Philips 60PFL8708S/12 - QFU1.2E LA - Error code 53

[kazbayadum]

Re: Philips 60PFL8708S/12 - QFU1.2E LA - Error code 53

Hello, I am going to share my experience with you.

A year and a half ago (2021 August) I got the famous error 53. I disassembled the TV and tried to reflow it without success on my own. As it seems to me something that requires some previous experience, I took the green motherboard to a store in the neighborhood, where they repair video game consoles. There they did a reflow.

The sales clerk did not charge me because he had no way of checking that he had done the job properly. I took it home and it has been working until last month. I don't use the TV much, to be honest, but it held up pretty well. The only thing I changed was the thermal paste. I used the same ceramic heatsink it comes with.

Now I am going to repeat the feat (2023 January). This time I will try to find a 54.6mm heatsink to replace the previous one, with two diagonal holes and some mechanical system that presses the heatsink against the board. I'm hoping that with that and a custom drill hole in the back cover, it will be enough. I don't like fans, I'm quite a fan of fanless systems.



On the other hand, while I'm here, I've never been entirely happy with this TV. The controls seem stilted to me, the smartTV is slow (and with a proprietary OS) and the keypad signals from the remote, don't transmit over HDMI as I expected would happen. Do you know of any hacked firmware updates that would improve these aspects? Maybe AndroidTV?


Best regards and if I remember I will send you pictures of the result!

[Maalobs]

Re: Philips 60PFL8708S/12 - QFU1.2E LA - Error code 53

Hopefully you fixed it for the final time now with the proper reballing.

The original TV from this thread is still the daily driver of my friend and it's working flawlessly.

[dominik0801]

Re: Philips 60PFL8708S/12 - QFU1.2E LA - Error code 53

Hello,
It's that time of the year again.

I have already blown the Fusion chip in my TV with hot air 10+ times, but the double blink always came back, so last time (2021 september 22) I did a proper reflow under the BGA machine.

I told to myself, that now we will see how long it will work, but the next time it fails I'm going to reball the chip.

Well.. today morning I woke up, wanted to watch some TV, but aww no, double blink..
I didn't even order stencil yet, and my gf comes to my place for a week, so I will need the TV fast. So I was like, okay.. reflow it once again, and next time I WILL reball it.

----------------

So I did the reflow today, which in fact made the TV come alive, so I put it together, but when I tried to switch to my TV BOX via HDMI, it wouldn't work..

I tried all 5 of the HDMI ports, but nothing.
It did recognize, that I plugged in a HDMI, but it would simply say "No video input" or something. Tried it with a laptop too, without any success.

Meh.. I really didn't want to wait for that stencil I ordered today, so I desoldered the CPU and reballed it without stencil - I did it already with GPUs, I knew it will be long and frustrating, but it is what it is..



I did the reballing, soldered it back and yes.. the TV works now, even the HDMI ports are functioning.



I didn't have 0.65mm balls, on other forums they said that it is the right size, I used 0.6mm, hopefully they won't fail.

If I was already there, fixed some loose-looking SMD caps and resistors, oiled my CPU fan, put a fuse on the +12V TCON jumper wire as it didn't have any when I soldered it in.

Let's see, how long it will go.

[Maalobs]

Re: Philips 60PFL8708S/12 - QFU1.2E LA - Error code 53

The TV has worked fine in my house during a rew days. Then it traveled a hundred+ km to my daughter's house. And there it started well but is beginning to stop working after some minutes.

Did you tap the corner of the chip while it was heated, as I described in this earlier comment?

And finally install some kind of clips on the heatsink to apply it as tight as possible to the SSB IC.

Looks like your model doesn't have the 54.6mm diagonal mounting holes, unless the heatsink covers the holes in your photo?
I think you are fine if you are using a thermal adhesive underneath the heatsink already, but if you find a way to push down the heatsink with mechanical force, please share it with us with pictures.

Something that would be convenient in this case is if there was a USB-fan product with an actuated arm like the Arctic Breeze Mobile, but also with a built-in fan-control:

I bought one of these to investigate the sound level of it.
It turned out that it runs at full speed and just the wind noise was too much, but the worst part was that having the fan out on the end of the actuated arm creates a mechanical vibration that reverberates into the SSB and chassis, which created a noise that was far too loud for this fan to be useful.
If only it had a way to dial the fan speed, it would had been a cheap and very simple solution for our particular problem.
That's why I went with the arguably overkill solution of a controlled PC-fan instead, which works very well but costs more.

[Davi.p]

Re: Philips 60PFL8708S/12 - QFU1.2E LA - Error code 53

My suspect is that by some manufacturers a tin with short life is carefully tested to shorten device life, so new solder must be put in, maybe also in the die if possible...

[William1967]

Re: Philips 60PFL8708S/12 - QFU1.2E LA - Error code 53

The TV has worked fine in my house during a rew days. Then it traveled a hundred+ km to my daughter's house. And there it started well but is beginning to stop working after some minutes. I will recover it later and I will tell you if the PCB is bent.
What I plan to do is to reflow it again in the oven, then reinstall the foam pad below the SSD board to keep it as flat and stable as possible. And finally install some kind of clips on the heatsink to apply it as tight as possible to the SSB IC.

Reballing is the only solution, the solder used in the factory is not good anymore. Heating it up in an oven might work but usually not for long.

[Zano64]

Re: Philips 60PFL8708S/12 - QFU1.2E LA - Error code 53

The TV has worked fine in my house during a rew days. Then it traveled a hundred+ km to my daughter's house. And there it started well but is beginning to stop working after some minutes. I will recover it later and I will tell you if the PCB is bent.
What I plan to do is to reflow it again in the oven, then reinstall the foam pad below the SSD board to keep it as flat and stable as possible. And finally install some kind of clips on the heatsink to apply it as tight as possible to the SSB IC.

[Davi.p]

Re: Philips 60PFL8708S/12 - QFU1.2E LA - Error code 53

Let us know how it goes... and tell me if the board watched from long profile is bent

[Zano64]

Re: Philips 60PFL8708S/12 - QFU1.2E LA - Error code 53

Some months ago I posted concerning the revival of my 47pfl7108h_12 TV set by blowing hot air to the SSB CPU.
Unfortunately after some time it began stopping and restarting without reason, till it completely stopped with the well known LED blinking.

I did not read the last post about a day cooking (sorry Davi.p), so I applied another method (inspired from another blog post) which I will call the "Kitchen oven reflow".
You will see on attachment the SSB PCB in the oven, set on bolts and nuts high enough to obtain more or less equal temperatures above and below the board.
I achieved the temperature curve (measured by a thermocouple) similar (I hope) to a pro reflux process. The most difficult part was that the oven thermostat is not at all accurate.

In addition to that kitchen reflow I replaced the heatsink by an aluminum one, see photos which this time I managed to post.

And the set is revived for the second time.
How long will it last ? No idea at all
See you later folks.

[Davi.p]

Re: Philips 60PFL8708S/12 - QFU1.2E LA - Error code 53

For anybody that comes here in the discussion and wants to make a reflowing job, it's important to slowly cook the chip for a day before attempting a reflow, this is to prevent the famous popcorn effect that destroys the chip in fact, for this purpose you can use an oven at about 100 celsius degrees connected to an external timer for mains interruption after 24h. So the reflow preferably mostly from the bottom or both..

[Maalobs]

Re: Philips 60PFL8708S/12 - QFU1.2E LA - Error code 53

Nicely done!

Although, that small fan looks like it could become a little loud and whiny on permanent max speed while the TV is running.

I bought a 75" Philips 75PUS6754/12 a couple of years ago and as a pre-emptive measure, I opened it up to inspect the state of the planned obsolescence in the device.
When I removed the heatsink assembly from the main processor I noticed that whatever they had used as thermal paste underneath it, could easily have been mistaken for a seagull's watery diarrhea.
I cleaned that mess out and replaced it with some Thermal Grizzly Kryonaut instead.
The heatsink still got very hot though and the entire system was clearly baking inside the chassis under prolonged use, so that wasn't going to be enough.

Then on further investigation I noticed how the USB-ports on this system actually power down properly when the TV enters the standby-state.
That's useful, so I bought a PWM-controlled 12cm PC-fan, and made a hole for it in the TV's plastic back case right above the heatsink, leaving the corners in place so I could drill holes there for the fan.
Then I mounted the fan into the chassis with the usual self-tapping screws that are included with the fan.
It worked out quite nicely, and I have open space behind the TV as well since it's standing on the feet.

I got a quiet Noctua fan, and they also sell a small separate fan controller device which has a dial on it to control the fan speed.
I already had a small USB-to-12V step-up adapter from somewhere, similar to this thing, which I opened up so I could solder the input of the fan controller directly to the output-pads on the converter's PCB:

Then I dialed the fan speed down to a tolerable noise level on the controller, which with the quiet and stable Noctua meant just reducing the woosh of the air blowing.
The fan also has rubber bumpers in the corners so there is no vibration carried into the big, wobbly back case either.
It has worked flawlessly for the past two years, and the fan turns off as expected when the TV goes into standby.




The point is; you also have to get the hot air out of the device, so make a hole above that fan as well.
You'll find that the plastic is surprisingly easy to work, even with crude tools!

[DamneD2000]

Re: Philips 60PFL8708S/12 - QFU1.2E LA - Error code 53

My solution to make my 42PFL6907T work was the reballing of the CPU and making active cooling with heatsink taken from my old VGA card.
Here is how it looks like

[DamneD2000]

Re: Philips 60PFL8708S/12 - QFU1.2E LA - Error code 53

My solution for the QFU platform (42PFL6907T) was the reballing of CPU with active heatsing cooling, made from my old PC VGA card. Some pics of the final improvement

[dominik0801]

Re: Philips 60PFL8708S/12 - QFU1.2E LA - Error code 53

Dominik; Wait, so you had access to a BGA soldering machine the whole time, but you didn't use it?

The guy who did the reflow for me, which was just a reheat and reuse of existing solder, he used a huge hot air station that is not meant for soldering I think but for other industrial purposes.
The handle was the size of the operator's remote control for a small industrial crane.
It was a few years ago now, but as I recall the actual heating took just a couple of minutes or something, and after that we chatted for maybe fifteen minutes while we waited for the board to slowly cool down.

He talk about how there was an art to doing this sort of thing.
He heated up the chip until it "danced" on the melted solder, then he gently tapped the corner of the chip with a pair of tweezers and watched the chip flow back into position again by itself.
Then he immediately stopped heating the chip any further.
Maybe he applied additional flux too around the edges, I didn't see that, I just heard him start up what sounded like a leaf blower in the back room.

I don't know what temperature or air flow he was running at, but he said the benefit of the big machine was that the nozzle was so wide that it heated the entire chip at once, and the machine was powerful enough to heat consistently across the entire area which also meant that he could work fast enough without heating the chip for too long.

The TV that this board was for, is still running perfectly fine.

And why knock on someone else's experimentation, finding out ways that don't work is also a valuable learning experience, if nothing else it narrows the field for later attempts.

Another method the guy showed me that was pretty cool, was how he quickly handled smartphones that people had let the battery drain on completely.
If you let the battery drain entirely, you risk a situation where the battery won't charge up anymore.
There is some careful charge-up protocol you have to use to recover the battery to a state where the normal charger can take over again, but he had reduced the process to art and skill again.
He had an old HP rack-powersupply up on a shelf with multimeter probes hanging down from the output plugs.
The power supply was set to some wierd voltage with two decimal precision and CC-mode for a specific current.
When a phone arrived for repair, he opened it to expose the battery contact pads and put the probe tips against the pads and watched the voltage drop on the power supply.
He had learned that when it dropped to a specific level, he would lift the probes, count a few seconds then put down the probes again, and repeat until he saw a familiar change in how slowly the voltage drop happened.
Then he'd plug in a USB-charger into the phone and let it charge the rest of the way.
It just took a couple of minutes and was like watching a virtuoso at work.

And I can tell you this, he didn't wake up one morning and come up with the exact method, this was the result of a long line of failed attempts and discoveries of how NOT to do it.
The trick is as always to never repeat the same mistake again.

Well, I bought the BGA machine this year in March, so when I first got the problem, I didn't yet have the machine. Then as you saw in the boring post, in february I had a heating so successful, that it lasted till this month, so I just didn't have to use it on the TV.

Also, when I got the machine, I wasn't so comfortable with it yet, popcorned a few GPUs. Not to meantion, that I later realized the machine had a factory fault that would have easily killed me, but when I did the glass removing and IR plate changing mod on it, I fixed that problem too.

Now I am comfortable with the machine, I am changing GPUs on VGAs weekly, so this is why I was ready to use it on my TV. Hopefully it will last long this time, bc I love it.

And yes, experiements are nice ways to learn new stuff, this TV is definitely an experiementing playground. :P

[Maalobs]

Re: Philips 60PFL8708S/12 - QFU1.2E LA - Error code 53

Dominik; Wait, so you had access to a BGA soldering machine the whole time, but you didn't use it?

The guy who did the reflow for me, which was just a reheat and reuse of existing solder, he used a huge hot air station that is not meant for soldering I think but for other industrial purposes.
The handle was the size of the operator's remote control for a small industrial crane.
It was a few years ago now, but as I recall the actual heating took just a couple of minutes or something, and after that we chatted for maybe fifteen minutes while we waited for the board to slowly cool down.

He talk about how there was an art to doing this sort of thing.
He heated up the chip until it "danced" on the melted solder, then he gently tapped the corner of the chip with a pair of tweezers and watched the chip flow back into position again by itself.
Then he immediately stopped heating the chip any further.
Maybe he applied additional flux too around the edges, I didn't see that, I just heard him start up what sounded like a leaf blower in the back room.

I don't know what temperature or air flow he was running at, but he said the benefit of the big machine was that the nozzle was so wide that it heated the entire chip at once, and the machine was powerful enough to heat consistently across the entire area which also meant that he could work fast enough without heating the chip for too long.

The TV that this board was for, is still running perfectly fine.

And why knock on someone else's experimentation, finding out ways that don't work is also a valuable learning experience, if nothing else it narrows the field for later attempts.

Another method the guy showed me that was pretty cool, was how he quickly handled smartphones that people had let the battery drain on completely.
If you let the battery drain entirely, you risk a situation where the battery won't charge up anymore.
There is some careful charge-up protocol you have to use to recover the battery to a state where the normal charger can take over again, but he had reduced the process to art and skill again.
He had an old HP rack-powersupply up on a shelf with multimeter probes hanging down from the output plugs.
The power supply was set to some wierd voltage with two decimal precision and CC-mode for a specific current.
When a phone arrived for repair, he opened it to expose the battery contact pads and put the probe tips against the pads and watched the voltage drop on the power supply.
He had learned that when it dropped to a specific level, he would lift the probes, count a few seconds then put down the probes again, and repeat until he saw a familiar change in how slowly the voltage drop happened.
Then he'd plug in a USB-charger into the phone and let it charge the rest of the way.
It just took a couple of minutes and was like watching a virtuoso at work.

And I can tell you this, he didn't wake up one morning and come up with the exact method, this was the result of a long line of failed attempts and discoveries of how NOT to do it.
The trick is as always to never repeat the same mistake again.

[Davi.p]

Re: Philips 60PFL8708S/12 - QFU1.2E LA - Error code 53

there was some chips on ebay many months ago.. yes my cpu can be popped you reminded me this.
So reflow must be made slowly by the bottom maybe..

[dominik0801]

Re: Philips 60PFL8708S/12 - QFU1.2E LA - Error code 53

In my opinion nothing can be better with reflow or quasi reflow than reballing, this is making me rivaluate what the man says about that resistor, in fact i've made a true reflow on it with all balls melting but the tv got worse, no more power up..

It could be that the heat aligned the electrons or microBGA inside the die in a bad way, it happens, and it isn't always a bad use of heat, some chips can be more sensitive than others.

Didn't your chip "popcorn"? I mean, its hard to know since the fusion has a lid on the top.

(btw, i know nothing about the resistor thingy, but does someone know anything about the fusion chips on aliexpress?)

[Davi.p]

Re: Philips 60PFL8708S/12 - QFU1.2E LA - Error code 53

In my opinion nothing can be better with reflow or quasi reflow than reballing, this is making me rivaluate what the man says about that resistor, in fact i've made a true reflow on it with all balls melting but the tv got worse, no more power up..

[dominik0801]

Re: Philips 60PFL8708S/12 - QFU1.2E LA - Error code 53

This last post is so boring and nothing useful... This cpu needs a serious reballing, also watch it by lateral vision when mounted on board, see if it or the board are bent.. Doesn't someone have tried the previously mentioned method of the resistor? Or is a fake?

Sorry that you found it boring and useless.

My last post was meant to help others to decide whether they want to try heating it or give it a proper reflow/reball.
Also an encouragement to put a little fan on the CPU heatsink. Mine went from 60c to 35-40c with a 5x5mm fan connected to 3.3v, which is barely audible.

Also, I shared all of this, because while you say the CPU needs proper reballing, other forums (and also another member of this forum, who also happened to post in this thread) says that not the balls, but the CPU itself is the cause of the problem, and it just needs HEAT to be revived (he states that 215c is perfect, because it won't melt balls, and the TV will still work due to heat shock).
I also read posts which say reballing is useless and/or makes things worse in some cases because of the thin PCB.

You see, there's really not just BGA balls to go bad with a flipchip design, but also microBGA under the die, and thin copper traces inside, that can get thinner over time due to electronmigration, can cause shorts or just go open.

Of course I am NOT against reballing, but in the end of the day, I would say a PROPER reflow is the way to go, since it won't just melt the balls, but also will give the chip a heat shock which can fix microBGA and electronmigration problems (even if the last two will not last too long, if adequate cooling is not applied).

[Davi.p]

Re: Philips 60PFL8708S/12 - QFU1.2E LA - Error code 53

This last post is so boring and nothing useful... This cpu needs a serious reballing, also watch it by lateral vision when mounted on board, see if it or the board are bent.. Doesn't someone have tried the previously mentioned method of the resistor? Or is a fake?