increase in voltage on wall adapter

[KevinW]

I have this 5vdc wall adapter that works with an smps transformer and after a year or more it has increased in voltage anywhere from 3 to seven volts.
Now I'm aware that some of these require a load to operate at their normal voltage but this one tested 5 maybe 5.2 volts without a load when new and has now jumped to 14 vdc and has settled at 8 vdc without a load.
I was using it for a clock and it killed the voltage regulator when it climbed to 14 vdc.
I'm wondering what would cause such an increase in voltage?

 

[Diver300]

I don't know the exact circuit, but it sounds like the feedback device has failed.

Power supplies like that often have a transformer with a primary and two secondaries. The transformer operates in flyback mode, where during one phase of operation, the current increases in the primary and there is no current in the secondaries, as it is blocked with diodes. In the other phase of operation, there is no current in the primary and the stored energy goes into the secondaries.

The two secondary voltages are in approximate proportion to their number of turns.

On secondary feeds the main 5 V output and the other supplies the control chip on the high-voltage side.

There is a feedback, so when 5 V is reached, that is signaled through and opto-isolator, and the control chip turns off.

If there is no feedback, the control chip keeps on running and the output voltage gets too large, which is what i think happened to your power supply.

In that state, the supply to the control chip may be too large as well due to the other secondary now having a larger voltage. A lot of control chips will detect that and turn off, which is hy the voltage isn't huge.

 

[KevinW]

I compared the primary voltage to other similar devices and the voltage is pretty much the same, this device has a single secondary but I can only check the voltage beyond that point and it is pretty constant throughout that side of the circuit.
I've tested most of the components and they all seem fine so I'm wondering if the transformer may be responsible.

 

[Nigel Goodwin]

I have this 5vdc wall adapter that works with an smps transformer and after a year or more it has increased in voltage anywhere from 3 to seven volts.
Now I'm aware that some of these require a load to operate at their normal voltage but this one tested 5 maybe 5.2 volts without a load when new and has now jumped to 14 vdc and has settled at 8 vdc without a load.
I was using it for a clock and it killed the voltage regulator when it climbed to 14 vdc.
I'm wondering what would cause such an increase in voltage?

Almost certainly due to the use of cheap low quality electrolytic capacitors, which have gone high ESR - it's an incredibly common failure. Usually it's the one on the output which fails most, this causes massive ripple on the output, which feeds back to control the output voltage, resulting in the output voltage increasing - often to the stage where it damages the equipment it powers. Quite often (but not always) the capacitor can be visually seen to be faulty, bulging and leaking - replace with a decent quality one, and it will never fail again.

Either repair the PSU (change the capacitor), or bin it - using it is likely to cause damage.

 

[KevinW]

I looked at the caps and they were the right value but I forgot to look at the ESR. I'll pull them again and retest them.
I don't have any intention to reuse this adapter, just wanted to find out what caused the increase in voltage.

 

[Nigel Goodwin]

I looked at the caps and they were the right value but I forgot to look at the ESR. I'll pull them again and retest them.
I don't have any intention to reuse this adapter, just wanted to find out what caused the increase in voltage.

It's the ESR that matters, the actual value of a duff capacitor is often within specifications. If you've got a scope then checking the ripple on the capacitors will show if it's duff. In a satellite receiver etc. it's often easier to just run a scope down the outputs of the secondary rectifiers than getting the board out to use an ESR tester.

 

[MrAl]

I have this 5vdc wall adapter that works with an smps transformer and after a year or more it has increased in voltage anywhere from 3 to seven volts.
Now I'm aware that some of these require a load to operate at their normal voltage but this one tested 5 maybe 5.2 volts without a load when new and has now jumped to 14 vdc and has settled at 8 vdc without a load.
I was using it for a clock and it killed the voltage regulator when it climbed to 14 vdc.
I'm wondering what would cause such an increase in voltage?

Some power supplies are meant to increase in voltage. I am not sure what type you have, but some modern ones communicate with the load and the load tells it what voltage it can take. This means in some cases the voltage can vary from 0 volts to something like 48 volts, although most only go from 5v up to around 12v.
Some 5v wall warts will go up to maybe just 5.5v due to communication with the load, and that's quite common these days.

However, if you have a power supply that is supposed to put out say 5v and it is putting out 14v, then clearly something is wrong internally. It could be a number of things. You have to take some measurements, and it helps a lot to have a schematic. If you have a schematic consider yourself lucky because then you could almost always fix it, but if not, you can make one yourself sometimes by tracing out the circuit and trying to identify the IC chips being used. Sometimes this is not too difficult, but other times it is almost impossible. This depends on a number of things, the most important is can you get inside the package to even see the PC board. Many wall warts are ultrasonically sealed, which means you have to cut the case apart. A Dremel comes in handy here, along with a small saw blade with no teeth. This kind of saw blade makes a very clean cut by melting it's way through the plastic. If you try to use a blade with teeth you may get injured as the teeth catch on the material and the Dremel tries to jump out of your hands. I had this unfortunate experience one time and cut two of my fingers. To make one of these saw blades, you just take a regular tooth saw blade and grind off the teeth, leaving you with a smooth edge disk with a hole in the center for mounting as the original saw blade had.

 

[KevinW]

I did find a bad 500 ohm resistor so we'll see what happens when that gets changed.

 

[Nigel Goodwin]

I did find a bad 500 ohm resistor so we'll see what happens when that gets changed.

What was 'bad' about it?, did you check the ESR of the capacitors?.

Your pictures are rather small and blurred, but there appears to be two electrolytics on the output (LH) side, one or both of those is the most likely culprit.

In the absence of an ESR meter or scope, just solder a capacitor across each one in turn, and see if that cures the problem. I don't know what value they are, but for testing purposes it doesn't matter much, a good 100uF is far better than a high ESR 1000uF.

 

[KevinW]

The ESR was good on both,
1000uf is 1054uf
680uf is 740
I'm changing both out anyway.
smd 511 (510) ohm resistor reads 503k ohms.
I didn't care for the pictures myself and will retake them.

 

[Nigel Goodwin]

The ESR was good on both,
1000uf is 1054uf
680uf is 740
I'm changing both out anyway.
smd 511 (510) ohm resistor reads 503k ohms.
I didn't care for the pictures myself and will retake them.

503k ohms is obviously way out for a 510 ohm, assuming you managed to make good contact with it? (not always easy with SM components). However, 510 ohm isn't a value which tends to fail, if they do it's usually because something caused them to 'blow' and the damage if usually visually obvious (big blast mark where it used to be )

However, looking at your much sharper (but still small) pictures, it appears to simply be a discharge/load resistor across the output capacitors. Have you tried testing the supply with an external load?, it 'might' be that the 510 was there as a load to keep the supply stable when not connected to an external load - and without it (or an external load) it can't regulate?.

You could just solder a resistor (510 ohms or less) across the output capacitors, and see if that cures the problem.

As for the capacitors, they commonly are connected by a choke, which causes big problems if the first capacitor goes high ESR, but in this case it looks to be a zero ohm resistor, so both capacitors would need to fail to cause problems (either one on it's own would still work fairly well).

 

[MrAl]

The ESR was good on both,
1000uf is 1054uf
680uf is 740
I'm changing both out anyway.
smd 511 (510) ohm resistor reads 503k ohms.
I didn't care for the pictures myself and will retake them.

Hi,

You sure that resistor is suppose to be 510 Ohms and not something like 510kOhms?
I ask because that's quite a coincidence that they both have a number close to 500 in them. If you read the multiplier wrong that could show the same problem.

 

[KevinW]

Looks like 511 on the resistor.
I wired up a red led to the supply with a limiting resistor and the voltage didn't change although that may not be a sufficient load.
This thing worked well for over a year.
I still have one capacitor and this resistor to swap out and we'll see what happens.

 

[MrAl]

Looks like 511 on the resistor.
I wired up a red led to the supply with a limiting resistor and the voltage didn't change although that may not be a sufficient load.
This thing worked well for over a year.
I still have one capacitor and this resistor to swap out and we'll see what happens.

Hi,

Ok, should be interesting.