Repair Generic Power Supply that can't go beyond 10V

[muhamadamru]

Hello
First i want to state that I am noob at troubleshooting and i got a generic power supply from a friend cause  i want to start studying practically "cause i am 3rd year student at electronics " electronic circuits and start doing alone some amplifiers, oscillators, voltage regulators, and so on.   
I made a video that explain what is happening  to the unit and i will try to write a description as much as  i can explain:

The power has fixed output which is 5V/2A and another variable output 0-30V 0-5A now :

1- the output of 5V is good and it maintain itself as 5V but the current using multi-meter Uni-t U33TA It shows me that it start at 3 Amps and then Drops to 1amp ( i don't know here may the multi-meter is not that good and doesn't has the load that need 2 amps idk "

2- on the variable output even if i don't connect any load   when i try to go over 10 it goes down and i hear voices like tick tick every time it drops :
https://www.youtube.com/watch?v=bvmcmP4toBg&feature=youtu.be

I made also some photos from inside so you can see maybe you could advice me :


Thank you in advance

[BAM5]

Hey Muhamadamru,

1- Sounds like when you wanted to check the amperage you shorted the supply with your DMM. This isn't how amperage works. If you want to be sure the amperage is in spec, you need to hook up a load to the power supply, and measure the amperage that goes through that. (In order to get a load to draw 2 amps from the supply you can use Ohm's law to figure out you need something with a resistance of 2.5 Ohms. But it will also need to handle 10 watts of power.)

2- Not sure what the clicking sounds are, they could be the relays on the first board J13 & J14. It's strange to me that the voltage is jumping around like that though. Actually, perhaps the circuitry is fine and the clicking you hear are relays switching between circuits for higher and lower voltages, and the cause of the voltage jumping around is a dirty pot?

[muhamadamru]

Hey Muhamadamru,

1- Sounds like when you wanted to check the amperage you shorted the supply with your DMM. This isn't how amperage works. If you want to be sure the amperage is in spec, you need to hook up a load to the power supply, and measure the amperage that goes through that. (In order to get a load to draw 2 amps from the supply you can use Ohm's law to figure out you need something with a resistance of 2.5 Ohms. But it will also need to handle 10 watts of power.)

2- Not sure what the clicking sounds are, they could be the relays on the first board J13 & J14. It's strange to me that the voltage is jumping around like that though.

yes i think i shorted it "i am stupid" i smelled something but i think that the power has protection in it because when i connect the it goes down the value on the screen of the volt to 0

10 watt  Resistor i think  this will cost me a  fortune 

now if i shorted it what i have to do ? replace the fuse ?

But i have to mention that the problem with the voltage drop and can't go over 10 and bounce before i do any ampere test

Also i have to mention that now when i read the voltage with volt-meter it reads exactly as it's on the screen +-1~2%

[BAM5]

I added a bit to my first reply.

Alright as I stated before, you can't do an amperage test without a resistive load. What you're doing when you put the meter on the contacts in amp measuring mode is shorting them out which is not what you should be doing. I recommend you go to a site like khan academy and look up a highschool level physics class that explains the properties and laws electricity.

[RJFreeman]

yes i think i shorted it "i am stupid" i smelled something but i think that the power has protection in it because when i connect the it goes down the value on the screen of the volt to 0

I would expect a bench supply like that to have current limiting built in, and you usually set the current limit by shorting out the supply and adjust the current control until you get the current you want.

the clicking sounds like 'tap switching relays' - these select different taps on the power transformer, based on the output voltage of the supply, this reduces the power dissipation in the regulator, so for example if the output voltage is, say 5 Volts (either because this is the voltage you have selected, or because the Constant current/current limit mode has kicked in) then rather than powering the regulator off the 30 Volt tap, the tap selector would switch down to a 12 Volt or maybe a 6Volt tap (depends on the supply).

so it might be that a higher voltage tap on the supply is open circuit, and so when the supply tries switching up it fails?

do you have any brand name or model number on teh supply?

[muhamadamru]

I added a bit to my first reply.

Alright as I stated before, you can't do an amperage test without a resistive load. What you're doing when you put the meter on the contacts in amp measuring mode is shorting them out which is not what you should be doing. I recommend you go to a site like khan academy and look up a highschool level physics class that explains the properties and laws electricity.

Ofc i understood that this is not the the way that one could  measure the amp,  and we need a load  but i mean  mistake have been done and now if i have to fix it what i have to do ?  I mean that  i shorted the supply this means that i have to replace the fuse ?

Or if it now gives me correct reading for the volts it's ok and nothing happened to him "despite the old problem with flickering output "

I attached for the 5V output and for the variable output :

[RJFreeman]

Viewing it again, the supply does get to 30 Volts, and then the tap switcher kicks in, or drops out which seems to be when the supply drops back to 10 volts.
This would suggest that the comparator for the tap switcher is playing up. Although it seems to be intermittent?

the tap switcher usually uses a comparator to compare the output voltage with an internal reference voltage and uses that to decide when to switch taps.

The Tap switcher will be independent of the voltage regulator as it needs to reduce the supply when the current limit kicks in.

[RJFreeman]

In the first 13 seconds of the video, it seems like the voltage slowly drops (which would be the Filter caps discharging) then the click would be the tap selector kicking in? then the voltage jumps up, suggesting a higher tap was momentarily selected and the capacitors charged up?

the voltage drop at 1:09 is due to the current limit kicking in (supplies might do this if you select the minimum current)  - this is indicated by the CV or Constant Voltage LED going out and the CC or Constant Current LED lighting up.

So yes this supply is deigned to be shorted out (at least in theory) and as I said before that is how you set the current limit.

[muhamadamru]

Viewing it again, the supply does get to 30 Volts, and then the tap switcher kicks in, or drops out which seems to be when the supply drops back to 10 volts.
This would suggest that the comparator for the tap switcher is playing up. Although it seems to be intermittent?

the tap switcher usually uses a comparator to compare the output voltage with an internal reference voltage and uses that to decide when to switch taps.

The Tap switcher will be independent of the voltage regulator as it needs to reduce the supply when the current limit kicks in.

the Tap switcher is fed interdependently  from the regulator
Where can i find the  tap switcher in the PCB and do i have to replace it
Really this is the first time i hear about such new stuff     

[muhamadamru]

In the first 13 seconds of the video, it seems like the voltage slowly drops (which would be the Filter caps discharging) then the click would be the tap selector kicking in? then the voltage jumps up, suggesting a higher tap was momentarily selected and the capacitors charged up?

the voltage drop at 1:09 is due to the current limit kicking in (supplies might do this if you select the minimum current)  - this is indicated by the CV or Constant Voltage LED going out and the CC or Constant Current LED lighting up.

So yes this supply is deigned to be shorted out (at least in theory) and as I said before that is how you set the current limit.

i will make longer video for better analyzing

[muhamadamru]

In the first 13 seconds of the video, it seems like the voltage slowly drops (which would be the Filter caps discharging) then the click would be the tap selector kicking in? then the voltage jumps up, suggesting a higher tap was momentarily selected and the capacitors charged up?

the voltage drop at 1:09 is due to the current limit kicking in (supplies might do this if you select the minimum current)  - this is indicated by the CV or Constant Voltage LED going out and the CC or Constant Current LED lighting up.

So yes this supply is deigned to be shorted out (at least in theory) and as I said before that is how you set the current limit.

if i want to set the current limit i have to shot the -  + outputs and change the current knob up to a chosen value ?

[RJFreeman]

he Tap switcher is fed interdependently  from the regulator
Where can i find the  tap switcher in the PCB and do i have to replace it

The tap switcher will be circuitry around a Relay on the PCB. without a schematic I could not comment further as it might be an spare op-amp in a package with other parts of the circuit or could be a standalone comparator.
This is where any details on the supply might be useful,

Really this is the first time i hear about such new stuff

they are common in variable Linear power supplies

[RJFreeman]

if i want to set the current limit i have to shot the -  + outputs and change the current knob up to a chosen value ?

Correct, that's usually how you set current limit on these types of supplies.

[ez24]

this is what I see when I look at your pictures - never seen this happen before on this forum.  So my guess you are doing something "funny"

[RJFreeman]

i will make longer video for better analyzing

probably not much point, the thing now would be to find a schematic and start taking measurements.

[crispy_tofu]

this is what I see when I look at your pictures - never seen this happen before on this forum.  So my guess you are doing something "funny"

Yeah, the pictures are acting up here as well. Can you please update them muhamadamru? 

[muhamadamru]

this is what I see when I look at your pictures - never seen this happen before on this forum.  So my guess you are doing something "funny"

Yeah, the pictures are acting up here as well. Can you please update them muhamadamru? 

Yes i will

[muhamadamru]

this is what I see when I look at your pictures - never seen this happen before on this forum.  So my guess you are doing something "funny"

you can't see any of the pictures ?

[RJFreeman]

you can't see any of the pictures ?

I don't think any of us can

[madires]

Your PSU looks like one of those Mastech clones. Please search for HY3003 or HY3005 schematics. The basic design is mostly the same, just minor changes for the different current ratings.

[muhamadamru]

I don't think any of us can

Now i uploaded them as attachment are they visible ?

[Len]

Now i uploaded them as attachment are they visible ?

Yes, thanks.

[tooki]

Alright as I stated before, you can't do an amperage test without a resistive load. What you're doing when you put the meter on the contacts in amp measuring mode is shorting them out which is not what you should be doing.

How is it wrong? With that type of bench supply, you must short the output to set a current limit, as others have said.

Besides, how is the resistance of the test leads and the internal shunt resistor in the meter not a resistive load? (It's a low resistance to be sure, but it is one nonetheless.)

[muhamadamru]

Today i wanted to do  a test to know the forward voltage for an LED though i connected a 1k resistor in series and  i also set the current limit to 0.8 A and try to test 
the power keeeeeps go down and up down and up and switch between taps

[jwm_]

Today i wanted to do  a test to know the forward voltage for an LED though i connected a 1k resistor in series and  i also set the current limit to 0.8 A and try to test 
the power keeeeeps go down and up down and up and switch between taps

Since you have current limiting in the supply, you don't need the resistor to limit current for the LED (assuming you set the current limit properly.)

To test it, set the voltage to something higher than the LED voltage, doesn't matter much, but 5V should do. Then turn the current all the way down to zero and attach the led directly without a resistor.

The voltage will be low (perhaps even zero) because the current limiting has kicked in (which is what you want). slowly turn up the current dial, and you should see a glow around 3ma for a modern LED and the voltage will be roughly at the forward drop voltage of the LED. you can then ramp up the current slowly and see how the voltage affects it. You should be able to go up to 25milliamps with any modern LED without damaging it and it should be nice and bright, but if you are willing to sacrifice the led, feel free to turn it up slowly even more and watch the failure mode.

[OldSchoolTechCorner]

Could be OP-AMP in current, or voltage control, or bad output transistor, which usually cause that issue. Without a schematic and measurements it a guess at best. Do you have a scope, or meter around to troubleshoot it? As you need to check taps off transformer, output transistors, relay section, big power resistors and OP-AMPS input and output. as for CV, you monitor the output voltage and feed this back into the inverting input of the op-amp driving the output, for CC, you monitor the voltage across a current sense resistor and feed this back into the same inverting input of the op-amp, is usually how it works as the circuit will automatically switch into CV or CV depending on which threshold is reached first. If current op-amp isn't working correctly, you won't have current control and voltage will drop at random, or no output at all.

First I would tap the board and make sure no cold solder joints and look for any obvious signs of damage components. Next step to take is to check transformer taps, relay section, big resistors, output transistors section and make sure no issues as shorted, or open transistors bad diodes, or resistors, ESR on capacitors and also getting correct voltages with the power section first and work your way back. Also check for excessive ripple. When you get to OP-AMP's assuming you check everything else. One trick is to remove the current OP-AMP and that will keep power supply from going into constant current mode and if you get stable voltage then current OP-AMP, or potentiometer is bad. Just make sure you don't draw more current then the PSU rated for, as you will blow the outputs, as you will have no current protection till OP-AMP is replaced.

To set current limit set power supply voltage to around 5 volts DC and short the + and - together to set current limit, then turn current pot to when the display read 25ma, so you don't blow LED, or put to little current in.

I watch the video, one thing I would like you to try is to short the + and - set voltage to 5 volts and turn the current knob and then tell us if it has a affect on current display, if current goes up, or down when you turn the knob? Plus if you can turn current knob down all the way on both coarse and fine and if CC light come on? One warning if current shoot up with no control pull cable, or turn off power supply. If OP-AMP is bad will cause lost of current control and issue with voltage dropping and output cutting off and on.

[OldSchoolTechCorner]

Watching your video do suspect current control OP-AMP is bad, or bad potentiometer, as notice CC light flashing when the relay clicks on, or off as it going in and out of constant current mode and did see you have current pot turn a little bit. Check the front panel potentiometer for current control and make sure they are not cracked, or broken, or dirty.

[muhamadamru]

Could be OP-AMP in current, or voltage control, or bad output transistor, which usually cause that issue. Without a schematic and measurements it a guess at best. Do you have a scope, or meter around to troubleshoot it? As you need to check taps off transformer, output transistors, relay section, big power resistors and OP-AMPS input and output. as for CV, you monitor the output voltage and feed this back into the inverting input of the op-amp driving the output, for CC, you monitor the voltage across a current sense resistor and feed this back into the same inverting input of the op-amp, is usually how it works as the circuit will automatically switch into CV or CV depending on which threshold is reached first. If current op-amp isn't working correctly, you won't have current control and voltage will drop at random, or no output at all.

First I would tap the board and make sure no cold solder joints and look for any obvious signs of damage components. Next step to take is to check transformer taps, relay section, big resistors, output transistors section and make sure no issues as shorted, or open transistors bad diodes, or resistors, ESR on capacitors and also getting correct voltages with the power section first and work your way back. Also check for excessive ripple. When you get to OP-AMP's assuming you check everything else. One trick is to remove the current OP-AMP and that will keep power supply from going into constant current mode and if you get stable voltage then current OP-AMP, or potentiometer is bad. Just make sure you don't draw more current then the PSU rated for, as you will blow the outputs, as you will have no current protection till OP-AMP is replaced.

To set current limit set power supply voltage to around 5 volts DC and short the + and - together to set current limit, then turn current pot to when the display read 25ma, so you don't blow LED, or put to little current in.

I watch the video, one thing I would like you to try is to short the + and - set voltage to 5 volts and turn the current knob and then tell us if it has a affect on current display, if current goes up, or down when you turn the knob? Plus if you can turn current knob down all the way on both coarse and fine and if CC light come on? One warning if current shoot up with no control pull cable, or turn off power supply. If OP-AMP is bad will cause lost of current control and issue with voltage dropping and output cutting off and on.

I post images for the PCB and the internal component at the first page for me i didn't see any obvious damaged component  and for Cold solder joints idk  i think all of them are ok but i will recheck them.

Next step to take is to check transformer taps "Where can i locate them  and the relay section and the big power resistor on the PCB ? "

Where i can find OP-AMPS input and output and how the readings must be ?

how to check the output transistor ?

For sure i have meter
When i turn   current knob down all the way on both coarse and fine and  i don't short + - the CC light is Off but but when i short  + - it will come on.

i thought i would be somewhat easier but i am very noob to know those techniques

[matseng]

Since you have current limiting in the supply, you don't need the resistor to limit current for the LED (assuming you set the current limit properly.)

To test it, set the voltage to something higher than the LED voltage, doesn't matter much, but 5V should do. Then turn the current all the way down to zero and attach the led directly without a resistor.

The voltage will be low (perhaps even zero) because the current limiting has kicked in (which is what you want). slowly turn up the current dial, and you should see a glow around 3ma for a modern LED and the voltage will be roughly at the forward drop voltage of the LED. you can then ramp up the current slowly and see how the voltage affects it. You should be able to go up to 25milliamps with any modern LED without damaging it and it should be nice and bright, but if you are willing to sacrifice the led, feel free to turn it up slowly even more and watch the failure mode.

The current limiting on my cheap psu's kicks in so slowly that they will happily destroy both standard as well as 1 & 3 watt LEDs if I hook them up wile the power is on (having already set the current limit(s) to reasonable values previously).

[mikerj]

The current limiting on my cheap psu's kicks in so slowly that they will happily destroy both standard as well as 1 & 3 watt LEDs if I hook them up wile the power is on (having already set the current limit(s) to reasonable values previously).

Even on expensive PSUs it's unwise to connect such devices whilst the output is enabled,  I can clearly remember a technician at work blowing up a load of expensive 980nm pump lasers by doing this with an HP 6630 series.  The cheap PSUs are worse though since they very often have a fair size electrolytic cap wires straight across the output terminals.

[OldSchoolTechCorner]

It not the OP-AMP circuit, as I stated it could be earlier, as current control seem to work and does go into CC and CV modes fine and you have current control, as you shown in third new video on the current meter when shorting the leads together and setting voltage to 5 volts. More likely issue seems to be in relay section, or output section.

[OldSchoolTechCorner]

Since you have current limiting in the supply, you don't need the resistor to limit current for the LED (assuming you set the current limit properly.)

To test it, set the voltage to something higher than the LED voltage, doesn't matter much, but 5V should do. Then turn the current all the way down to zero and attach the led directly without a resistor.

The voltage will be low (perhaps even zero) because the current limiting has kicked in (which is what you want). slowly turn up the current dial, and you should see a glow around 3ma for a modern LED and the voltage will be roughly at the forward drop voltage of the LED. you can then ramp up the current slowly and see how the voltage affects it. You should be able to go up to 25milliamps with any modern LED without damaging it and it should be nice and bright, but if you are willing to sacrifice the led, feel free to turn it up slowly even more and watch the failure mode.

The current limiting on my cheap psu's kicks in so slowly that they will happily destroy both standard as well as 1 & 3 watt LEDs if I hook them up wile the power is on (having already set the current limit(s) to reasonable values previously).

Because you get a spike as it falls in and out of regulation, when you first turn on, or off the power supply and charge store in capacitors. Reason why quality power supplies you have button to enable output manually, once power supply is turned on. Use power supply in CC mode and set current limit, as it removes the possibility of transient inrush current protecting the LED, or any other sensitive component from damage. The inrush current will make for a bad day and destroy some sensitive components with no issue.

[RJFreeman]

Hang on, sequence is:
start at 19V
voltage drops off (capacitor discharge)
Voltage reaches around 9.8V
"Click" relay (tap changer?) kicks in
Voltage goes back to 19V
sequence repeats

now here's the thing, I would expect to hear relay kick in and kick out again (two clicks)
So the selector monitors the output voltage (and is usually independent of the regulator) so as teh voltage drops, I would expect it to drop out the higher voltage tap at probably 9.8Volts, which would mean the voltage wouldn't rise again until it went above maybe 10-11 Volts (which would presume the tap to be at 12V or so)

It is almost as though the tap selector is wired arse about, and so as the voltage drops, instead of dropping out the higher voltage tap, it kicks in the higher voltage tap, Then the output voltage has increased so the tap selector changes it mind and drops out immediately

So it all merges into the one click? (not sure, but think I can hear a slight echo?)

either way you need too open it up, check the taps from the transformer (measure the output voltage) and see what the relay is doing (is it kicking and dropping out immediately?)  then you might need to identify the drive circuitry for the relay (it will most likely have a transistor and channel on an op-amp comparing the output voltage with a reference voltage).
Unfortunately no one here can do this for you, and without a schematic we can't even suggest what to look at.

And to have any chance of finding a schematic, we need any detail (model numbers, brand names etc) of the power supply, there may be some  info on circuit boards inside....