Repair of a switched power supply

[brue]

Hi there..

Im currently trying to repair a switched power supply. Its from an Canon MP610 Printer.

The printer dont want to work anymore completely. I checked the power supply and measured no output voltages.

I checked the 3 fuses, all okay. I have the 318V DC on the big primary cap. I measured between Gate and Source of the MOSFET, no voltage. 318V between Drain and Source on the MOSFET, okay.. MOSFET: K3562
Its a little driving IC on the PCB, labeled with 2A07D88 / 5518. I cant find any datasheets on that device.

The transformer has 6 connections primary and 4 on the secondary side. Can anyone say how the connections internally works? Cant find any datasheets on that device, too..

On the Output connector is an connector that trace is going to the Q5 device, i presume a transistor (Labeled with "XN" or "NX") Any ideas for what that connector is?


Thanks for any help, dont wanna throw the thing away..

[Gall]

Can you draw the schematics?

I suspect the driving IC has some drop-in replacement. Most likely the IC is broken.

[SeanB]

Check R2, it probably is open circuit or dry jointed. It is the startup supply to the controller.

NOTE - discharge C2 first...........

The transformer has 2 secondary windings and 2 primary ones. The secondary ones are used to make the 2 output rails. The primary has one driven by the power switch mosfet, the other is for both powering the chip after it starts up and for sensing when to switch off the output device when the core saturates.

Q5 is used to switch the unit to standby, where it produces low output voltages, but still runs. Normally then the 30v rasil drops to 10V or so

[dr_p]

  I almost spilled my drink. Yesterday I started repairing a SMPS of a Canon MP830. The switch mode controller is a Fuji Electric FA5518 and the MOSFET is a Toshiba 2SK3562.

Symptoms are:
-fuses intact,
-semiconductors intact(as far as i can measure while offline),
-HV startup present on IC pin 8,
-primary capacitors not dried;
-10V present on Vcc of IC
-constant  (?!) 10V on output pin, BUT
-no voltage on MOSFET gate, thus closed.

to be continued

I am currently reading this tutorial into repairing a SMPS, which I find to be great.

[SeanB]

Change those primary capacitors, 10V is not enough to start up with, it needs about 15v before it will try starting.

[kripton2035]

yep not enought voltage on the ic. it's like the uc384x, the capacitor that's on the vsupply is usually dead, no capacity or dry or both.
check with en esr meter if you have, or better change it
it's often a small capacity like 10uF. bad cap leds to no oscillation and no mosfet commutation.

[dr_p]

I replaced it with a 47uF and now Vcc is 12.8V. Scoped it and it's stable.
First screen is Vcc, second is OUTPUT on chip, third is voltage on gate of MOSFET. It seems like it's trying to start, but quits. 

I have 5W/ 250 ohm dummy load on both outputs  (should be 24V and 30V). They are warm. Outputs were scoped are stable at 8V and 12V.

Optocoupler LED gets 1.25V and the photodiode is always on (0.6V drop). So IC Feedback pin voltage is 12V.

edit: I'm not thread hijacking, but I think the OP and I have the same problem, so we might as well help each other.

[brue]

Hi

I've checked now the power supply on Pin 6 of my controller, got 25.1V. So i think this should be okay in my case.

The 2 small primary electrolytic caps on my supply are 1uF both. They look okay. I check they tomorrow with the ESR meter. Problem is, havent got 100V electrolytic caps here.

Checked R2, its fine. 318V on Pin 8 of the controller.


thanks everyone

[dr_p]

I just realized the MOSFET should get 10V PWM on gate-source. It's getting 4V PWM. (frequency depends on the load current, and that seems to by like described in the controller's datasheet)

I already posted the waveforms, but now I've drawn a hand schematic. Mind you, the 4V PWM is between gate and ground, but I also scoped between gate and source and it's 3.86V, since there isn't much drop on the power resitor (0.22 ohm).

Now, I'm thinking the diode does nothing (for now) since it's reverse biased. So I have a 6V drop on the 820 ohm resitor. that translates into 7mA that the MOSFET sinks through it's gate. That isn't normal, is it?

When I get home from work I will try replacing the transistor with a similar one.

Hi

I've checked now the power supply on Pin 6 of my controller, got 25.1V. So i think this should be okay in my case.

The 2 small primary electrolytic caps on my supply are 1uF both. They look okay. I check they tomorrow with the ESR meter. Problem is, havent got 100V electrolytic caps here.

Checked R2, its fine. 318V on Pin 8 of the controller.

I think kripton2035 is reffering to the capacitor on the IC's Vcc pin( in your case 25.1V). It's a small electrolytic between pins 6(Vcc) and 4(GND). Not the mains rated 400V capacitors.

[flolic]

Check IC2, TL431 reference

[nixxon]

Check out the video on

https://www.eevblog.com/forum/chat/nice-video-tutorial-on-switching-regulator-power-supplies/

Randy Fromm (1992)

After initially talking about how the power supply works in general (components and their functions) he mentiones a few probable causes of failure (or failiure?). He divides the failures into two groups: The failures that make the fuse blow and the ones that don't make the fuse blow. Maybe you find something useful here.

[dr_p]

Check R2, it probably is open circuit or dry jointed. It is the startup supply to the controller.

Q5 is used to switch the unit to standby, where it produces low output voltages, but still runs. Normally then the 30v rasil drops to 10V or so

you were right

from here:

I wanted to first check the supply and took it out of the instrument. The connector has 5 pins, the 2 black are GND, 2 of the others are 24V and 32V and the one on the end toward the edge of the board is Sense. [...]. AND you will see 7.5V and 10V until the sense line is active, the sense sits near GND if the AC adaptor is taken out of the printer. The sense line goes through a 1k resistor and to the base of an NPN transistor on the power board. [...] Anyway, I put 3.3V on the sense pin (almost anything will work because of the 1k resistor) and GND on one of the GND pins. That made the other two pins go from 7.5V/10V to 24V/32V. So apparently there is a low power mode and normal power mode.

[dr_p]

Hi

I've checked now the power supply on Pin 6 of my controller, got 25.1V. So i think this should be okay in my case.

The 2 small primary electrolytic caps on my supply are 1uF both. They look okay. I check they tomorrow with the ESR meter. Problem is, havent got 100V electrolytic caps here.

Checked R2, its fine. 318V on Pin 8 of the controller.


thanks everyone

also, check that the output diode D10 is not shorted

[Nermash]

also, check that the output diode D10 is not shorted

When repairing smps, if primary transistors and fuse are OK, I always turn my attention to secondary diodes. If they are shorted, control IC will try to start and immediately shut down the gate drive due to excess current.

[zaoka]

Measure output voltage at the time you apply power to the unit. (you can use MinMax feature of your meter, range 30V).

If you have voltage for a second it usually means that regulation circuit is not OK, if there is no voltage at all suspect that IC in primar.

[Gideon]

Hello How do i : please help : I put 3.3V on the sense pin (almost anything will work because of the 1k resistor) and GND on one of the GND pins. That made the other two pins go from 7.5V/10V to 24V/32V

How do i do that so thatthe ower supply stay on 24v and 30v ?

Hi there..

Im currently trying to repair a switched power supply. Its from an Canon MP610 Printer.

The printer dont want to work anymore completely. I checked the power supply and measured no output voltages.

I checked the 3 fuses, all okay. I have the 318V DC on the big primary cap. I measured between Gate and Source of the MOSFET, no voltage. 318V between Drain and Source on the MOSFET, okay.. MOSFET: K3562
Its a little driving IC on the PCB, labeled with 2A07D88 / 5518. I cant find any datasheets on that device.

The transformer has 6 connections primary and 4 on the secondary side. Can anyone say how the connections internally works? Cant find any datasheets on that device, too..

On the Output connector is an connector that trace is going to the Q5 device, i presume a transistor (Labeled with "XN" or "NX") Any ideas for what that connector is?


Thanks for any help, dont wanna throw the thing away..

[the_grinch]

Hi,
I found this topic on google and I have the same problem.
Everybody think it's a power supply problem (and then buy a new one) but actually not at all.
The problem comes from the motherboard inside the printer with the yellow wire (the "sense" wire) which makes the transistor Q5 turning on.
Im gonna look for a solution.
Buy !