top of it. I guess it could have been designed so the zener didn't end up under the capacitor. But then the device would last many years :)
very nice video. thanks for the contents. I am reaching to this conclusion that Capacitors are bad :D
they are - but also necessary unfortunately!
How much power is this wasting when the towel radiator is off. With that zener at 95+ degrees C all the time, it must be wasting several watts. What a crappy design. I would modify it so when the relay switches off, the whole thing switches off. You could have another push button to turn it on.
Yes, I am now turning it off from the main switch every time! It's so silly!!
To modify it, you cut the trace that goes from live in to the input of the film capacitor and run a wire from the output of the relay to the input of the film cap. Then you add a momentary switch between live in and live out to turn it on. Once the relay is energised, it keeps its self on. When the relay turns off, the whole timer turns off completely. You save money on your electric bill and the zener and capacitor last longer.
the "momentary switch" can be a problem. It's a bathroom, I cannot drill a hole into the case! I mean, I could, but might not be the greatest idea!
It can't waste several watts when we're talking about half a watt diode here to begin with ;)
Just because something tiny is hot doesn't mean a huge amount of energy is wasted.
Capacitor droppers are used in almost all low power double isolated devices, you'll probably be surprised how many things you own have that same "crappy" design inside them.
learning new things here! Thanks!
nice, but the logic has a fuse?
I don't think so. I suppose if something goes wrong, there is not much to burn in there, traces are small. Does the big 1Mohm resistor act as a current limiter?
@@Tony359_2 i think that the capacitor and the resistor act as voltage reduction, but yes is a cheap object
A DMM definitely is too slow to properly troubleshoot this sort of problem. There is obviously something happening in the circuit that happens many times per second (causing the light to flicker and the relay to keep cycling on and off), and a simple meter will not show you anything that fast. You really need an oscilloscope if you want to have any real idea what's going on in this circuit at those speeds. Basic oscilloscopes really do not cost that much these days, so if you don't have one I'd really recommend getting at least a cheap one. There are lots and lots of things you just can't see or understand a lot of the time if all you have is a multimeter.
Also, your explanation of what was happening I think is not quite correct. Just because there's another zener there does not necessarily mean the IC is "fine". Also, the relay wouldn't actually care that much about ripple in the supply voltage (it's just a coil, its own (quite high) internal inductance will actually counteract most ripple or other small variations in the power it sees anyway).
What was almost certainly happening is actually that the capacitor was just good enough to supply power to the whole setup as long as the relay was not engaged; however, the moment that the relay was turned on, the capacitor could not compensate for the abrupt change in load current, which caused the power supply voltage to suddenly drop, and that caused the IC to "brown out" (undervoltage). When that happened, it turned off the relay, at which point the supply voltage bounced back up to normal. Then the IC wakes up and says "oh, wait, I'm supposed to be turning on the relay now", turns on the relay agaiin, and the whole cycle repeats.
A zener does not act "just like a capacitor" in terms of smoothing. One important difference in this case is that a zener will only pull *down* the voltage if it tries to go above its limit, but it will do absolutely nothing at all if the voltage goes *too low.* And that is almost certainly what is actually happening in this case.
(Oh, and yes, it is very common and normal for even reputable electronics to contain chips with removed identifiers and such. IMHO, it is actually *more* common to find this with many reputable higher-end devices, because the manufacturers making them actually spent their own R&D time and money developing a working circuit design, and they don't want some knock-off shop just stealing their whole design and churning out tons of fake copies for cheap, so they do this to try to make it harder for people to clone their hard work. The really cheap manufacturers don't care as much if somebody else clones their design, because it wasn't actually their design to begin with anyway, so they are actually the ones who often don't bother doing things like scraping off the numbers. Filed-off identifiers is actually often a *good* indicator regarding circuit quality, because it means somebody has actually spent their own time and effort to design a good circuit and they have pride in what they produced and don't want other people stealing their work.)
Oh, I do also agree that putting a capacitor right next to a zener that is constantly drawing a significant amount of current (and producing a fair bit of heat) at all times was not a very smart design decision, though, and almost certainly is the reason why this device failed so quickly. If there is room in the case, I might actually recommend soldering some slightly longer wires on that capacitor and moving it a bit further away from that zener, so the new one is not subjected to the same kind of heat stresses that the old one was.
Thanks for your comment. I do have an oscilloscope! I didn't use it on this occasion.
Yes, I appreciate mine was a crude analysis - with some inaccuracies. What I mean about the zener on the IC was that it is used to regulate the voltage from 24V DC to 5DC. Unless there is a MASSIVE ripple on the 24V line, that should not affect the IC which is regulated by the zener 19V below the expected voltage. I assumed that the capacitor still had a little capacitance left and could at least cover for those 5V required by the IC. Maybe I was wrong.
Thanks for watching!
I have a feeling Chinese electronic components quality in 2020's is like the quality of Chinese cars in the 1990's. Specially capacitors inside cheap electronics.
So far, and after watching so many repair videos, the no.1 failed component is a Chinese made capacitor. :(
Please don't get me wrong, Chinese factories build component to a price. If you want a Chinese capacitor with quality similar or close to a Rubicon one, you can find, but not in the cheap electronics that floods Amazon and other markets.
a good point I never considered: good capacitors are Japanese. That said, I suspect that a low-grade Nichicon, rated 85C, running in an oven for 3 years might have had a similar outcome :)
@@Tony359_2 You have a point and I suspect a PCB using jap caps like Nichicon will have a better design where the capacitor will not be running all the time but only when activate the boost function.
Anyhow, great video.
Possibly the capacitor has constant current in it and is always charged. that would have explained its short life.
Maybe. But I assume it's the worst of the worst quality, which means something like 1000 hours at 100C. If it's been running at 60C constantly, 24/7, it won't take much to die!
@@Tony359_2 Okay. I experience more and more of one thing in the country where I live. The quality control does not seem to exist and more and more shops sell AliExpress cheap type of electronics.
Tldr: zener cooked main smoothing cap end of the story
Total classic of failure
You could wire the cap with wires away from it to save it from the heat
Even a good caps are rated maybe at max to like 10k hours under 70-90c
Well, the zener is at 95C, the cap is on the other side. Let's say 50/60C all the time? If the cap is "crap" and rated 1000h at 85C, then it won't last long :)
The IC is probably a microcontroller, there's not much to say except that the analysis was very didactic. In Chinese devices, you should always be wary of electrolytic capacitors. The average lifespan is about 5 years, but when they were manufactured in the USA, Japan, European countries, etc., they lasted decades. As for resistors, I worked for a year assembling boards in the 80s. Up to 4 colors (digit, digit, multiplier, tolerance) I knew the value as soon as I saw them, for more color bands it was a mess.
I don't work with resistors very often so I always find it challenging - particularly because sometimes it's easy to read them the other way round! :)
Capacitors yes, I guess it's easy to make a crap one and then pretend they're good on the datasheet.
well, the capacitor is -40+105 degrees which is 65, so...the manufacturer was correct :) (joke)
LOL!
Most producers would tell You, that this is perfect design when device fail right after warranty period. This looks like aliexpress quality, but I saw a lot better power supply's there. If I'm not mistaken, currently EU demands, that device in standby need to draw less than 0.1W, so I guess that this constantly being hot would not fit in that description. Something shady, in my opinion, is going on here.
this has indeed failed just after warranty! :D
@@Tony359_2not necessarily. In the UK the warranty is 6 years despite what they try to tell you. You do have to prove that it's a design fault or manufacturing defect that was present when purchased though.
Source?
Well, good luck with that!
@@Tony359_2 link appears to have vanished, but easily found online. Citizens Advice explains it.
@@Tony359_2 Previous 2 replies have disappeared, you may have them in some kind of moderation queue?
It is always a crapacitor!!!
ahah always!
@@Tony359_2 99.99% of the time it is and if anything else is blown you will also find a blown crapacitor somewhere
Everyone buys in China, just few do better testing before buying. Don't ask how I now.
I probably don't want to know! :D
The device is "safe". It has double insulation and if something goes wrong I really don't think anything too bad can happen. As usual, a £20 item sells better than a £30 which is designed to last 3 times longer.
We can clearly see that you don't know anything about capacitive psu. Please don't do this kind of videos, you are guessing things...
Yes, I've just educated myself about that, I was not aware!
How do you learn things? :) How would have learnt about this type of capacitive PSUs if I hadn't tinkered with this PCB and made this video?
That's the whole point of this channel... So Tony can learn and show his findings to us... Stop gatekeeping. When people like you do things like that, no one learns, knowledge is lost and everyone ends up less enthusiastic and worse off as a result. From watching this video, I learnt plenty. Especially from the comments but yours wasn't useful at all unfortunately :(
@@Tony359_2 It is very remarkable that you reply to del4you2 the way you did! Kudos for that You got a subscribe and a like today.
Maybe it's more "I can clearly see..." than "We can clearly see...". You are not sharing my opinion...