PS3 PS3 (CECHG) NLOD

jacobsson

jacobsson

Senior Member
I got a fat worthless PS3 CECHG from the interwebs which doesn't show a red led during standby. When pressing the power button I get the triple beeps just like a YLOD unit, except no lights are flashing in the front.

I used this unit to pratice a NEC tokin replacement (I'll soon receive a YLOD CECHA which is gonna get this treatment).
IMG_20191012_105927.jpg
B85ZN3T.jpg
The mod seems intact and have not changed the behavior of the unit, therefore I thought why not try to fix the real issue.

I anyone have any suggestion of what components to remove/check/measure?

Things I've tested so far:
  • Replacing the PSU with a known working unit
  • Replacing NEC tokins for the GPU and CPU (top side only)

Thanks!
 
Last edited:
sandungas said:
In SEM-001 the switches and leds are integrated in the motherboard (bottom-right corner in this photo)
https://www.psdevwiki.com/ps3/File:SEM-001_Bottomside.jpg

Have you took a look at that area?, maybe there is something fryed in it or in short circuit
Click to expand...

Thanks mate, I glanced over that area yesterday but didn't see anything suspicious. I'll make sure to measure the components for shorts or if somethings dead. I don't like the thought of binning it w/o giving it a chance.

By the way, I looked at some videos removing NEC Tokins, they use screwdrivers to pop them off in 10s w/o any residual on the solder pads. I tried the same thing but had so much residuals left from the bottom layer connected to the pads that it took an hour to properly remove (trying not to damage anything). Have you attempted this mod yourself?
 
jacobsson said:
Thanks mate, I glanced over that area yesterday but didn't see anything suspicious. I'll make sure to measure the components for shorts or if somethings dead. I don't like the thought of binning it w/o giving it a chance.
Click to expand...
I think it worths a chance because is a rare failure, sometimes rare failures are easy to fix
Btw, the front panel leds and buttons are connected with syscon, the only posible reason that could cause the standby red led to dont turn on are:
-led fryed
-resistor of that led fryed
-transistor of that led fryed
-syscon is not sending the signal to turn the red led on

The green led is not turned on either, so probably what is happening is syscon is not sending the signal, anyway, i made an schematic of that part of the circuit, is from a PS3 slim but the way how it works is the same
https://www.psdevwiki.com/ps3/images/0/04/SYSCON_SWx_JTP-001_JSD-001_HSW-001_CN101.jpg

Syscon sends an small voltage to the transistor, and the transistor connects the negative (-) pin of the led to ground
The resistor of the led is always located in his positive (+) pin
jacobsson said:
By the way, I looked at some videos removing NEC Tokins, they use screwdrivers to pop them off in 10s w/o any residual on the solder pads. I tried the same thing but had so much residuals left from the bottom layer connected to the pads that it took an hour to properly remove (trying not to damage anything). Have you attempted this mod yourself?
Click to expand...
I never did but i would not do it like that, i think is safer to touch it with a solder iron on top to overheat it entirelly
This would cause lot of residues and fumes, but is easy, and i would use tape all around to prevent the shit to spread all along the motherboard, then cleanup with alcohol, etc :)
 
sandungas said:
I think it worths a chance because is a rare failure, sometimes rare failures are easy to fix
Btw, the front panel leds and buttons are connected with syscon, the only posible reason that could cause the standby red led to dont turn on are:
-led fryed
-resistor of that led fryed
-transistor of that led fryed
-syscon is not sending the signal to turn the red led on

The green led is not turned on either, so probably what is happening is syscon is not sending the signal, anyway, i made an schematic of that part of the circuit, is from a PS3 slim but the way how it works is the same
https://www.psdevwiki.com/ps3/images/0/04/SYSCON_SWx_JTP-001_JSD-001_HSW-001_CN101.jpg

Syscon sends an small voltage to the transistor, and the transistor connects the negative (-) pin of the led to ground
The resistor of the led is always located in his positive (+) pin

I never did but i would not do it like that, i think is safer to touch it with a solder iron on top to overheat it entirelly
This would cause lot of residues and fumes, but is easy, and i would use tape all around to prevent the shit to spread all along the motherboard, then cleanup with alcohol, etc :)
Click to expand...

Thanks mate, always so helpful!
I'm gonna have a look at your list and schematic to check if any passive component is dead (hopefully).
Both the eject and power button works by touching them to turn the system on, so signals should be getting sent from the touch circuit at least.

About the NECs, there is actually a pretty good way that combines our methods, which is gonna be my approach for the CECHA01.
1. Pop of the plastic lid (very easy with a side cutter)
2. Peel off most layers of the NEC with a small screw driver or knife (tedious but easy)
3. Remove any residual isolating plastics from the NEC (minor PCB scratch risk), a lot less fumes for step 4.
4. Heat up the residual bottom layer of the NEC till it let go of the of the pad (need a lot of heat due to the ground plane)

The last part (4) is the part that was really risky using knife or screw driver, so this should be much safer using heat.
 
Last edited:
In step (4) you should also add lot of tin to create a big "blob" of tin in a side of the tokin (or both sides at the same time, not sure)
This way you can "insert" the solder tip inside the blob, this helps a lot to transfer the heat from the solder iron tip to the motherboard
It also allows the tin to conserve the heat for a few more seconds (this way you can move the solder to the other side)
And las reason (but this is probably the best) it allows the factory tin to alloy with the new tin you are adding, this changes his melting point, in my experience any tin you have at home is going to have a lower melting point than the one in the PS3 motherboard
 
sandungas said:
In step (4) you should also add lot of tin to create a big "blob" of tin in a side of the tokin (or both sides at the same time, not sure)
This way you can "insert" the solder tip inside the blob, this helps a lot to transfer the heat from the solder iron tip to the motherboard
It also allows the tin to conserve the heat for a few more seconds (this way you can move the solder to the other side)
And las reason (but this is probably the best) it allows the factory tin to alloy with the new tin you are adding, this changes his melting point, in my experience any tin you have at home is going to have a lower melting point than the one in the PS3 motherboard
Click to expand...

Yes, that is a very good point. The solder at home also has some flux mixture in it which helps a lot, I'm also an advocate of that you can't add too much flux (as long as you do a good cleaning after) when doing this kind of work, although more flux = more fumes.
I've started wearing a filtered mask too and not just ventilation.

Anyhow, I'll get back to you when I've got the time (hopefully later today after work) to troubleshoot my board :encouragement:
 
As a simple way of testing the SMD LEDs use ur multimeter and set it to diode mode and probe the + and - , u should be able to see the LED light up very faintly to see if its working, that way u can back probe from there. If for some reason ur DMM doesn't work in diode mode (not all cheap meters do) u can use a simple 1.5V battery which will do the same thing. it wont send enough current to destroy the LEDs so u will be fine.

With the tokin caps my method is basically what sangas has mentioned the 2nd time. my solder tip is a flat chisel type tho, if u can get one of these that will help a lot.

1. Pry the caps off like u stated
2. break away all plastic around the caps outer edges, small pointy nose pliers works best.
3. use the thinnest and smallest flat head screw driver u have, size 00 works best so u can pry the layers up slightly.
4. use same pliers to pull the layers right off and u will be left with the last ground layer pad and the positive pads still soldered to the board usually.
if u get to the ground layer pad u will have plastic between the ground pad and both positive points still.
5. use long thin tweezers if u have them to remove the last remaining thin plastic between these 2 points.
6. put 2 blobs of solder on the ground pad directly under each motherboard ground pads, one top and one bottom of the ground points.
7. use same tweezers to help lift the ground pad up from one side of 1st blob while u heat it, once its clear u can do the same on other blob and it will remove completely.
8. do the same blobs on both positive points to remove the positive pads, use the same tweezers.
9. This is important i find but u may not have it for every day use but use plenty of flux paste and get some solder wick to clean up all the remaining solder. this will create fumes from the flux so if ur worried use protection. it will get messy but nothing IPA cant fix to clean up the motherboard. Dont forget to clean both sides, the flux will go through the vias and onto the other side of the board.

Gezz after reading this back it seems like a lot of steps haha but its not really, i find its the best way to not get any plastic fumes but also get a nice neat job ready for ur new caps.
 
@wrx884 @sandungas
I actually did the probing in diode mode earlier today before reading your post.
The diodes are intact, so are the connecting resistors.

The transistors I couldn't really tell for sure, they had these very large resistance differences when measuring the different legs in combinations, which I understand is good, so I think they're fine too.

I had a look at all fuses I could find, they all had continuity.

The only thing I know I have to do is trying to measure the voltage rails after the psu, any directions for test points?
Other than that I guess my syscon is borked then maybe?
 
Last edited:
You should try to meassure voltage in the "base" pin of the red led transistor... in standby (there is no need to turn on the PS3)
Is a direct connection in between the base transistor pin and a syscon pin

And next thing you can check (without voltage on the motehrboard), is to meassure continuity in between this same base transistor pin and a syscon pin... just to see if the line is "broken" at some point... not probable but anyway

Next ting i would try to check is the voltage regulator of the schematic i posted before, that voltage regulator "feeds" syscon, it have 2 output voltages (and both are 3.3v if i remember right)
 
@sandungas Hi mate. Just wanted to say that I haven't had the time yet to check the voltages. I wonder though, the syscon on my board is on the opposite side of the board (right by the battery), correct?

Reference image:
s-l1600.jpg
 
jacobsson said:
@sandungas Hi mate. Just wanted to say that I haven't had the time yet to check the voltages. I wonder though, the syscon on my board is on the opposite side of the board (right by the battery), correct?

Reference image:
s-l1600.jpg
Click to expand...
I cant identify it in that photo, but should be located close to cell/southbridge/flash
And for a SEM-001 motherboard it should be the CXR713120-20XGB
https://www.psdevwiki.com/ps3/Syscon_Hardware
https://www.psdevwiki.com/ps3/CXR713120-201GB

Is like that photo of the first generation, this is bad news and good news at both times, is soldered by BGA so is tricky to check his pins and some of them could be inaccesible
But probably is the better documented of all the PS3 syscon familly, the keys/exploit published some weeks ago are for it
 
sandungas said:
You should try to meassure voltage in the "base" pin of the red led transistor... in standby (there is no need to turn on the PS3)
Is a direct connection in between the base transistor pin and a syscon pin

And next thing you can check (without voltage on the motehrboard), is to meassure continuity in between this same base transistor pin and a syscon pin... just to see if the line is "broken" at some point... not probable but anyway

Next ting i would try to check is the voltage regulator of the schematic i posted before, that voltage regulator "feeds" syscon, it have 2 output voltages (and both are 3.3v if i remember right)
Click to expand...

Hi. I had some time tonight so I measured the voltage over the transistor leg. during standby, I only saw a couple of millivolts so basically nothing (as inspected since the diods work). I found that the "left leg" (facing the ps3 normally) of the transistor had a via that connected to some adjacent passives of the syscon. I have to dig some more tomorrow and see if these passives are getting any voltage. If not, maybe one of the 3.3v regulators are dead as you said
 
@sandungas
this is what I get during standby and power cycle ("during pwr on"):
SUbsSbS.jpg


So there is at least one 3.3V to SYSCON present during standby. After many sequencial the CPU get hot, that's a good sign at least.
 
Last edited:
jacobsson said:
@sandungas
this is what I get during standby and power cycle ("during pwr on"):
SUbsSbS.jpg


So there is at least one 3.3V to SYSCON present during standby. After many sequencial the CPU get hot, that's a good sign at least.
Click to expand...
Question about the capacitors, I've read through another forum on this site that gives in detail the steps to replace the capacitors. But I thought that you only had to connect the capacitors with a wire if you replaced all of the capacitors on the board ?
 
You must log in or register to reply here.

Similar threads

Back
Top