PlaystationTV help

[Mrwoodchuck]

Hi all,

so I am still learning things and trying to figure out diagnostics and could use a hand.

In this picture I have noted that I have 5v all the way up until after the bigger capacitor "or what I think is a cap" and the component marked 077A


I found that there is a short between ground and everything past those two components, which makes sense why it wont power up.
I am assuming that the part marked 077A is the culprit but its just a guess, I think its some kind of vrm?

Does anyone know what this part is? I have been searching and have come up with nothing.


Back side of the board in the same section.


Thanks for any help!

[sean0118]

It's probably some kind of buck converter with an integrated mosfet and diode.

You could try to remove the inductor, the one with A8 printed on the top. With it removed you'll be able to measure if the short is in the buck, or if another component is pulling its output low, depending on which pad is 0 ohm with the inductor removed. It's common for ceramic capacitors to go short too, so if the problem is on the buck side it might not be the IC itself.

[Mrwoodchuck]

Thank you for the reply,

I followed your advice, and everything was still shorted that was before.

so i removed the big cap that was next to the "buck converter" no change, removed the other two caps, no change.

lastly i removed the "buck converter" and nope no change.. all the path that i had highlighted yellow was still shorting to ground.

i was able to put everything back on but the inductor got damaged, i measured it before it fell apart and it was 2000 Microhenries, so at least i know what to buy.

but im not sure what to do next.

[sean0118]

I think the large part highlighted in yellow is meant to be GND, hard to be sure from the photo though.

Were you able to measure both inductor pads to GND? Only one of them should be shorted.

2000uH seems too big, someone else might know the exact replacement though, you also need to know it's current rating etc.


edit: I don't think that smaller trace you marked in yellow should be ground, so that might indicate a problem in the buck IC. That looks like a resistor divider from the output being used for feedback.

[Mrwoodchuck]

I think you might be right..

the single pad closer to the 077A is saying ground but the other is not.

on the meter i had it set to the 200uH and it was reading 10.0 so  200x10 = 2000 uH ?

thanks for all the help.

[perieanuo]

next, you try to identify which voltage your stepdown dcdc is doing, cut your dcdc output with precision knife and inject that voltage from a power supply with current limiting, increasing the current slowly from 100 mA. At one moment, the defective component (most likely some capacitor) wil start to get hot, you can check with the back of your hand
if you did that before removing inductor and dcdc chip, you had already found you bad apple
sometimes you have to raise the current up to several amps
for the bold, you can apply directly the correct dcdc output target voltage (with your dcdc output cut from the powered circuit) with current set to 5-10 amps, your died component pop up in a second, but that method risks to vaporise one or 2 tracks
if you don't know what voltage your dcdc is supposed to fabric, look at the datasheet for the ic's powered by the dcdc, it may be 3v3, 1.1v or who knows. usually it's easy to get this, if you're lazy, just limit your ps to 1V and start raising the current, but this is not ideal if you have another dcdc's after that one you dismounted, and they are trying to start from smaller voltage that the one they normally do, it will work also but not really a good idea to keep pwm ic's in continuous startup situation.
stick yourself to first phrase procedure, try to get the principle i just talked about, then repairing this king of equipment gets easier
that method is not some mumbo-jambo, they use this when designing dcdc converters, for viewing how the dcdc/smps starts and shts off because of over/undervoltage, how it behaves when the fluctuation form mains 220Vca are coming (cause they are coming for sure at one moment). it comes from experienced hardware designers

[Mario87]

I think the large part highlighted in yellow is meant to be GND, hard to be sure from the photo though.

Were you able to measure both inductor pads to GND? Only one of them should be shorted.

2000uH seems too big, someone else might know the exact replacement though, you also need to know it's current rating etc.


edit: I don't think that smaller trace you marked in yellow should be ground, so that might indicate a problem in the buck IC. That looks like a resistor divider from the output being used for feedback.

100% agreed with this, all those capacitors are likely for decoupling and that big area of yellow (GND) in the first pic between the IC and the inductor is likely meant to be a ground plane for decoupling of the supply before and after.

The only concern is that 1 small trace that has been highlighted in yellow on the output of the IC. That most likely should not be GND.

I think you might be right..

the single pad closer to the 077A is saying ground but the other is not.

on the meter i had it set to the 200uH and it was reading 10.0 so  200x10 = 2000 uH ?

thanks for all the help.

No, when you have you meter set to 200uH that is it's range, it means it can measure up to that value, not that it jumps up in increments of that value. So 10uH is your inductance of that part.

Also, this isn't really making sense now, as you said when you remove the IC and inductor there is still has a short to ground, but then you say 1 of the legs of the inductor is not shorted to ground, however that particular leg that is not shorted to ground is the one linked to the small trace on what appears to be the output or feedback of the IC.

So can you clarify, now that the inductor is removed, does that small trace next to the '6' on the IC still show a short to ground?

I'm betting that now the inductor has been removed the short to ground has disappeared also on that output trace.

However you mention the pad for the inductor nearer the IC has a short to ground, that is probably your issue, that pad should not be shorted to ground.

An inductor is a coil of wire, it is by definition, a short circuit between its 2 connections, it should not be connected between power and ground on it's own, only when in an RLC filter with a capacitor in parallel. So the fact that pad for the inductor is shorted to ground indicates to me a problem elsewhere on the board where that pad connects to.

As has been said earlier, the "island" of copper you highlighted yellow is likely supposed to be a ground plane for decoupling purposes which is fine, but if the inductor was meant to also be coupled to ground the common sense approach for the designer would have been to join that pad to the "island" on the top, but they have not, that inductor pad which is shorted to ground is on its own and 99% disappears within the board to other layers because it is not meant to be connected to ground at all.

I would try to find out where that pad connects to on the other side of the board (easier said than done) and start inspecting components on the other side for shorts to ground, as it looks like you have narrowed down the issue to being the connection on the side of the inductor not directly connected to the IC. Likely there is nothing wrong with the IC itself or anything in the area you have been inspecting.

[Mrwoodchuck]

Thank you for the info on reading the lcr meter, as you can tell im still kinda a rookie lol.

You were 100% right Mario87 the little trace is no longer grounded.

unfortunately the backside has nothing in the location of the grounded pad, so i think it is using an internal plane.

do you think at this point i should try perieanuo's suggestion of power injection?

[Mario87]

Thank you for the info on reading the lcr meter, as you can tell im still kinda a rookie lol.

You were 100% right Mario87 the little trace is no longer grounded.

unfortunately the backside has nothing in the location of the grounded pad, so i think it is using an internal plane.

do you think at this point i should try perieanuo's suggestion of power injection?

My next point of call would be to look at what that IC is outputting without the inductor in place now that it is not shorted to GND any more. Check the voltage (3.3V, 1.8V, etc), if you have an oscilloscope that would be best as it will confirm for sure if it's genuinely just a DC-DC buck converter or of its some other kind of device outputting a signal of some specific frequency (not likely, but good to check).

Once you have that info I would inject the same signal you just checked in the last step into the pad of the inductor which is shorted to GND and current limit to about 50-100mA. Then check with a thermal imaging camera to identify which component is getting hot and causing the short.

I appreciate that you might not have access to a thermal imaging camera, however 99% isopropyl alcohol can work too, just put a little onto each component while powered on and if one is shorted to GND and getting hot the IPA will sizzle off instead of just sitting there. That will identify your failed component, then once you know that you can decide what to do next.

[Mrwoodchuck]

Ok so i just had the chance to check things out,

I kind of messed up on my reflow of the buck converter so I just reflowed it, then I connected the normal power and was poking around with my probes.

I now have voltage on the inductor's single pad closer to the 077A  0.480v  !!!

it is no longer shorted... i have no clue why or how...

so now now i would have re attached the inductor but thats damaged, can i jumper it for a short time to see if it powers on?

Thanks!

[Mario87]

so now now i would have re attached the inductor but thats damaged, can i jumper it for a short time to see if it powers on?

No, you can't just use an jumper wire instead of the inductor. The inductor is an energy storage device, as the buck converter is rapidly switching on & off to provide a lower average output voltage, the inductor is taking that energy, storing it, then releasing it as a smooth continuous output instead of the rapid switching of the buck converter.

Chances are it possibly won't turn on without the inductor in place. You should be able to get one from RS, Farnell, Mouser, DigiKey, etc by next day shipping. Even eBay could probably get it to you in a couple of days.

[Mrwoodchuck]

Well part came in and installed and .480 volts across but still not working.

on the 077A ic starting at the pin one mark going clock wise pin 10 now has a voltage of 2.146

man i was hoping that would have been it.

to bad i dont have a schematic or even data sheets for the dang parts.

[RJSV]

Old-School factory provided 'MILLI-OHM METER':
  For repair by tracing for shorts, a Technician guides and pokes the two probes, along traces. You then can use a 'bigger or smaller' method for zeroing in, on the minimum 'trace to nearest ground trace'. Also checking trace to trace, keeping in mind components along there.

This worked for finding tiny hairline solder shorts or also hairline defects as a new shipment of PC Boards.
The difference, in resistance, as it is displayed to thousandths, can be read as only a few centimeters has large enough change (in milliohms).
   For high volume factories, they can afford an industrial 'MilliOhm Meter'. Not sure how low prices are in hobbyist markets...

[Mario87]

Well part came in and installed and .480 volts across but still not working.

on the 077A ic starting at the pin one mark going clock wise pin 10 now has a voltage of 2.146

man i was hoping that would have been it.

to bad i dont have a schematic or even data sheets for the dang parts.

It’s not easy to tell in the pictures, is pin 10 of the IC the one that goes to the inductor?

2.146 is not a common voltage, I’m betting it’s supposed to be 3.3v and there is still a partial short which is pulling down the voltage.

Did you measure the voltage output of the IC without the inductor installed? What was it? If it is say 3.3v without the inductor but drops to 2.146 with the inductor installed it basically confirms a partial short pulling down the voltage.

Sure, schematics and data sheets help a lot, but there is still more you can do as long as you have the time and patience to keep digging deeper.

[Mrwoodchuck]

Good idea, on the 'MILLI-OHM METER'
problem is anything that is not cheap Chinese branded seems to start around $200.

a little over my budget for this project..


so after removing one the inductor "with out destroying it this time" I still have the same voltage of around 2.146..

The yellow pin is the one that is 2.146, all red pins are 5v and black is ground.

the ones uncolored are to hard to reach or No signal, not ground but no voltage.
right now the one that has the little trace that goes to the inductor is no signal because the inductor is off, with it on it was about 83 mV...

thanks for all the help I really appreciate it!

[Mario87]

right now the one that has the little trace that goes to the inductor is no signal because the inductor is off, with it on it was about 83 mV...

Ok, so this is what matters as it appears this is the output of that IC. An 83mV output will be insufficient to power anything.

Can I ask, with your LCR meter, are you able to measure the 3 components that are between the pin on the IC and the inductor? Looks like it is some kind of LCR filter setup. There is 1 capacitor, but I'm not sure if the others are resistors or inductors.

I THINK it looks to be that the first component that goes between the trace and GND will be a resistor and then the capacitor will be parallel with an inductor and that would give a band-stop filter, but that doesn't make much sense if this is just a simple DC-DC buck converter, why would you have a band stop filter, surely they would just have a low-pass filter or decoupling capacitors to ensure a clean DC output with little noise.

If you can measure those components and confirm first of all what they are and what their values are it might help us better understand that is happening in that area of the circuit (ie, what it's supposed to be doing).

Also, can you check for continuity between the pad of the inductor near the IC (the one that was previously shorted to GND) and the pins on the IC? Just because you mentioned re-flowing the IC fixed your short to GND previously, I suspect that pad will internally be linked back to the IC for feedback on one of the layers of the PCB, but would be good to confirm this and also to confirm which pin. As the fact your re-flow "fixed" the short to GND to me says 1 of 2 things....

1. This IC I suspect will have a GND pad on the bottom and your previous solder / rework attempt resulted in a short between the GND pad and the pin on the IC being used for feedback, hence that inductor pad close to the IC was shorted to GND and a reflow of the IC fixed it. This to me is unlikely as the chances of that 1 pin of the IC being shorted and no others is slim.

2. There is an internal fault with the IC and the heat from re-flowing has somewhat helped with resolving the internal short to GND on the IC feedback pin. In which case the only solution is a new IC.

However if you can check the value of those components with your LCR meter and confirm if the inductor pad closest to the IC does indeed feed back to 1 of the pins on the IC it will help gain a better understanding as to what is going on and how things are connected so we can better diagnose what is happening.

[Mrwoodchuck]

sounds good, do you want me to test in circuit or take them out?

[Mario87]

Preferable to take them out to test then re-fit as we just want to determine what they are exactly and their values. Having them in circuit would mean other components can affect the readings.

[Mrwoodchuck]

Understood, sorry this might take me a week or so, I am working on buying a house and it has taken up a lot of my time.

[Mrwoodchuck]

I had time today to get this done,

I have numbered them from 1 - 3 to make it easier.

#1 reads 22pf
#2 reads 150K ohms
#3 reads 30K   ohms

no other measurements work working for each device, so I just assume that it is 1 cap and 2 resistors.

Thanks for waiting.