Switch Mode Power Supply (SMPS) Repair - Low output voltages whilst under load.

Thread Starter

MC22

Joined Sep 12, 2022
3
Hi,

I’d like help please troubleshooting some circuitry. I have some knowledge of electronics repair but are eager to learn more!

I have an audio mixer with a faulty switch mode power supply. When switching on the desk the buttons will flash and the screen lights up, but the unit does not boot. Switching the power on/off a few times will eventually get the mixer to boot up but not operate fully. Only half of the boards would light up and respond.

I have since went ahead with the obvious fix and replaced all 21 of the Chinese branded electrolytic capacitors with comparable Panasonic and Nichicon caps purchased from Digi-Key.

No change! Still the same non-working status.

The output voltages under *no load* are close to the specification, but when the 7 control boards are connected the voltages drop like a stone! Connecting up a single PCB on the output has the same affect. I have included the values measured below – in order.


Manufacturers Specification:
+48v
+16v
-16v
+6.6v
+6v

*No load*:
+46.2v
+15.1v
-15.1v
+6.6v
+5.0v

*Load connected (1 PCB)*:
+8.75v
+2.4v
-2.4v
+2.2v
+0.9V

My thoughts now are, could there be something failing in the IC feedback circuitry to increase the voltage when required? This would this cause all the voltages to be affected.

Am I right in thinking? How would I go about checking this?

I have added attachments to help diagnose.

Kind Regards,
Matt
 

Attachments

Dodgydave

Joined Jun 22, 2012
11,285
No that sounds like overcurrent sense , there must be more leaking current in the circuit which is sensed by pin3 on the switchmode chip (UC3844) .

It could be bad capacitors on the output side or output diodes.
 

Ian0

Joined Aug 7, 2020
9,671
Check that the value of R6 is really what it should be.
If it had previously suffered some fault which resulted in damaged components, then R6 could have been overloaded and its value may have changed. If it is now higher that it should be, the supply will current limit at too low a value.
 

Thread Starter

MC22

Joined Sep 12, 2022
3
Thank you to everyone who replied. Apologises for the late response. I am back to work again on the power supply!

I will check out the components suggested by renemonte and Ian0.

No that sounds like overcurrent sense , there must be more leaking current in the circuit which is sensed by pin3 on the switchmode chip (UC3844) .

It could be bad capacitors on the output side or output diodes.
I have replaced all of the electrolytic capacitors so they shouldn't be a problem.

If I test the output voltages individually with a load, would that identifiy which output diode (if any) is at fault?
 

Dodgydave

Joined Jun 22, 2012
11,285
Thank you to everyone who replied. Apologises for the late response. I am back to work again on the power supply!

I will check out the components suggested by renemonte and Ian0.



I have replaced all of the electrolytic capacitors so they shouldn't be a problem.

If I test the output voltages individually with a load, would that identifiy which output diode (if any) is at fault?
Remove the diode from the circuit and test it with a multimeter, and your method will work as well.
 

Thread Starter

MC22

Joined Sep 12, 2022
3
Is there any reduction in the main power input rectified voltage across C7 when the psu is loaded?
I have taken some measurements and found there is a drop in voltage when under load.

No load: 245 V

With load: 221 V


What does this indicate is at fault?
 

Dodgydave

Joined Jun 22, 2012
11,285
I have taken some measurements and found there is a drop in voltage when under load.

No load: 245 V

With load: 221 V


What does this indicate is at fault?
Check the bridge rectifier D1 and the thermal resistor R3, . Also check diodes D12 & D13.
 

Hymie

Joined Mar 30, 2018
1,277
Looking at the circuit diagram (which is quite complex), if the transistor of opto IC2B is switched on, pin 1 (comp) of the primary switching IC will be grounded (along with the circuit to the left of the IC), which will most likely shutdown the PSU.

IC2A, the opto diode is driven/contyrolled by the voltage 15V/1, however I am unable locate this on the circuit diagram. The value may be in error and should be 16V/1 which is on the 16V rail out before fuse F4.

If this 16V1 is not within spec, due to an overload etc, then this may cause the PSU to shut down – therefore measure this voltage with the other output voltages loaded, and then with the 16V rail loaded and see what happens.
 

The Cleric

Joined Dec 5, 2023
4
With no load here's what I'm seeing:

+42.2V
+14.7V
-14.7V
+5.1V
+5.1V

With load (attaching it back into the sound board):

+44.3V first measurement, but 58.3V when I double checked
+15.1V
-15.1V
+5.0V
+5.0V

To be very clear I have a slightly different version of the one in the PDF, but that PDF is the only one they publish (this sound board comes in two versions, a traditional one and a rack mount, I have the rack mount whereas all their documentation only ever refers to or describes the traditional, from what I can tell they are MOSTLY the same).

I will attach some pictures of my PSU from my phone in a separate message.
 

The Cleric

Joined Dec 5, 2023
4
As well in my case I replaced all the caps and that did allow it to boot, but no inputs or outputs for the sound board seem to work. I tried tracing every connection I could in one of the input channels and noticed that the input disappeared at an amplifier chip that needs the +48V, so I’m under the assumption that the PSU is still to blame though I’m very new to electronics and could be very wrong.
 

R!f@@

Joined Apr 2, 2009
9,918
I have taken some measurements and found there is a drop in voltage when under load.

No load: 245 V

With load: 221 V


What does this indicate is at fault?
I would check the filter cap for proper capacitance in this case
 

R!f@@

Joined Apr 2, 2009
9,918
I also would measure the AC voltage across the cap under load.
This would tell me the ripple if there is. Of course you need a good DMM fro this.
If the AC is more than 0.5 or so ( my assumptions ) then the filter cap needs to be replace.
You will get a much higher AC value if the cap is bad.
 
Top