PS3 Overheat problems & CELL delidding

My PS3 FAT has nearly 10c difference between CELL and RSX but overall i can keep CELL/RSX at acceptable limits with reasonable fan speeds. I can play all games including the most demanding ones. So i have decided to delay the surgery. Patient feels ok for now so no urgent need to put in under a knife. Maybe next year, i will think about it.


@sandungas - A question. Is it possible that Temp sensor of CELL becoming faulty? Though its highly unlikely. But can the Temp sensors of GPU and CPU give wrong readings? These sensors are also electronics after all. I know that these kinds of sensors can go bad in cars.
 
Last edited:
My PS3 FAT has nearly 10c difference between CELL and RSX but overall i can keep CELL/RSX at acceptable limits with reasonable fan speeds. I can play all games including the most demanding ones. So i have decided to delay the surgery. Patient feels ok for now so no urgent need to put in under a knife. Maybe next year, i will think about it.
Heheh, yes, is better like that :)
The only payback you have is the bigger noise because you need to use higher speeds... so is ok

If at some point that difference becomes bigger (lets say, over 15ºC), then you should starting worrying
Personally i think something in between 10-15ºC is the limit... more than that could be dangerous because the effect is the DIE heats very violently (because it cant trasnfer well the heat to the IHS)... and that violent changes of temperaure could damage it internally
But i dont really know if what im saying is true... is mostly speculation... but obviously is not designed to work like that... so is not good

For PS3 FATs like yours (and well... all PS3 in general) that have the factory thermal paste under IHS since years ago there is something that needs to be considered when talking about this

Lets say... the paste under IHS when was applyed at factory had a performance of 100%
After several years of use... lets say the performance has been reduced to 80%... this is something normal, as expected
But the important thing is the thermal paste under IHS in both CELL and RSX should have the same degradation

So... the difference of temperatures in between CELL/RSX the first day the PS3 was turned on (with the paste new from factory at 100% performance)... should be mostly the same you have now after years of use with the paste degraded to 80% performance

We know your PS3 have a difference of 10ºC now.... but we dont know how much it had when it was new... but overall we can deduce is not so bad
Maybe you have one at 80% performance... and the other at 70% performance
So... one of them have been degraded more than the other... but is ok, is still inside aceptable values

What is not aceptable is to have one paste degraded to 80% performance and the other to 40% performance
 
Wow i just came from work and i can see so much information. I think we all can agree that something is wrong. I will make a video of me dissassembling my ps3 so that all can see and maybe recognize some faulths. I will clean my board and the paste from the chips and add new paste.. i'm going to buy new paste today and hopefully i will have it today if not then i will buy it online. Good points were written here like : not enough pressure on chips and about the how to test temps.

Thanks guys for this exciting experience.
 
Last edited:
Well, the FATs were made 11 years ago. So you can imagine the condition of thermal paste under the IHS.

Wow i just came from work and i can see so much information. I think we all can agree that something is wrong. I will make a video of me dissassembling my ps3 so that all can see and maybe recognize some faulths. I will clean my board and the paste from the chips and add new paste.. i'm going to buy new paste today and hopefully i will have it today if not then i will buy it online. Good points were written here and especially about thet how to test temps.

Thanks guys for this exciting experience.

Take good photos of following and post here.

Your motherboard photo of side with CELL and RSX. And also your heatsinks photo. Take these Photos before cleaning and replacing thermal paste. We can check if your thermal paste was adequately spread and making proper contact with heatsinks or not.

You should use and test your old thermal paste to know the best applying method. And then when you know the best technique, you should use your new thermal paste after completely removing old thermal paste with IPA. Normally your IHS and heatsinks should be have thermal paste on full area of contact without leaving any pockets or corners. That will look like a thin layer of thermal paste on both IHS and CELL/RSX. Apply thermal paste on CELL/RSX and seat your motherboard. Now remove it back, and see if IHS & heatsink is fully covered with thermal paste or not. If you find that there are some pockets or corners where thermal paste has been left out, that means you need to reapply thermal paste in better way. If you find that thermal paste has been spread fully on IHS and is also overflowing too much out of IHS, that means you are applying too much paste. In case of old FAT, a slightly too much is better than too little.


To know more about PS3 IHS repair. Read this. Take note, that this is very risky procedure.

https://quade.co/2017/ps3-delid-ihs/
 
Last edited:
Well, the FATs were made 11 years ago. So you can imagine the condition of thermal paste under the IHS.



Take good photos of following and post here.

Your motherboard photo of side with CELL and RSX. And also your heatsinks photo. Take these Photos before cleaning and replacing thermal paste. We can check if your thermal paste was adequately spread and making proper contact with heatsinks or not.

You should use and test your old thermal paste to know the best applying method. And then when you know the best technique, you should use your new thermal paste after completely removing old thermal paste with IPA. Normally your IHS and heatsinks should be have thermal paste on full area of contact without leaving any pockets or corners. That will look like a thin layer of thermal paste on both IHS and CELL/RSX. Apply thermal paste on CELL/RSX and seat your motherboard. Now remove it back, and see if IHS & heatsink is fully covered with thermal paste or not. If you find that there are some pockets or corners where thermal paste has been left out, that means you need to reapply thermal paste in better way. If you find that thermal paste has been spread fully on IHS and is also overflowing too much out of IHS, that means you are applying too much paste. In case of old FAT, a slightly too much is better than too little.


To know more about PS3 IHS repair. Read this. Take note, that this is very risky procedure.

https://quade.co/2017/ps3-delid-ihs/
Roger that. In that case i will not go and buy a new thermal paste in a store but will order it online.
I will start in like 20 min with my ps3 journey. i will take photo's but also video and post it on youtube.
I just hope it is allowed to post links to other sources like youtube.
 
There is no need to make a video, but a couple of good photos of the heatsink and IHS surfaces before cleanin up the actual paste could be very handy to post here (with the titan paste you have now)

As @snkplkn mentioned keep an eye at how the paste surface has been "deformed" by the pressure, if it covers the whole square of the IHS (included corners), if there was some area not touching the other surface, etc...
There are a lot of things that can be deduced just by looking at that

Also... i mentioned you before that you should dissassemble the PS3 "a couple of times" to check this deformations of the paste surface several times (by trying different things) to see the better results
And for this tests you can use the titan paste... actually, you can reuse it several times... you know, by doing this the paste is degrading, but doesnt matters because at the end of the experiments you need to remove it with isopropil alcohol

Actually, you could do this tests with toothpaste


-----------
One detail some people misses is...
You know the PS3 motherboard is covered with 2 metal shields at the sides like a sandwich, right ? (and the heatsink is attached to one of the metal shields)

Well... that metal shields needs to be attached to the motherboard in a specific order, and following a specific movement
Keep attention at the border where is located the HDMI connector
In that border.... one of the shields have like some "holes"... and the other shield needs to be "inserted" in the "holes"
After that... you should "close" the shields (like closing a book)
Not sure how to explain it, english in not my native language... but the point is that border of the shields is like the "hinge" of the book

If you dont cose the metal shields like that... then one of them is deformed, and the heatsink doesnt seats correctly in CELL/RSX
Maybe thats what is happening to you... so check what im saying patiently
 
There is no need to make a video, but a couple of good photos of the heatsink and IHS surfaces before cleanin up the actual paste could be very handy to post here (with the titan paste you have now)

As @snkplkn mentioned keep an eye at how the paste surface has been "deformed" by the pressure, if it covers the whole square of the IHS (included corners), if there was some area not touching the other surface, etc...
There are a lot of things that can be deduced just by looking at that

Also... i mentioned you before that you should dissassemble the PS3 "a couple of times" to check this deformations of the paste surface several times (by trying different things) to see the better results
And for this tests you can use the titan paste... actually, you can reuse it several times... you know, by doing this the paste is degrading, but doesnt matters because at the end of the experiments you need to remove it with isopropil alcohol

Actually, you could do this tests with toothpaste


-----------
One detail some people misses is...
You know the PS3 motherboard is covered with 2 metal shields at the sides like a sandwich, right ? (and the heatsink is attached to one of the metal shields)

Well... that metal shields needs to be attached to the motherboard in a specific order, and following a specific movement
Keep attention at the border where is located the HDMI connector
In that border.... one of the shields have like some "holes"... and the other shield needs to be "inserted" in the "holes"
After that... you should "close" the shields (like closing a book)
Not sure how to explain it, english in not my native language... but the point is that border of the shields is like the "hinge" of the book

If you dont cose the metal shields like that... then one of them is deformed, and the heatsink doesnt seats correctly in CELL/RSX
Maybe thats what is happening to you... so check what im saying patiently
Ok i just started doing this.( I had visitors). I will make pictures
 
I saw that i had no cooling paste anymore. so i'm going to continue adding paste tommorow.
See pictures. Those cables at first image are in case i will need them in future. (sbe to ground)
 

Attachments

  • a.jpg
    a.jpg
    421.5 KB · Views: 174
  • b.jpg
    b.jpg
    548 KB · Views: 169
  • c.jpg
    c.jpg
    466.4 KB · Views: 194
  • d.jpg
    d.jpg
    621.9 KB · Views: 215
  • e.jpg
    e.jpg
    511.2 KB · Views: 164
  • f.jpg
    f.jpg
    669.2 KB · Views: 189
  • g.jpg
    g.jpg
    464.9 KB · Views: 219
  • h.jpg
    h.jpg
    642.7 KB · Views: 199
  • i.jpg
    i.jpg
    469.8 KB · Views: 184
Good photos, it can be seen the heatsink is seating very well and with good pressure in one of them, at left
That "peaks" and "rifts" at left spreaded along all the square is exactly what you would espect to find, in few words, lets say... is perfect

At the right we dont know if is seating well... because the paste was not well spreaded, also, it looks like some bits of the paste are like "water" (it can be seen is wet, but is not grey).. i guess the paste itself decomposed, maybe was the ending of the syringue
mlcZxhq.jpg


In the other surface is mostly the same... one is perfect, the other we dont know
So... you need to repeat the test... apply new thermal paste, make some temperature tests and if there is still a problem, open it and look at the paste again to try to see if it has that "peaks" and "rifts"
VvTTL8D.jpg


The weird thing is... in the photos the one that looks good is CELL... and RSX looks bad
But you said the opposite... you said CELL was 30ºC over RSX o_O ...so i guess you need to re-check that temperatures
I am now in the middle of opening my ps3 and redo the cooling paste. The reason for that is because i think i have put too much cololing paste. I have read from control fan utility cpu 85° and RSX 43°

Based in the images... it looks like it was the other way around... CELL was ok, and RSX had problems
In that case is nice, you have found the problem, the paste of RSX was bad :encouragement:
 
try to use a rubber like something to cover that small hole at the back of the second processor in picture number six, I try this in my days of using fat ps3 and it worked for me. who knows it might also work for you.
 
Good photos, it can be seen the heatsink is seating very well and with good pressure in one of them, at left
That "peaks" and "rifts" at left spreaded along all the square is exactly what you would espect to find, in few words, lets say... is perfect

At the right we dont know if is seating well... because the paste was not well spreaded, also, it looks like some bits of the paste are like "water" (it can be seen is wet, but is not grey).. i guess the paste itself decomposed, maybe was the ending of the syringue
mlcZxhq.jpg


In the other surface is mostly the same... one is perfect, the other we dont know
So... you need to repeat the test... apply new thermal paste, make some temperature tests and if there is still a problem, open it and look at the paste again to try to see if it has that "peaks" and "rifts"
VvTTL8D.jpg


The weird thing is... in the photos the one that looks good is CELL... and RSX looks bad
But you said the opposite... you said CELL was 30ºC over RSX o_O ...so i guess you need to re-check that temperatures


Based in the images... it looks like it was the other way around... CELL was ok, and RSX had problems
In that case is nice, you have found the problem, the paste of RSX was bad :encouragement:
You're right i was at the end of my syringe and when i saw your comment about that i was like "he knows"!
i will take a picture to show temps, that will be better. I will also clean all the board with isopropyl because it might be that there is paste somewhere and is causing a shortage that generates heat..
 
try to use a rubber like something to cover that small hole at the back of the second processor in picture number six, I try this in my days of using fat ps3 and it worked for me. who knows it might also work for you.
Thanks for the tip, i will do what you mentioned. It bodered me from the first time i opened my ps3.
 
You're right i was at the end of my syringe and when i saw your comment about that i was like "he knows"!
i will take a picture to show temps, that will be better. I will also clean all the board with isopropyl because it might be that there is paste somewhere and is causing a shortage that generates heat..
Hehehe, i mentioned it because in this photo can be seen the thermal paste had some problem
WeTJyz6.jpg

Rought explain.... basically the thermal pastes are composed by 2 substances:
-Some particles of "magic"
-Some liquid

I say "magic" in general, because are chemical formulas under patent, very complex, etc... i dont pretend to summarize how is made in a single sentence, so lets say is magic
And the liquid could be water, oil, or whatever... the liquid doesnt makes anything... other than keeping the magic particles together

But in that photo can be seen in the areas marked by the arrows there is only "liquid" (but no magic)... thats bad, is a failure of the thermal paste

-------------
There is no need to cleanup the motherboard with alcohol btw, i dont think that could cause heat
you should keep an eye at the thermal pads though (are that gummy squares on top of some chips) incase are missing or very weared out could cause problems, are not going to overheat CELL/RSX but eventually the chip that had the thermal pad could overheat and get damaged, so are important
 
Ok , i have bought a cooling paste from a store. They didn't had arctic or grizzly.. I bought the logon brand.
I cleaned my ps3 motherboard with isopropyl, i removed with isopropyl the cooling paste from the chips and heat sinks.
added new cooling paste, assembled my ps3 and when powering on and coming in the xmb menu i started hearing same old song..
fan started to speed up, then faster and faster,,,,,, after a minute or so i got message at top right ps3 overheating
CPU 85° RSX 46°
Before jailbreaking there was nothing like this.... after jailbreak --> overheating problem..
Should i desolder those two wires?? maybe the ground wire from the main power is giving ground through metal case that touches the sbe or wire? Should a jailbroken ps3 with sbe to ground active ,, overheat?

In my pictures i uploaded here i just want to show what kind and how much i added cooling paste. I tried to put a bit more then normal.
 

Attachments

  • 69173387_2385313718457390_9001171233781317632_n.jpg
    69173387_2385313718457390_9001171233781317632_n.jpg
    577.5 KB · Views: 168
  • 69289635_494175634483740_8722938850004959232_n.jpg
    69289635_494175634483740_8722938850004959232_n.jpg
    627.7 KB · Views: 167
  • 69342874_1220442304792399_1635169950249254912_n.jpg
    69342874_1220442304792399_1635169950249254912_n.jpg
    74.8 KB · Views: 145
That looks too much pase, and the way how you are applying it is not the best imo
The way i do it is pretty much exactly like this, a very thin layer of paste covering all the IHS + a tiny "rice" in the center
I do the layer the most thinner posible (almost transparent), and i spread it with a transparent plastic like butter (the transpart plastic allows me to see under it)
The tiny "rice" is not really needed but i do it because sometimes the IHS and heatsink surfaces are a bit curved or have imperfections (are not completly flat)
maxresdefault.jpg


The weird thing is you are still having CELL a lot more hotter than RSX, is starting smelling fishy
Open it again, check how the paste was spreaded by the pressure (the rifts and peaks i mentioned in the previous posts)
And after cleaning it... do a new test... try to grab the CELL IHS with your fingers and "pull" from it and try to rotate it a bit... if it moves this is a problem becaue it should not move at all

As i mentioned... at this point you can "recycle" the previous paste, there is no need to use new paste for every test because you seem to be at a point where you are going to need to repeat this several times
 
The way you applied thermal paste is not satisfactory. But now the Difference of 40c between RSX and CELL is indicating a different problem. There shouldn't be so much difference.

When you separated your motherboard and heatsink this time, did you see that your heatsink was covered with paste as well? Now i strongly believe that problem is not the thermal paste.

@sandungas Can DEX cause PS3 overheat? I don't think so. Can PS3 CPU sensors give wrong reading? Is it possible?

If i were you, i would do the following. Take out extra wires etc from console. Install thermal paste as @sandungas advised. See how Temps turn out. If its still hot, then delidding is the only option. I would also try to go back to cex if possible.
 
At this point i really think the CELL IHS has been moved in the past (maybe it happened when pulling from the heatsink because the CELL IHS was sticked very strong to the heatsink). And that movement of CELL IHS have "damaged" the paste under CELL IHS a lot

I mean... there is a thing we know for sure... the problem is HUGE, and is something that should look very bad
Is just we cant take a look under CELL IHS... but at this point you made 2 tests that points to it
 
At this point i really think the CELL IHS has been moved in the past (maybe it happened when pulling from the heatsink because the CELL IHS was sticked very strong to the heatsink). And that movement of CELL IHS have "damaged" the paste under CELL IHS a lot

I mean... there is a thing we know for sure... the problem is HUGE, and is something that should look very bad
Is just we cant take a look under CELL IHS... but at this point you made 2 tests that points to it
Ok before we gain pain in our brains (lol) i will be uploading in 30 min a picture here of how the paste looks like on the cpu and rsx as i'm dissassembling it. I will also desolder the wires. For this attempt i will put the normal amount of cooling paste.
Also it bothers me that the ps3 before jailbreaking was not having this issue..
Also converting the ps3 to CEX is now tough as the ps3 is giving me message that the ps3 is overheating.
This message appears after 20 seconds in xmb (after hearing the fan speeding up), making much noise
 
So surgery is the only option. Do not rush on it. Read carefully about delidding. There is plenty of information on internet. There are many videos also on youtube. But take note, that some people are doing it incorrectly and some methods of delidding are riskier than the others. You will need the right tool. And a lot of patience.
 
(maybe it happened when pulling from the heatsink because the CELL IHS was sticked very strong to the heatsink). And that movement of CELL IHS have "damaged" the paste under CELL IHS a lot

I don't think its possible. My PS3 was 11 years old when i changed thermal paste for first time. When i opened up PS3 and separated Motherboard from Heatsinks, it was easy. The old paste on IHS was completely dried up and crumbled....and it was not having any stickiness. So separating Motherboard from IHS was not a problem.

We are pretty sure that something is wrong under the IHS and original Sony paste under IHS is not working anymore. But i can't give any specific reason for that....except the fact that its too old.
 

Similar threads

Back
Top