Samsung LE32S73BD Flashing standby LED no backlight

[dicky96]

Re: Samsung LE32S73BD Flashing standby LED no backlight

I am measuring the AC in directly on the 220V connector to the PSU board. See pic.

I am measuring boost between the heatsink which is connected direct to -ve of main smoothing cap and a link on pcb that connects direct to the +ve of the cap

I made a video to show what I described - AC input drops to 10VAC with PSON activated and the PSU generates 12V on 24V rail (I did not video this 24V rail, but it does). When I am saying I am 'switching on the power' I mean I am switching the 2.5V from bench psu to the PSON pin via 390R resistor

https://youtu.be/FiN9FSYmPGE

As I stated with the 105W lightbulb instead of 60W, it flashes briefly (and dull) on first application of 220V, it does not light at all when I activate the SMPS with PSON supplied from bench supply, the boost is reading 398V, and 24V output is reading about 12-14V.

I don't have much experience with PCF to be honest. What should I expect to see on main cap boost voltage with 220VAC in? Is 398V about right?

So now you have all seen the video - did I accidentally invent something lol

[budm]

Re: Samsung LE32S73BD Flashing standby LED no backlight

As you can see in your Video, when the lamp is not lit you do get over 200VAC to feed the board so the cap got charge up to around 300VDC, and when the lamp lights up then the AC drops down to 10VAC range since the cap was charge up it will take awhile for the DCV on the cap to discharge to lower lvel, if you keep the AC input to the board at 10VAC and monitor the Voltage on the cap long enough you will see that it will not maintain at 309VDC.
So back to the 24VDC power supply section, if you do not use any lamp in the AC power, do you get steady Voltage around 12~14VDC at the output with may be using about 1/2 A load on it.

[dicky96]

Re: Samsung LE32S73BD Flashing standby LED no backlight

That sounds about right bud. With the 60W limiter and PSON energised I get about 309V on main cap, slowly discharging.

If I do the same and then turn off the AC mains, the main cap discharges considerably faster.

Anyway, on to the other tests

I made up a load from 8x 4R7 15W resistors I had lying around, I wired them in series and used a croc clip so I could select the number of resistors in the load.

Expecting about 14V on the 24V rail, I connected the load using six of the resistors giving 6 x 4.7R = 28.2R so it should give me 0.5A dissipating about 7W

Anyway with the following conditions:
No light bulb limiter
main board and inverter disconnected
28.2R load connected to 24V output
PSON at 2.5V from my bench supply via 390R resistor

I am seeing a steady 3.65V on the 24V rail.


Repeating this test without the load I am seeing an initial 12V on 24V output rising to 15.65V over about 10-15secs then stable

Yesterday evening I removed all the electrolytics on both hot and cold side and tested each individually for ESR and Capacitance.

All were good, capacitance correct or slightly higher than marked, ESR around 0.010 ohms or thereabouts on 1R range, but one had a slightly low capacitance (CM856 read 960uF rather than 1000uF) so I replaced that one anyway even though it was within tolerance

Already checked the power semiconductors and they look OK in circuit best I can tell. Anyway the PSU 'runs' so I doubt there is a problem here

Gonna look around all the small semiconductors next for short or open circuit. And check low value resistors as I could imagine one of these going high causing something like this.

Rich

[dicky96]

Re: Samsung LE32S73BD Flashing standby LED no backlight

OK I had a look at all the through hole resistors (seem OK in circuit) and semiconductors (nothing obvious on diode test DMM)

Only thing I notice is the diode side of one of the opto isolators PC805S has a lower forward voltage drop than the other two of the same type. This is in circuit so could be erroneous.

I may consider pulling the optoisolators and making a circuit on breadboard to test.

Or swap the one that reads a little different with one of the others and see if the symptoms change.

The optoisolator PC803S which is a different type reads quite different forward voltage


I had a look at as lot of the smd devices on the rear of PCB but not all as I had another idea to maybe try to narrow this fault down.

It occurred to me that if the PSU was running at minimum duty cycle, and the duty cycle did not change when a load was connected due to some feedback or regulation fault, it may cause a fault a bit like I am seeing?

So I tried to determine this. As mentioned my isolation transformer and variac are still back in the UK

So I tried resting my scope probe across the SMPS transformer primary at 1V/div(or was it 0.5V/div) and 50uS/div to see if I could determine whether the duty cycle changed when I connected the load. I've not tried this method before it was just an idea.

Anyway the results are not quite what I thought I may see - don't know whether someone with better knowledge could explain them?

The tranformer is clearly ringing especially with no load, the Mosfet 'pulse' if that is what I am seeing is pretty much the same width with or without the load but the frequency is higher with a load. 3 divisions (150uS) compared with 4 divisions (200uS) See pics- 1st pic is with no load.

Interesting?

Oh - is it possible/safe to connect my scope ground to main capacitor -ve via a high voltage cap so I can probe the MosFET drive without isolation transformer?

Rich

[dicky96]

Re: Samsung LE32S73BD Flashing standby LED no backlight

I've not given up on this one.

Does anyone have a proper datasheet for the F9222L? I just find a one page pdf that does not tell a lot about it.

When I did those checks with and without load and my scope probe near to the transformer, I didn't expect to see a change of frequency when the load was present on 24V, I just hoped to see a wider pulse when the mosFET is on.

Now thinking from basic principles could someone help work out what could be going on?

From what I understand of smps design, when the mosFET conducts it draws current through the primary and stores energy in the coils field, then when it switches off this field collapses and energises the secondaries. So the greater the pwm duty cycle, more energy is stored - until the transformer core saturates and it goes bang.

For some reason this PSU seems capable of generating only minimum energy in the secondaries.

I can think of a few reasons for that and my observations upon them. With reference to schematic on post #3



A shorted turn or excessive load somewhere on the secondary.
Considering I can't find any shorted semiconductors and all secondaries generate some voltage that seems to have been eliminated. Also the primary seems to 'ring' quite nicely with no load and is not being dampened yes? So I don't think I have an excessive load condition

Excessive resistance in the primary drive circuit eg. high resistance mosFET D-S, high resistance series resistor (probably a current sense resistor).
Not sure how I could check forward resistance of the mosFET which is part of F9222L anyway. Anyway wouldn't such a problem cause the mosFET to get excessively hot and blow?

Doesn't appear to be a low value current sense resistor in series with primary/mosFET on the schematic.

I'm not certain about the role of CM802. It looks like all the primary coil pin2-pin4 supply voltage flows through this cap when the mosFET switches. If so I can see how a faulty cap here could cause low outputs from the PSU. However the cap checks ok on capacitance meter. I am not really sure what the circuit designer had in mind here anyway. It seems very stressful passing all that energy through A cap. I've seen 110V motors driven from 220V mains this way and the caps tend to fail. Can anyone help explain why this circuit is the way it is?

Secondary winding pin5-6 looks like some sort of gate drive for a high side mosFET built into F9222L?

Secondary winding pin 7-8 looked like a running supply for the F9222L but it isn't as there is no rectifier or electrolytic to smooth it. Some sort of feedback?

PWM to narrow (low duty cycle) due to drive or feedback fault
This would cause low power output from the PSU I am sure. But how to test?

PWM Drive frequency too low
I'm pretty sure this would cause low output as well. I would have excpected the PSU to run at a set frequency and variable PWM but form my tests it seems to run at variable frequencies - and they seem a bit low to me. (about 6.6KHz no load, 5KHz with load on 24V)

But with no datasheet for F9222L I honestly don't know. Should it be fixed frequency? Does it have internally set oscillator or use external RC? Not sure what pins STB, CON, CB, CS, COMP, VREF aresupposed to be doing without a datasheet.



Can't think of any other causes for these symptoms off top of my head. Am I thinking right here guys? Or missing something.


Guess I could change F9222L and possibly the CM802 - they ain't overly expensive and this job has turned more into 'investing in my own knowledge and experience' rather than making a profit vs time


Advice or corrections welcomed
Rich

[vinceroger69]

Re: Samsung LE32S73BD Flashing standby LED no backlight

if this was my set as im still learning fault finding myself i would just swop those two parts if cheap enough and easy too get, atleast you have ruled them out then.

[dick_barton]

Re: Samsung LE32S73BD Flashing standby LED no backlight

There's a schematic using an F9222L that could be of some help


https://www.youtube.com/watch?v=02Re-UmG9Gc

[dicky96]

Re: Samsung LE32S73BD Flashing standby LED no backlight

I've ordered a F9222L and the 56nF 640V capacitor, The owner is in no rush so I will pick these up next time I am back in the UK which will be a couple of weeks or so

Rich

[dicky96]

Re: Samsung LE32S73BD Flashing standby LED no backlight

FIXED
------

Replaced F9222L and the CM802 56nF 630V capacitor at the same time and the TV running fine now.

Thanks to all who assisted
Rich

[budm]

Re: Samsung LE32S73BD Flashing standby LED no backlight

Thanks for the follow-up.

[dick_barton]

Re: Samsung LE32S73BD Flashing standby LED no backlight

Good news