12V 2A switching power supply - test, teardown, schematic, analysis
ฝัง
- เผยแพร่เมื่อ 25 มิ.ย. 2020
- Today, let's take a look at a nice 12V 2A switch mode power supply (SMPS) module. I identify its components, test it and show some waveforms on my oscilloscope. Then I draw the schematic and explain in detail how does it work. I also analyse its safety, disassemble the power transformer and count the numbers of turns. I talk about possible failures. I also show how the output voltage could be modified and how to turn it into an adjustable voltage power supply.
Please support this channel on Patreon:
/ diodegonewild - วิทยาศาสตร์และเทคโนโลยี
I just love these videos! I've learned so much about SMPS design from watching them. Thanks!!
Yoooo! Hello there lol
I really admire your dedication to reverse engineering, love your videos (even the unedited ones!)
The hissing test will be my standard test as well.
It does beat the 'OUCH ***' test
Thanks to this video, I managed to change the o/p voltage and current on an unused SMPS i had laying around 😊
i need more of these cause it increases my knowledge and ability to design circuits. Also the new techniques used at different phases are really inyeresting.
Keep this up and your be starting your one CE test house. We destroyed quite a few good transformers during CE testing. The test house's don't trust what a manufacture states about their transformers + some manufactures won't provide full construction details anyway. great video as usual.
Yay, a power supply analysis, hope it isn't too dodgy
Edit: Not as dodgy as it looks. Seems like even salvaged circuits are better than brand "new" chinese ones. No wonder that these are often more expensive. Another flyback you reviewed had a "2A" output and "5A 100V schottky" diode that was sizzling hot. And hot output caps. And a hot switching transistor. And a hot transformer. And the list goes on...
thats really HOT
You are right. A lot of salvaged circuits are better made than newer cheapy Chinese ones, as long as your salvaged circuit isn't a cheapy Chinese one :)
@@EsotericArctos yeah, by the looks of it, this was likely taken out of a recycled or faulty power adaptor. It's fortunate that it's cheaper for them to make new boards
@@lakiza55 Agreed. This particular one that was disassembled was quite good. But it is cheaper to by a cheap, less well isolated supply from China than to repair it.
Excellent analysis, thank you DGW!
I have the 12V version at my cotage (in shed to drive some heatsink fans), I am happy with it. It works over a year now.
I really like these ebay salvaged/repurposed switchers. you can get them in various voltages up to 24V and 1-5A current capabilities. theyre always good quality and look pretty safe. I must have used a dozen of them in various projects over the last few years. I tend to recap the output as a matter of course, and for some reason they often seem to have the remains of output wires tacked in before the final capacitor/inductor filter, or have unpopulated space like this one..
also worth watching out for is bizarre wiring schemes, Ive had a few where the cropped remains of wires used red and black for the 230V AC input, and blue and brown for the 5V DC output, ie the exact opposite of standard european wiring convention. if someone wasnt paying attention they could easily misinterpret that and connect 230V to the output. cap go bang.
if you put 5V at the output you get 230V
I miss the old heavy wall power bricks. Yes, the chunky transformer wasn't really efficient, but I have seen many broken switch-mode PSU's compared to very few chunk-former ones
Captain Diomedes - In a conventional PSU, the transformer itself has reasonable efficiency. And (assuming the transformer is a double bobbin type) they are much safer than poorly designed / poorly made (cheap) SMPSUs (switch mode power supply units).
The inefficiency comes from getting a smooth regulated and ripple free DC supply from the a.c. output of the transformer. Often a series pass transistor circuit was used, which drops the ‘unwanted’ part of the voltage fed to it to chop off the ripple of the top part of the waveform. In doing this, it converts this unwanted voltage into heat...
If a 78xx series regulator chip was used (e.g. 7805), these can now be replaced in most applications with a DC-DC switching regulator. This vastly increases the efficiency as no metal heatsink is then needed.
In a conventional PSU, the components used are not normally under and undue stresses. The voltages are all normally well below the components voltage rating. Of course there have been some where the current rating of the components was too low, or the electrolytic capacitors were not good enough, or the heatsink was too small, or there was a lack of suitable ventilation slots, and this will lead to failure. But overall, reliability is very good. In my job, conventional PSUs run continuously for many years (30 to 40 years plus). Well designed industrial SMPSU typically last 20 to 25 years. But in the consumer market, SMPSUs may only last for 2 to 10 years. Whereas a conventional PSU may last for more than 30 years. Obviously these figures are approximate and it will depend on usage, loading and the ambient temperature in which they are used.
Well, one of my other hobbies, apart from electronics, is listening on the Short-Wave radio bands. I have to make sure any device that uses a Switch-Mode supply is turned off (including LED light bulbs, monitors, phone chargers, etc). They produce tons of interference all the way up the bands into the 30MHz region and beyond. Old-fashioned linear power supplies don't have this problem. I even change my light bulb to an incandescent one when i'm using my radio as the LED ones produce interference too.
@@risvegliato - Compliant switch mode PSUs don't produce significant radiated interference in those bands.
However many PSUs don't comply with the EMC requirements. Theoretically that makes them illegal, but nobody cares about that stuff.
@@johncoops6897 They all do produce horrible harmonics, sorry. compliant or not. If you look at the SW bands using a spectrum analysis program such as Spectrum Laboratory, you will see stripes all the way up the bands. Usually spaced so many 10's or 100's of kHz apart. All these SMPS's produce harmonics of their 'chopping' frequency. The worst culprits are Flat-Screen monitors and TV's in my experience. Try setting up a long-wire and a sensitive receiver and you will see what I mean.
@@risvegliato - I know exactly what you mean. There are limits on radiated and conducted EMC that SMPS (and devices that contain them) must comply to. Unfortunately consumers don't care about EMC etc so the majority of products don't actually comply (despite CE, FCC, RCM, etc labelling. I've spent most of my working career dealing with compliance of electrical products, so it annoys me how much work I did on making products compliant, yet opposition suppliers didnt bother and nobody cared anyway.
Small token of appreciation
Long videos are not bad! It is very interesting!
Love your long educational videos! Keep it up!
Thank you, I’m so glad that I’ve found that channel, so informative. 👋 🇷🇺
excellent schematic. its very neat. good job!
I totally admire the way he talks 😊
Excellent work as always
Great job keep it up. Thanks for sharing your experience and knowledge. I love the oscilloscope, Made in USSR (sdelano v SSSR) 🙂
Make small videos with good content. You've taught me a lot please continue it...
nice analysis and review!! thanks
Yayy diodegone wild video
Very educational. Thanks.
Good review video about SMPS.
That was a nice little supplt. It was a shame to destroy such a nice transformer. Good isolation there.
Amazing power supply 👍🏼👌💪💪
Great informative video. I learned a bit more from you.
Your cat 🐈 seems to approve too.
Helpful video. I liked it
Great info and explanation! Cute cat too
thanks for your destructive testing it reveals teh truth.
You have a great voice.
It seems more common to see the class Y capacitor hooked between the lower side of the primary and secondary windings (positive HV and common on the primary/secondary) as is done above on higher power flyback supplies. Back when I was designing my own I was trying to figure out what the advantages/disadvantages are to hooking it to either the live or neutral rail on the primary side. I would assume potentially having it hooked up like above may provide a more direct return path for capacitively coupled interference back to the primary side, and also keeps the lower portion of each winding at the same RF AC potential by shunting out the stray capacitance. Not really sure.
Thank you. Very nice analysis. Nice PS it was.
Hi... Nice video 👍👍👍
Just a question... What is the inductance of the primary coil of the flyback transformer??
Thanks
The annoying white goo (which also comes in other colors) is fast-setting adhesive compound. It's to prevent components, especially heavy ones like the primary capacitor from bending back and forth at their solder joints due to vibration and physical shock so they don't eventually break off from the board.
3:33 here's the diode.!!😅
I like the way his cat accompanies him.
But that diode didn't gone wild because it wasn't too hot 😀
Very good explanation. I wish to know how can be increased output current of this 12vdc supply.
I am just one minute in the video but I can tell you this is branded for 12/1.5 power supply for netgear routers etc as I have one exactly like this. I suspect it can give 2A so thats good to know!
I love your explanations...I guess this was in a decent piece of kit originally ?
I guess the 5 amp fuse is there in the event that the MOV fails short...which I believe is a common failure mode.
I believe when the MOV absorbs spikes its voltage drops a bit... So its voltage quickly drops to nothing once it inches down to peak mains voltage. Just wanted to share my understanding of why they fail short.
The MOV and Fuse are not included in these circuits to "protect" the circuit by absorbing any voltage spikes, but rather it's there to allow the product to pass electrical safety testing.
The safety standards require that the item fails safely when subjected to a short pulse of high voltage at high amperage. The MOV conducts to clamp the mains voltage, so input current BLOWS the fuse. That test is only for "safety" and the unit does not need to be operational after that test.
This is the lowest cost way to ensure compliance with safety standards, compared to designing the whole device to cope with the high voltage test.
@@eDoc2020 - The way that MOVs work is NOT like a physical switch which closes at a certain voltage. In fact they act kind of like a Zener Diode, and there is a "knee" in the voltage to resistance curve.
So, as the mains voltage raises to very close to the "specified voltage", the MOV resistance begins to decrease and they begin to slightly conduct. This small power flow creates heat, which the MOV is unable to radiate. As the material gets hotter, it's resistance reduces further so it conducts more power which makes it even hotter. In the end they simply overheat and fail "open" (not short circuited).
This "thermal runaway" is a common failure mode for MOVs, and they burn at fairly low currents because they are so small. A 5A fuse at 230V is like 1150 Watts.... a MOV would have vapourised long before the associated 5A fuse pops.
As the MOVs age, the material inside is literally "blown away" by each Transient absorbed. It's not so much that their "voltage decreases", but more that their ability to dissipate heat is reduced... in effect their "power handling" is reduced.
Anyway, they don't "fail short" because they are specifically designed to short. They fail by over heating, and that happens at only a few Watts, not the >1000W that a 5A fuse would protect against. I hope that makes sense :-)
@@johncoops6897 That makes perfect sense. I must have picked up misinformation somewhere. I also know power strip surge suppressors can stop working without blowing the fuse, so that is definitely failing open.
@@johncoops6897 thank you John, your explaining is so clear
the kind of videos i love 😍 good luck for u
ur cat is so special and funny 🤣😁
As ever a good and interesting video!
Thank you your videos.... Can you share a video while drawing this schematic please? I am wondering how you created this scheme....
I was always wondering: How long does it take you to draw these schematics? How many tries does it usually take before you get it right? Do you have any tips for reverse engineering schematics? Thanks :)
Neat PSU, I like it too!
Nice to see your video without tesla coil
I have a question. How do you decide/calculate the number of windings in primary, secondary and aux (in the transformer)?
Can you make video on AC motors and induction motors BTW I love your content
And it would be great if you would solder the missing parts and show the difference at the output on the oscilloscope.
I was thinking about it, but the video got bloody long even without it...
@@DiodeGoneWild oh trust me we love your videos. never get bored.
@@DiodeGoneWild plz make video on induction motors
That tiny heatsink is useful and works for that package type thas is mostly plastic, if the package was the one that the rear plate is metal its usefulness is reduced a lot and is near marginal in a lot of cases where in the plastic type is useful.
Love ths Videos! I'm looking for a Motorola T64210 8647 . Do you have any suggestions?
Hi friend! I'm having issues with a power supply wich has an IC wich manages a Power Transistor like most of the cases you have shown. The problem is that the PT is ok when I measure it isolated from the board, but when is conected, it has no voltage out, so it doesn't excite the transformer (and it's auxiliary winding), and because of that the whole system can't start up. How can I measure if the IC high-frequency output signal is OK? ( I don't have an osciloscope unfortunately) I want to thank you again for sharing all these videos. I have been learning quite a lot! Best regards!
The 6 pin IC could be similar to FAN6862RTY , my guess. The "glass MOV" I believe is a - glass gas discharge tube / Surge arrester.
Your cat expects at least the same attention you always give to your electronics :-)
Would like to see an analysis of a big server power supply and a modification for variable output voltage :-)
Small but cute power supply 🙂🥰
I have a 24v/5a 120w power supply.when i checked the circuit board i found that a copper plate(same size of pcb) is connected between the output ground and heat sink at the bottom of pcb.i don't know why they used that copper plate🤔
Shielding due to interferences
So, the temperature of the power supply depends a lot on the enclosure it is in. If it is in a plastic enclosure, it will run much hotter than if it is in a larger piece of equipment. Would have liked more heat dissipation on the diodes, though. Just a little aluminum goes a long way.
Could you please make a review of anker, Samsung or baseus power supply ? That types of Power supply which has pd and quick charge 4 i know you are smart and have the knowledge for it 😊
I thought I saw the output jump to 17V with no load when you turned it on. I wonder about startup regulation.
@DiodeGoneWild I have a question. If I send you thorough pics of a ip camera power supply, could you help me troubleshoot why the voltage is to low? It's 5V/ 2000mA output and it won't power the camera because it only outputs 4.5V. Camera gives a led fault. Camera works fine with another identical supply.
Do a video with a tour in your workshop! :)
I am confused about how the chip starts up. How can it use unrectified AC to charge the 10uF cap up to starting voltage if the net DC current from an AC supply is zero? Does it start up fully within the first half-cycle of the AC waveform? Doesn't it need to complete a half cycle to get any power through the transformer to the auxiliary winding?
The chip in sot 23-6 package is probably CR6850S. Novel Low Cost Green-Power PWM Controller according the datasheet.
It could be :). or OB2263 as somebody else mentioned. Both have a matching pinout.
17:23 SOT23-6 case chip look like OB2263 (PWM control IC, Manufacturer On-Bright)
It could be... the pinout seems the same.
Can we use induction cooktop circuit as ZVS driver for flyback transformer and induction heating os tesla coil driver. It would be quite powerful
The small glass think is probably a TVS Diode
The device might be a power supply for halogen lamps. These usually provide a voltage slightly lower than 12 V in order to prolongue the bulbs life (so do iron core transformers, which usually provide 11.5V effective voltage
Do tell more because i thought halogen lights used line voltage
@@ceilingfanmusic6597 There are halogen lamps labelled GU10 or R7S which use line voltage (e.g. 230V in Europe) However, there are 12V Types eg with G4 or G6 sockets. These I mentioned...
Niccce
You got
500 View
In
Just
20 minute 😁
( that show how much people love your video )
Well the video was actually uploaded yesterday
After 6 day
He got 17k view...... 😑
My god!, a quite reasonable power supply and 2 amps seems very fair :-D
The transformer looked decent, maybe you can rewind it and have a handy little supply :-D
LOL i like your mobile power supply, not cat-astrophic in any way :)
I would modify it for lower voltages by replacing the two starter resistors with a power resistor an a liniar ragulator so it dosent need the aux winding. The aux winding would just be for protection. Id also swap out the zener with a lower one of my choosing. Or maby even put afew zeeners on a selector switch so i can choose between afew voltages and still have protection for each.
31:38 = I am a skeleton. watched all your videos very minutely.
How i can remove short circuit protection /over load protection from this 12v smps ?
Nice soviet oscilescope bro!
Apropos increasing voltage much. Do you know these used HP ProLiant server power supplies (ODM manufactured by Lite-On) that you can get for like 15€? They are usually 600W and up, and they're fully 12V, no minor rails, and they have an edge connector, and they're half-height so two of them go redundantly in a server. The PSU chassis that they go into in a server generates minor rails like 5V and 3V3 from 12V. Anyway they get thrown out after their time runs out or at first potential sign of trouble, this is how they land cheaply on eBay. Some have issues like bad PWR_GD circuit or something else, or maybe their fan failed to spin up once for a few seconds and it's been logged to management software.
Actually there's a number of ODMs, but Lite-On are the ones that don't sound like a rocket taking off, and the ones that usually land cheaply. Delta and a few others happen too, so you just look at the UL E listing number to decode the ODM.
So i was wondering, can you stretch them to 24V? Also would you be curious to just take a peek at one?
Does it have a potentiometer to regulate output voltage?
Where's the properly drawn (final version) of your sstc schematic?
He is gonna draw it in a future episode of the SSTC series.
Sir where are ur other videos on playlist
Pls update ur playlist
At last your doing something else from Tesla videos
That's might be a point. LOL
A bit rude, don't you think? Don't get me wrong, i prefer these videos myself, but i still appreciate the time and effort he puts in the tesla coil videos.
@@lakiza55 i love this channel Dude and i love the way he explains
Chill Dude!
@@lakiza55 And also i love the cat too but i have dog.
@@lakiza55 off course I also do but I was simply saying that something different happened in between
good
On next video pls 12v 10a 120w power supply .
He already made a video on such a power supply.
0:57 the cat :umm this board seems to be dodgy
Is it possible to convert the TV remote control from IR to RF ?
What is the part number of contoller IC? Could not identified? I think it's too simple design.
The website link which shows on video for lead acid battery charger don't work
Plz put all the website links in description always...... When mention on video....
Wish it was glitch due to the poor multimeter but the voltage that go to 17V at startup is pretty bad (:
Flyback controller seems like 6 Pin variant of AT3263.
Can you tell me why some power supply output shock a little despite the output voltage is 12v??
The live is connected with ground through a capacitor and the negative output is grounded. So this means that if the power supply is not grounded, it can give you a slight shock.
This is the Y capacitor at the top of DGW's schematic. The larger its value the greater the leakage current. There's also stray capacitances in the transformer. Of course, there could also be improper separation in cheaply made supplies.
❤
i also have calculator like yours but different body ...its not working means addition and substraction not working other function are working ...plz make a video on that calculator
Tenéis éste vídeo en español? Gracias por tùs vídeos!
Random combination of Google keywords suggested me OB2273. The pinout looks like it might work, and the spec switching freq is 65 KHz which is also a match, but the topcode is wrong. A clone?
thadits nyyyyyyiceee
Did any of these power supplies you test actually blow up?
Hi can you please send me a guide on how to design a transformer for adaptors
@12:43 Well, your O'scope is showing that it's not quite fully loaded. At 5uS/div, the total duty cycle looks to be about 2.6 divisions, while the waveform shows there is still about 0.8 divisions left available. Since this is at 1.8 amps, it means the power supply can provide a maximum of 1.8(2.6/1.8) = 1.8*1.44 = 2.6 amps total. It held at 2.5 and will probably hold up to 2.6 (at least until that diode overheats). But will likely become unstable at anything above that.
Your multiple power supply tear downs, reverse engineering, and schematics is teaching me everything about PSUs except a couple things I would love to learn. First is how to alter the voltage output. Not just how to change it, but all things I should consider. Voltage ratings on caps, diodes, and certain resistor wattage ratings, etc.. And not just adjusting the voltage up, but also down.
I'd love to see a couple videos on this, one for a simple flyback, and another on an ATX PSU with over-voltage protection. There's another video that shows a PSU, but he runs through it fast, doesn't discuss much, bypasses the protection circuits entirely, then abandons the project as "unsafe".
Tell me sir...transformer primary..start pin..and ..end pin..connect the mosfet..drain side..with..feed back..winding..start and..end pin...how to connect...
Wow, I've never been this early :D
Lol? How it can be like this?
@@oskarsabo What do you mean?
The video is just uploaded hours ago, and you commented here 1 day ago...
@@oskarsabo Patreon supporters get 1 day earlier access :)
You have any circuit without ic so please explain it.
At minute 21 you says that the max voltage is 240+10% (264v), but here in italy (or is just my city) the max voltage is 230+20% ( 276v)
Do you really have +/- 20% tolerance? That's quite a lot. Most 220V-240V countries have 10%. Some 100-120V countries have 15% because it's harder to keep it accurate at lower voltage where resistive drops show up more.
276V should kill old light bulbs very fast... cant imagine that
DiodeGoneWild in my electricity bill, it says 230+-20%, i have an ups that constantly monitors the voltage, usually here at my house i have 245v but it depends when you meausure it, i have actually seen 255v some times.
Dr. House uhm, never thought about that but here this is a common problem. Most of the chandeliers at my house does not have all the light bulbs because they blow up really fast and 2 light bulbs are enough, even if you are supposed to put 6 of them in the chandelier
Even in places with tighter mains voltage there can be temporary voltage spikes and surges. Equipment should be able to handle that without damage.
Nice video and I like your wet tissue thermostat, I guess it doesn't need much calibration as long as you know the pressure, you should sell the wet tissues on ebay :-)