AC/DC power supply with 19.5V 7.9A issue

[spark45]

Hi!

I have an AC/DC power supply with 19.5V 7.9A output from a desktop computer (a kind of All in One). Apparently it is a source similar to that of a laptop charger, same look.

For some time now, despite feeding the equipment well and working without cuts, if I disconnect it from the AC plug, I can no longer use it in hours or days... If I reconnect it, it doesn't work and then the LED that has does not light up, and you have to leave it, as I said, disconnected to use it again.

I have no advanced knowledge of electricity, I have disassembled it and apparently everything is correct, there are no inflated or burst capacitors.

Can you help me?

Thanks in advance.

 

[Sharpend]

Is this the psu for the computer or something else?
Perhaps suggest looking for a replacement psu?

 

[Richard-the-Ninth]

Spark45, as you have no understanding of electronics, and no way of testing anything, it would be cheaper / easier to buy a new one. I would look on Amazon or ebay

 

[spark45]

Spark45, as you have no understanding of electronics, and no way of testing anything, it would be cheaper / easier to buy a new one. I would look on Amazon or ebay

Thanks, but I can solder/desoldering components. I see a big capacitor into the power suply, (3)7 DU, it's a brown big one. I see some videos on YouTube about similar power supply. I have no more info about this capacitor, can you help me?

 

[Richard-the-Ninth]

I have no more info about this capacitor, can you help me?

Changing random components is never a good idea. You need to establish which component has failed and replace it with a working one of the same value.

Stating the colour of a component if of little use. Here is a blue capacitor, unless it is proven that this capacitor is at fault what would be the point of changing it, and just because it is blue does not state what its value is. As I have already said, it would be easier /cheaper to buy a complete new power supply.

 

[spark45]

Changing random components is never a good idea. You need to establish which component has failed and replace it with a working one of the same value.

Stating the colour of a component if of little use. View attachment 14865Here is a blue capacitor, unless it is proven that this capacitor is at fault what would be the point of changing it, and just because it is blue does not state what its value is. As I have already said, it would be easier /cheaper to buy a complete new power supply.

Ok.

And regarding the symptoms, couldn't you tell me anything approximate? Why does it stop working hours or days after unplugging it from the electrical line, and then it starts working again? It's very weird

 

[Richard-the-Ninth]

Something overheating?
Buy a replacement, problem solved.

 

[spark45]

Something overheating?
Buy a replacement, problem solved.

Nobody say overheating. All is explained on first post.

 

[binky]

Ok.

And regarding the symptoms, couldn't you tell me anything approximate? Why does it stop working hours or days after unplugging it from the electrical line, and then it starts working again? It's very weird

Sounds like a failing connection within the power supply such as a dry joint. Could possibly be the capacitor, but even if you did manage to replace that, the nature of the way it's failing tends to produce power spikes that can cause damage to other parts of the power supply. It really is easier and probably safer to buy a new one.

 

[SPECIAL LOCATION]

Ok.

And regarding the symptoms, couldn't you tell me anything approximate? Why does it stop working hours or days after unplugging it from the electrical line, and then it starts working again? It's very weird

Welcome to the forum..
I admire your eagerness to try and fix this yourself to save waste etc..

But from the information you have given its almost impossible to guess the cause....
It could be numerous possibilities...
Damaged input lead...
Failed component(s)..
Failed joint on PCB..
Failed output lead...
etc..

If you could get hold of manufactures data sheets/component lists/wiring diagrams.. etc... etc..
And you have access to some suitable test equipment..

You may stand a chance of sussing out a probable cause....
that you may then be able to evaluate if its worth trying to repair...

But as you say you have limited electrical knowledge....
Although you can solder..
You really do need a comprehensive understanding of identifying various components..
and the ability to read what their products numbers/values are...

If it was me...
I would probably just start googling an equivalent power supply..?
Buy it, and save time and hassle..
Unless you actually do PSU repairs as your day-job and have the kit readily available..

Do you have any computer repair shops locally to you..
advertising PC, Laptop, MacBook repairs upgrades etc..
(We've probably got 3 or 4 within 2 or 3 miles.. )

It could be worth asking locally if any will repair your PSU..
Or do they have an equivalent new or refurbished item you could buy?

 

[johnb2713]

Hi!

I have an AC/DC power supply with 19.5V 7.9A output from a desktop computer (a kind of All in One). Apparently it is a source similar to that of a laptop charger, same look.

For some time now, despite feeding the equipment well and working without cuts, if I disconnect it from the AC plug, I can no longer use it in hours or days... If I reconnect it, it doesn't work and then the LED that has does not light up, and you have to leave it, as I said, disconnected to use it again.

I have no advanced knowledge of electricity, I have disassembled it and apparently everything is correct, there are no inflated or burst capacitors.

Can you help me?

Thanks in advance.

Hi,
I have a computer retail shop and we get customers bringing PC's in with this sort of issue on a regular basis. It really is not worth the time and effort to fault find and repair these power supplies. Ebay has loads of them for sale, identify the voltage and the current or power, search ebay find the one with the right tip, job done. Steer clear of the ultra cheap ones and you'll be fine. It's likely we will have some used ones at a good price if it helps.

 

[Geoff1946]

Sounds like a failing connection within the power supply such as a dry joint.

Agreed. The majority of electrical faults, especially intermittent ones, are really mechanical failure.
Expansion and contraction of metal and other materials, as equipment warms up and cools down causes stresses and eventual microscopic breakage. This can be internal to components or where they are soldered to the printed circuit board.
It is extremely difficult, even if you have some understanding of electronics, to trace such faults without a circuit diagram and test equipment. Further, switching power supplies are especially difficult because they have protection circuits which monitor for over-voltage/ over-current/ over-temperature, etc, and shut the circuit down.
Returning to your hope of finding the problem the only thing I can think of to try is to establish if it is temperature related. Warm it gently with a hair dryer. If it then starts working try cooling individual components or joints with a freezer spray. You may just strike lucky but I wouldn't advise you to put your wages on it!
Above all, be careful ! Even if it isn't working there is mains voltage at one end of the board!

 

[Andy1733]

Nobody say overheating. All is explained on first post.

Manners cost nothing

 

[runningpiglet]

The fault is likely in the startup section, once running the supply controller powers itself happily from its own output but for the first few milliseconds uses a dropper resistor from the mains. It is not a rare problem and I seen several cases . It can be as simple as a resistor gone open circuit. However the difficulty is finding it! Really the simplest solution is to get a new one.

 

[spark45]

Give me some days. I will change 450v 100uf x1 (the big brown : ) ), 25V 220uf x2 and 25V 1000uf x2 capacitors

 

[user 30718]

These are common on dell laptops (pre usb c).
Recent dell units are 130w. Older ones are 180 and 200w.

It is a switch mode supply (smps) so expect over 350v on that 450v cap (nasty. Ask any one who worked on valve amplifiers)

The guys on here often suffer 'compassion fatique' and if you can ignore the 'I told you once' arrogance, there is useful house wiring info to be had.

Electricians, like everyone, do not know 'what they don't know'. Low level concepts can escape them. To conceal this, they may be defensive. Do not be put off. While random part swapping may not be the best use of time, it can be a great way to learn.

Sadly, buying a new unit is usually cheaper if you are paying today's inflated 'engineer' prices.

Lookup swps on electronics or amateur radio websites. The front end (mains diodes, chokes and caps) are easy to test. The output side is a lot harder.

Have fun, do not be put off, be carefull of the absolutely lethal voltages. Remember that an inquisitive mind is far superior to a closed one. But if you don't understand why the DC on the input side is so dangerous - please read up on it.

Good luck.

 

[Sharpend]

Whilst I take on board and often encourage people that the best way to understand is to learn by trial and error, I don’t necessarily advocate such principles on a public forum for the very reason that you give, unknown lethal voltages being present. A little knowledge and dangerous voltages as you say can be fatal.

 

[johnb2713]

Sadly, buying a new unit is usually cheaper if you are paying today's inflated 'engineer' prices.

So you work for £1.50 per hour do you?

Lookup swps on electronics or amateur radio websites. The front end (mains diodes, chokes and caps) are easy to test.

But unlikely to cause this kind of fault.

 

[user 30718]

So you work for £1.50 per hour do you?


But unlikely to cause this kind of fault.

Spikey! All prices are inflated to people of our age, be it a car wash or a boiler install. I don't work for £1.50, I bearly work at all. Bit I guess this is more about you and how you justify your charges (to yourself)

 

[user 30718]

So you work for £1.50 per hour do you?


But unlikely to cause this kind of fault.

Indeed. A failed diode was not uncommon in early Sony stuff. Quite dramatic too. The real point of mentioning it is to give an easily understandable example of the hard ware. Hoping to build up readers confidence and encourage personal learning. Kind of opposite to many of the superior toned posts