Im pretty sure you will get ylod in future. Can you replace tokins?
PS3 PS3 Undervolting Thread
- Thread starter raidriar
- Start date
Ganjo
Member
Hello everyone!
I got a CEHC04 (~11 days runtime) COK-002 and I have soldered Syscon wires for a nice, permanent setup in order to test values around.
I was able to undervolt it from CELL ? (didn't bother to check with a multimeter) to 1.0875V and its bin 3 toshiba RSX from 1.3125V to 1.05V.
Tested many games for hours such as GT6 demo, TLOU, MGS4, Yakuza (PS2) it ran perfectly fine without any freezes. Temps stayed under 55C on both chips during heavy gaming, with the RSX being cooler with 28-30% fan speed.
Now sadly whenever I turn on the console on a fresh new day after it has not been powered on overnight, I get YLOD in form of 1601 and 1701 whenever I launch any ISO. Turning it back on again lets me run the ISO again for hours with no issues. Sometimes it takes 2 times, sometimes once
I tried raising the CELL to a known stable value I once used (1.1125V), but it sadly still did it. So as a last resort, I changed RSX to 1.15V (which I knew was stable) and as of now, under the same testing conditions it loaded the ISO fine first try....
Has anyone met this issue before? I can only find one reddit post about it kind of unanswered [emoji17]
I guess its the RSX being kind of too low when loading a game from coldboot not being able to process the sudden "jump" on load so I will raise it up to 1.075V or 1.0875V and see..
Yes, I was gonna post about this. I have done a lot of these recently and found that a perfectly stable system that can run GTAV for 10+ hours over multiple days of testing can fail after a day or two of being turned off (power switch off). Its happened on 3 systems so far and its always the RSX that causes it. 2 of 3 were under 0x27. 2 of 3 had TOKINs replaced.
This is what I have in my log of systems. The first 2 are 40nm the 3rd is 90nm
| 1 caused a system to need a video reset anytime it was removed from AC power |
| 1 caused a ylod after days of being stable. No error codes RSX needed to be changed |
| 1 caused ylod after sitting. Rsx 3c/1c too low after working fine. 801701 |
WailkyS5M5
Member
Maybe? I have a JCD 8898 there is a possibility im able to solder in tantalizers but I have never done that before which makes it sound hard to me
Rn I set the rsx to 1.0875V I'll see if that is stable enough, if it works why bother replacing them now rather until they fail?
Oh shoot that could make sense, I always turn off the AC power switch on my cechc when Im done using it for the day. Thought it would be something healthy to do but I guess the capacitors discharge quickly there?
Should I just leave the standby on most of the times? At some point overnight therr wouldn't be any as I always cut the power off of my entire setup
So you are telling me that 2 of 3 consoles that have these symptoms already have new capacitors? That's kind of weird, is it just some random unstable thing with the undervolt or does it lead to something serious?
Last edited by a moderator:
WailkyS5M5
Member
An update, 1601 1701 was momentarily fixed with rsx 1.0875V it didn't do it after evening till mid-day without power
But when I used it after a night, a full on day and half a day it did it again, even when starting irisman
So its definitely time relevant. I have no Idea what to possibly diagnose?
But when I used it after a night, a full on day and half a day it did it again, even when starting irisman
So its definitely time relevant. I have no Idea what to possibly diagnose?
WailkyS5M5
Member
Try it out running any app or game after it being without power for 1 or 2 days to be sure
TriForcer
Member
Tried for 20 hours (off grid) and it started without problems. But I'm going to leave it for a few days as I've working on another PS3. But if it's OK for about 3 days, can I leave it at these voltage levels? Do I need to take into account that the NEC chip/CPU/GPU ages and steps up a bit?
WailkyS5M5
Member
About this, my PS3 now shares the same symptoms like yours with the prompting video reset after its been not plugged to AC for a while. 1601 1701 on starting apps/games didnt occur anymore
What changed it is I removed the stupid underclock gimmick I had where xmb would run below stock values and games/apps at stock values.
Was this PS3 fixed by reverting the undervolting, like is it still alive or dead? (because of the symptoms)
DoublesAdvocate
Member
I've been testing mine over the last few days and it doesn't seem like I have much room to undervolt. My CECHA has about 5000 hours of use so it could be that the RSX has worn down with age (no idea if this is true or not).
I also decided to measure the voltages on my 25XX and 43XX to see what sort of voltage the later revisions ran at, I was surprised to find the 28nm RSX was running at almost the same default voltage as the 40nm (-37mV) There's only -179mV between my 90nm and 40nm so maybe I'm expecting too much, or maybe I just have a large amount of space to drop the default and still keep it stable.
I also decided to measure the voltages on my 25XX and 43XX to see what sort of voltage the later revisions ran at, I was surprised to find the 28nm RSX was running at almost the same default voltage as the 40nm (-37mV) There's only -179mV between my 90nm and 40nm so maybe I'm expecting too much, or maybe I just have a large amount of space to drop the default and still keep it stable.
WailkyS5M5
Member
Thats a nice table, but if I understood it right I would like to update on my CECHC, it runs absolutely fine on both 1.0875V. My lowest was 1.0875 CELL and 1.05V RSX which were stable on many games, but that gave 1701 issues whenever loading anything on a fresh new day. 1.0875v on both still didnt fix it up until I stopped underclocking the RSX when on the xmb as that got rid of the issue
Tested after 7 days no power and game load fine with no YLOD and no video reset on startup
DoublesAdvocate
Member
My theory on those 1601 and 1701 errors was actually that the TOKINs are cold when they've been left for a day or two and thus can't perform as well when still warming up for the first time, I'm curious if your console would still have the same issues if you turned it on, then allowed it to warm up for 30 mins before trying to start a game.
WailkyS5M5
Member
Could be, but symptoms were not present anymore after I stopped underclocking the RSX on XMB which was a gimmick eitherway (-1°C difference).
First time after removing that underclock when booting up the console it would have an extra beep as in prompting a video reset by itself. After that it didnt do it anymore and si left it without power for an entire week and tried the same thing that gave me issue ans everything loaded fine with no YLOD or video reset
Had it at 400mhz ish something on both for xmb, and stock for in game. I think the console gets a type of "shock" whenever doing this jump paired with undervolt
So now it's running absolutely fine at CELL 1.0875V RSX 1.0875V [emoji16]
TriForcer
Member
Stock clock on CPU/RSX/VRAM
Stable in GT6 (loop demo mode) and TLOU.
Disconnected from the mains for at least 3 days and started up again without problems:
Stable in GT6 (loop demo mode) and TLOU.
Disconnected from the mains for at least 3 days and started up again without problems:
- CECHA01 CELL: 1,0750V [20], RSX(40): 0,9125V [07]
- CECHB01 CELL: 1,0750V [20], RSX(40): 0,9125V [07]
- CECHC04 CELL: 1,1125V [1E], RSX(40): 0,9000V [27]
- CECHE01 CELL: 1,0750V [20], RSX(40): 0,9125V [07]
- CECHA01 RSX(40)
- CECHC04 RSX(90)
- CECHC04 RSX(90)
- CECHL04 CELL(65), RSX(65)
DoublesAdvocate
Member
Did you happen to measure the default voltages for these 4 units? I'm looking to add that info to my list so we can get a rough idea of the voltage ranges set from the factory.
TriForcer
Member
For CELL I currently only have an assumption on these PS3's (1.25V) at 'FF'
On RSX I have looked from what I can read out of LV1 (I don't know if this is the right way to do it?) and readed BEFORE undervolting.
So I have NOT read the bits on the BUCK controllers or actual values on BE_VDDC or RSX_VDDC, on these:
CECHA01 (NOS RSX):
rsx40 a01 500/650 vpe:ff shd:3f [N3N174-11:0:4:b:9:11:6:0:2][24:0:e:0:1:0:1][1:1:0] --> 24 --> 0,975V
have tantalum + mlcc (direct on board)
CECHB01(NOS RSX):
rsx40 a01 500/650 vpe:ff shd:3f [PMH234-15:0:4:f:8:16:6:0:1][23:0:a:0:1:0:1][1:1:0] --> 23 => 1.0V
have only tantalum (TantalACEr)
CECHC04(NOS RSX):
rsx40 a01 500/650 vpe:ff shd:3f [N4W563-08:0:4:8:19:d:6:0:1][1e:0:a:0:1:0:1][1:1:0] --> 1e => 1,125V (!)
have only tantalum (TantalACEr)
CECHE01-MG (NOS RSX):
rsx40 a01 500/650 vpe:ff shd:3f [PMH234-15:0:4:f:9:d:6:0:1][24:0:a:0:1:0:1][1:1:0] --> 24 => 0,975V
have only tantalum (TantalACEr)
I'll try to get some mlcc in there, since there's room for them on the TantalACEr-board. I just don't know if it's a hassle to solder them in there directly or if it's easier to solder the entire ACE board off? I can solder - but this is so small...
Something that cross my mind is that the one I reported earlier (CECHC04 90/90nm/stock NEC-tokin, 103 days) is stable at CELL 1.0625V [1], RSX 1.075V [20] and that I started up (after 2 weeks off grid), and now runs on GT6 loop demo mode without any problems. Could it possibly have something to do with the extra filtering on CELL/RSX and that the MLCCs are important for stability on low voltage, as the NEC-tokin (when they work) are really really good? ...or maybe I just won the silicone lottery on this machine...
On B/C/E I have a jumper on the RSX voltage regulator, which I don't really know what it does yet?
Haven't had time to check it out yet (see picture).
The ones I still haven't undervolted yet, I can measure the bits on the BUCK controllers and/or actual values on BE_VDDC or RSX_VDDC before undervolting. (when I get the chance)
On RSX I have looked from what I can read out of LV1 (I don't know if this is the right way to do it?) and readed BEFORE undervolting.
So I have NOT read the bits on the BUCK controllers or actual values on BE_VDDC or RSX_VDDC, on these:
CECHA01 (NOS RSX):
rsx40 a01 500/650 vpe:ff shd:3f [N3N174-11:0:4:b:9:11:6:0:2][24:0:e:0:1:0:1][1:1:0] --> 24 --> 0,975V
have tantalum + mlcc (direct on board)
CECHB01(NOS RSX):
rsx40 a01 500/650 vpe:ff shd:3f [PMH234-15:0:4:f:8:16:6:0:1][23:0:a:0:1:0:1][1:1:0] --> 23 => 1.0V
have only tantalum (TantalACEr)
CECHC04(NOS RSX):
rsx40 a01 500/650 vpe:ff shd:3f [N4W563-08:0:4:8:19:d:6:0:1][1e:0:a:0:1:0:1][1:1:0] --> 1e => 1,125V (!)
have only tantalum (TantalACEr)
CECHE01-MG (NOS RSX):
rsx40 a01 500/650 vpe:ff shd:3f [PMH234-15:0:4:f:9:d:6:0:1][24:0:a:0:1:0:1][1:1:0] --> 24 => 0,975V
have only tantalum (TantalACEr)
I'll try to get some mlcc in there, since there's room for them on the TantalACEr-board. I just don't know if it's a hassle to solder them in there directly or if it's easier to solder the entire ACE board off? I can solder - but this is so small...
Something that cross my mind is that the one I reported earlier (CECHC04 90/90nm/stock NEC-tokin, 103 days) is stable at CELL 1.0625V [1], RSX 1.075V [20] and that I started up (after 2 weeks off grid), and now runs on GT6 loop demo mode without any problems. Could it possibly have something to do with the extra filtering on CELL/RSX and that the MLCCs are important for stability on low voltage, as the NEC-tokin (when they work) are really really good? ...or maybe I just won the silicone lottery on this machine...
On B/C/E I have a jumper on the RSX voltage regulator, which I don't really know what it does yet?
Haven't had time to check it out yet (see picture).
The ones I still haven't undervolted yet, I can measure the bits on the BUCK controllers and/or actual values on BE_VDDC or RSX_VDDC before undervolting. (when I get the chance)
Attachments
Last edited:
TriForcer
Member
Should it be replaced with a "real" one or is this good enough?
No, it is fine. It can be applied to 40nm only
TriForcer
Member
Thanks
- Good to know!My previously reported CECHC04 has a Toshiba RSX if that could also be relevant for undervolting.
b08 500/650 vpe:ff shd:3f [AP0006043:1:1:c:8:d:2:f:2][16:3:1:1:1:3:1][0:0:0]
Last edited by a moderator:
DoublesAdvocate
Member
Hmm okay. I'm still in the early stages of collecting data on this but I'm reasonably certain that the VID values found in the LV1 dump can't be directly used to measure the actual voltage provided to the RSX, or at least they don't actually match the values found in @RIP-Felix 's voltage chart. I'm not sure how many people paid careful attention to that video but he says that the Voltage/hex values listed in his spreadsheet are just taken from the closest public equivalent since the controllers seem to be a Sony specific part.
Just as an example I can point to some of the ones I've measured on my consoles.
eg. I took measurements on 2 consoles of mine (a CECHA and CECHE), one measured at 1.185V and the other at 1.190V, but both have an LV1 VID of 39 which should provide 1.2250V
Similar threads
-
PS3 Abnormally high temps on newly delidded ps3 slim !!!!
- Started by Jinno365
- Replies: 4
-
PS3 Interesting PS3 Errors (1802, 1701, 1601) and Capacitor Replacement- Started by Cheshire UK
- Replies: 2
-
-
