PS3 YLOD

[sumsum]

Hello rod905,

how can I post an image ... in here

you have to use a imagehoster like imageshack or anything similar.

If you have uploaded the image, you will get a link to it (or you have to search for it by yourself). The imageadress has to be in a code

Code Auswählen Erweitern

[img]http:\\www......jpg[/img].

[rod905]

can you people tell me if I am to scrap the ps or try and buy a new reballed rsx. trhank you sumsum

I think that at 1st glance I lifted 4 pads

[Takeshi]

Hoy have to clean the board more, there is to much old flux. But there are some solder pads broken. I don't know if they are needed.

[rod905]

to clean the board would be no problem, although there is no flux on it.  I think some pads lifted yes, and I don't know how. I lifted the board at 252deg cel. I don' t know if it is possible to try and fill the hole under the pad with solder. I guess I better dump everything and wait for the ps4

[Takeshi]

to clean the board would be no problem, although there is no flux on it.

There IS, I see it

I lifted the board at 252deg cel.

Too much.

I think (-> I don't know) all pads are only GND or main Voltage, so it's not the worst case.

[rod905]

so if you were in my place you would 'risk' buying a preballed rsx and fit it? (my rsx has popcorned during a previous reflow at 180 deg c).

I think if I had kept the temperature lower than 252 I would have pulled more pads because the solder would have been more solid, or no?

[Takeshi]

Yes, I would.

I only heat it up to ~230°C and it wirks fine. If you need 250°C, then the heat distribution was bad, you had hot spots.

[rod905]

Thanks Takeshi. I managed to find the pinout number and the respective schematic for this board so now I know where every pin connects and what it does. basically you have to short the pins skipping one. I will upload the images here for everyone to make use.

[rod905]

From the schematic and pinouts I discovered that from the 5 missing pads, the one in the centre is ground sense, it links to nowhere so Ill leave it as it is, the other 4 are vdd and ground. they need to connect with the one 2 spaces from each. I tested with the multimeter underneath a lens and found that this is correct (part of the trace should have been left in the 'hole' under the pad because I can get ontact to it after some trying with small gauge wirw attached to the probe. I will try to fix this issue with silver conductive paint (used to draw tracks).

My question is because I am really afraid of popcorning this bga that I am waiting for from ebay, and I have no way and means of prebaking it:

If I heat the motherboard to 210deg (gpu area from underneath mostly) then put the gpu in place, I will fix guides so it will go exactly in place (already did fix guides before I removed the old gpu so that I will mantain the exact location) will this work?

This way the gpu will never gets heat directly on top of it but the leaded balls at the bottom will make contact with the board that is at 205deg.Then obviously let it cool down by itself.

What do you think?

[rod905]

Here is an image of the board at the moment. I did clean it more like you (Takeshi)advised circled in red are the missing pads and the silver lines is to where they should connect. kindly advise on the method I mentioned before of heating only the motherboard and if this board is clean enough in your opinion or needs more

http://img339.imageshack.us/img339/6456/boardm.jpg

[Takeshi]

I wouldn't fix the pads with silver conductive paint, because it will work, even if the pads are missing. There are much more close to the missing pads. And if you try to fix them, it's possible, that you make mistakes and thereby a short. So leave it as it is.

If I heat the motherboard to 210deg (gpu area from underneath mostly) then put the gpu in place, I will fix guides so it will go exactly in place (already did fix guides before I removed the old gpu so that I will mantain the exact location) will this work?

Yes, the BGA automatic moves to the right place, IF the position is nearly correct.
And I would put the BGA in place and then heat the board. You should heat the BGA directly, otherwise the contacts could be bad. If the BGA ist preballed with Pb-Solder, then the zinn melts at ~183°C and you don't need more then 210°C. 200°C up to 210°C would be good.

To prevent popcorning, heat up the mainboard (without BGA and flux) up to ~130°C and hold this temperature for an houre.

Now the board looks great

[rod905]

Hi Takeshi what I was asking is:

Would it work if I heat the mainboard to 200deg then put the BGA on it and let it settle ther? Therefore the contacts of the preballed BGA will touch the mainboard which is at 200 and melt Nad the BGA will never get heated over directly. I just heat the mainboard from bottom and when it reaches 200 I put the BGA (RSX) on it and never heat the BGA.

[Takeshi]

As I said before, I wouldn't do it like this. It coukd work, but this way isn't really good.

[rod905]

To prevent popcorning, heat up the mainboard (without BGA and flux) up to ~130°C and hold this temperature for an houre.

I was saying I am afraid of the RSX popcorning not the mainboard.

and another thing, is there a way to check if an RSX is good or not before installing?

[Takeshi]

I don't think you'll get popcorning of the RSX. If you have popcorning of the RSX, your temperature from the heatgun is to high. If the RSX is mounted with Pb-free solder (original), you have to heat to 220°C ore more. The risk of popcorning is high. But if you use Pb-solder, you only need about 180°C (better 200°C) and then the risk is lower
If you don't heat the RSX directly, it's possible that the temperature of the solder is not consistent and very soon you may get the YLOD again.

To check the RSX, look at this guide first. By measuring the resistance between the the power supply and ground, you can see if there is a shortage. Now if you follow the conductor, you get the pads where the power supply and the ground is.
By doing this you don't absolutely know, if the RSX is OK, but if there is a shortage, you definitely know that the RSX is broken. But I never had a RSX with "good" values which didn't work.