PS3 Fault finding YLOD with the SYSCON - First steps and Error reporting

[vyktormvmpay25]

Well after spending about 3 and half hours cleaning bridged contact pins I'm about ready to send this board to someone and have them solder everything on or just buy another nand clip and use that.

My joints just aren't up to the task of soldering on that fine a level anymore.

Luckily the pins look clean but I'm going to use my DVM and make sure none of them are bridged in the morning.

Anyone got a nand clip they wanna sell me? [emoji23]

Sent from my SM-G965U using Tapatalk

As I said many times you should have tools ,anyway focus on top where legs of nands start from ic, most of time I had microshorts on top of there is kind of unfinished suport and had always shorts there.Is quite difficult to see without microscope camera.
Check very carefully that white area.

 

[lovetohax]

Use amtech gel flux, it makes all the difference. Don't worry about bridges, a small amount of solder comes away with your clean tip each time you touch the blob. So all you have to do is clean the tip, touch and repeat until the bridge clears. There is no danger of screwing up. Just use flux and keep at it until you get it right.

Stay calm and carry on!

Have you tried ChipQuik? I tried Kester and MG Chemicals that come in a bottle and they both suck. But I hear ChipQuik is good.

My current favorite is Amtech, but I am exploring other options.

 

[Pacorretaco]

Btw yeah I tested different board (SEM 001) and actually the 4v with the missing coils seems normal.

I guess it's time to look at the erm, tokins then. I was not really sure how the tokins could be responsible for 2+v. But maybe it's possible after all. Or a special kind of bad RSX. Or wait maybe the coils themselves?
I'll try some of those things now

 

[RCWD21]

Use amtech gel flux, it makes all the difference. Don't worry about bridges, a small amount of solder comes away with your clean tip each time you touch the blob. So all you have to do is clean the tip, touch and repeat until the bridge clears. There is no danger of screwing up. Just use flux and keep at it until you get it right.

Stay calm and carry on!

That's exactly what I did, along with a razor blade to scrape some of the larger build up away so it wouldn't just reflow around the pins.

I'm heading out here soon to get a better magnifying setup and I'll post back when I get it done.

As for the nand dumps should I post on here or in a separate thread?

Sent from my SM-G965U using Tapatalk

 

[vyktormvmpay25]

Btw yeah I tested different board (SEM 001) and actually the 4v with the missing coils seems normal.

I guess it's time to look at the erm, tokins then. I was not really sure how the tokins could be responsible for 2+v. But maybe it's possible after all. Or a special kind of bad RSX. Or wait maybe the coils themselves?
I'll try some of those things now

Probably I'm not right but if yes @RIP-Felix can you give right explanation? I'm not god at math and nor have time.

 

[vyktormvmpay25]

That's exactly what I did, along with a razor blade to scrape some of the larger build up away so it wouldn't just reflow around the pins.

I'm heading out here soon to get a better magnifying setup and I'll post back when I get it done.

As for the nand dumps should I post on here or in a separate thread?

Sent from my SM-G965U using Tapatalk

Ask @sandungas if he has time to check test, use some new ps3 check tools from github, pyps3tools, flowrebuilder 5.2 last one may work as well.

 

[RCWD21]

Ask @sandungas if he has time to check test, use some new ps3 check tools from github, pyps3tools, flowrebuilder 5.2 last one may work as well.

Will do! I was just reading about those last night actually.

Sent from my SM-G965U using Tapatalk

 

[Pacorretaco]

Yeah the coils are not the problem. A parasitic aluminum polymer capacitor was not enough to make difference either.
Which makes sense I suppose. This never looked like a typical tokin fault.

Now I should start removing tokins and if the problem persists... It should mean I have a special kind of dead RSX.

Because I think it's safe to say that the IOR regulators are not the problem if the 4v was normal?

 

[vyktormvmpay25]

4 v is normally because that power supply around cpu/gpu is calculated very well and it should vary depending how consumers will act. So as example if add under 2 ohms instead of 3 ohms power think its likely an short I think. Well the basic method is about the same. That 1.2v is an node of circuit compared with Ohm second law.
If I am right in further test I can test power supply by removing that coil and separate to get constant 1.2v. I do not suggest that to anyone without a lab power supply where it can see/set voltage/amps. Something may be fried or I may not be right.

 

[RIP-Felix]

Probably I'm not right but if yes @RIP-Felix can you give right explanation? I'm not god at math and nor have time.

I can't say I fully understand it either, but I have been trying. I'm certainly no engineer!

Here is a good article on 2 stage filter design, that pretty much sums up what SONY is doing here (I think). The voltage ripple in figure 2a is exactly the type of noise curve you get with bad tokins on the CPU. A sinusoidal curve with HF spikes corresponding to switching events. This artical and that YT video I posted before are quite a good introduction to how this filter works. The article has the math laid out if you want to try calculating it. I did this awhile back and it's...well, it's just hard to know if a calculation makes sense or not, since there are a bunch of non-intuitive concepts like Q, resonance frequency, and feedback loop bandwidth, etc. A while back I used an oinline RLC calculator to "attempt" to figure out the resonance frequency of the second stage filter, and It was just hard to know what kind of numbers I should be getting. I did get numbers in the 100-500KHz range, which I though might make sense, but I should examine this again, to see if it makes more sense now that I've been staring at this stuff for a while. Alot of it still seems like gibberish, but I understand more than I used to.

EDIT: Here's a diagram I made to hopefully make it easier to figure out. The full diagram on the left is the CPU. The RSX is similar, but with one fewer VRM nodes.

 

[vyktormvmpay25]

Yes that fig 2 in pdf is quite similar our case, forgot to add feedback from that point of 1.2v to switching power . I will try more with time to simple get to a feedback of 1.2v without ic(separated from power line by that coil. But it may not work without signal generator of specific Hz to keep opened ior and mosfet. It takes time and will be same results if board is not working properly even after reball I will pass on next.

 

[RCWD21]

Alright! I took all day to do it but I have the teensy soldered to the nand chips for the dual nand configuration.

All that's left is the 3.3v source from the motherboard! Can someone help me with that? My brain is shot from such miniature work all day [emoji854]

Edit: I still have the voltage regulator that you solder to the Teensy but if I remember correctly that isn't enough to power both nand chips???

Sent from my SM-G965U using Tapatalk

 

[RCWD21]

Ok so I have a regulator set at 3.3v and have that supplying to the 5v pin on the teensy like in the wiki, but do I need to solder across the 3v pads on the teensy since the 5v trace has been cut? Or can I solder the regulator that I got to solder onto the teensy and run both nand chips with that? I don't want to supply power yet and cook a chip so I'm awaiting a confirmation on here first...

Sent from my SM-G965U using Tapatalk

 

[RCWD21]

Alright, I'm confused again. I see on the wiki that there's only 15 wires per chip. Then in the NORway master zip I see there's a vcc and a ground on the nand chip connections. Do I need those or is the board being in standby mode giving power to those pins?

Also I can't locate the port that the teensy is on in dev manager I think because the virtual serial driver keeps giving me an error but teensy.exe works and recognizes the teensy as being connected..

Sent from my SM-G965U using Tapatalk

 

[RCWD21]

So far I've confirmed that the teensy is being recognized by the computer since I can program the nandway dual boot hex file to it.

Now when I try to use WAY-launchers v2.04 by littlebalup using the info command I get the error "Teensy not found. Command aborted."

Any idea what to do next? I'm about to pull my hair out as I've been at this all day and the board is as ready to go as I can get it but now programs aren't wanting to work correctly [emoji19][emoji58]

Sent from my SM-G965U using Tapatalk

 

[vyktormvmpay25]

So far I've confirmed that the teensy is being recognized by the computer since I can program the nandway dual boot hex file to it.

Now when I try to use WAY-launchers v2.04 by littlebalup using the info command I get the error "Teensy not found. Command aborted."

Any idea what to do next? I'm about to pull my hair out as I've been at this all day and the board is as ready to go as I can get it but now programs aren't wanting to work correctly [emoji19][emoji58]

Sent from my SM-G965U using Tapatalk

Ok it may not have time those days but I will try to help you with this if you would like to open another thread about this subject to say Downgrade with teensy2++?It's a different time zone and I will post in the next day, had my nights free time, now work flow starts to grow for me and you must be patient.

 

[RCWD21]

Ok it may not have time those days but I will try to help you with this if you would like to open another thread about this subject to say Downgrade with teensy2++?It's a different time zone and I will post in the next day, had my nights free time, now work flow starts to grow for me and you must be patient.

I'll start a thread in the morning, I'm gonna take a break from this for a day and clear my head.

I'll call it something along the lines of Cechb01 teensy troubles

Sent from my SM-G965U using Tapatalk

 

[sandungas]

I realized about something handy related with the syscon UART testpads, if we order them in groups based in his location on the motherboards they matches exactly with how we grouped the testpads for the hardware flashers using the concept of "layouts", are the same groups that appears in this list:
https://www.psdevwiki.com/ps3/Teensy++_2.0#Schematics_by_motherboard_.28retail.29

There are 6 syscon UART testpads layouts:
Layout 1 = COK-001, COK-002, SEM-001 (PS3 fat, mullion syscon, NAND flash)
Layout 2 = DIA-001, DIA-002 (PS3 fat, mullion syscon, NOR flash)
Layout 3 = VER-001 (PS3 fat, sherwood syscon, NOR flash)
Layout 4 = DYN-001 (PS3 slim, sherwood syscon, NOR flash)
Layout 5 = SUR-001, JTP-001, JSD-001, KTE-001 (PS3 slim, sherwood syscon, NOR flash)
Layout 6 = All PS3 superslim models (the testpads are very close to syscon chip in all them)

And is not a coincidence, the reason why they matches is because that concept of "testpad layout" we has been using unofficially represents the "jig-pins" used officially... you know... it should be some kind of frame with tenths of pogo-pings (i guess made of metal or with a 3D printer) and a thick cord with tenths of wires going to a PC
If at some point they needs to interact with a different motherboard they disconnects the "jig-pin" frame for "layout 2" and connects other for "layout 5" and attachs it to the motherboard like a sandwich

-------
@RCWD21 yeah, if this talk about the flasher extends too much is better to create a new thread... at this point you are not sure if is going to extend too much but anyway... as an intro...
The point is... when you "inject" the 3.3v externally you are doing it by soldering the wire into a "3.3v rail" that powers the flash chips but also some other components of the motherboard around, in some cases the voltage regulator that can be soldered on the teensy could not be enought, most people using teensy with PS3 motherboards with NAND was powering the whole thing externally with a ATX PSU (the orange wires)

You can solder the 3.3v wire in several places, directly to the motherboard... or in the teensy
But remember... this should be the only voltage active in the PS3 motherboard, in other words, dont power the PS3 with the PS3 PSU... actually the best thing you can do is to remove the PS3 PSU

 

[Garlic Bread]

I have a YLOD ps3 viewing the syscon log I found the error A0801002 I corrected it by changing the rsx capacitors.

I noticed that the tantalum sensors in the back were getting hot, so I rearranged them (it worked), then I did the same for the ones next to the rsx

now I have the error A0093004 so I think it could be a fuse

 

[lovetohax]

I realized about something handy related with the syscon UART testpads, if we order them in groups based in his location on the motherboards they matches exactly with how we grouped the testpads for the hardware flashers using the concept of "layouts", are the same groups that appears in this list:
https://www.psdevwiki.com/ps3/Teensy++_2.0#Schematics_by_motherboard_.28retail.29

There are 6 syscon UART testpads layouts:
Layout 1 = COK-001, COK-002, SEM-001 (PS3 fat, mullion syscon, NAND flash)
Layout 2 = DIA-001, DIA-002 (PS3 fat, mullion syscon, NOR flash)
Layout 3 = VER-001 (PS3 fat, sherwood syscon, NOR flash)
Layout 4 = DYN-001 (PS3 slim, sherwood syscon, NOR flash)
Layout 5 = SUR-001, JTP-001, JSD-001, KTE-001 (PS3 slim, sherwood syscon, NOR flash)
Layout 6 = All PS3 superslim models (the testpads are very close to syscon chip in all them)

And is not a coincidence, the reason why they matches is because that concept of "testpad layout" we has been using unofficially represents the "jig-pins" used officially... you know... it should be some kind of frame with tenths of pogo-pings (i guess made of metal or with a 3D printer) and a thick cord with tenths of wires going to a PC
If at some point they needs to interact with a different motherboard they disconnects the "jig-pin" frame for "layout 2" and connects other for "layout 5" and attachs it to the motherboard like a sandwich

-------
@RCWD21 yeah, if this talk about the flasher extends too much is better to create a new thread... at this point you are not sure if is going to extend too much but anyway... as an intro...
The point is... when you "inject" the 3.3v externally you are doing it by soldering the wire into a "3.3v rail" that powers the flash chips but also some other components of the motherboard around, in some cases the voltage regulator that can be soldered on the teensy could not be enought, most people using teensy with PS3 motherboards with NAND was powering the whole thing externally with a ATX PSU (the orange wires)

You can solder the 3.3v wire in several places, directly to the motherboard... or in the teensy
But remember... this should be the only voltage active in the PS3 motherboard, in other words, dont power the PS3 with the PS3 PSU... actually the best thing you can do is to remove the PS3 PSU

Are there more images for the test pads for the other Syscon solder points?

I feel like these should be added to the official guide.


Sent from my iPhone using Tapatalk