Hardware PS5 Slim (EDM-140) – BLOD, RAM calibration / impedance error – diagnostics so far Error code: 80801510

  • Thread starter Thread starter slimmy182
  • Start date Start date
  • Views Views 1,593
  • Replies Replies 10

slimmy182

Active Member
Newcomer
Joined
Mar 23, 2022
Messages
35
Reaction score
1
Trophies
0
Age
36
Location
Italy
XP
230
Country
Italy
PS5 Slim (EDM-140) powers on with beep + blue LED for ~2 seconds, then shuts off. No second beep.

PSU / basics
• PSU tested standalone: 12.15 V stable
• 12 V remains stable on motherboard during shutdown
• All fuses OK, HDMI visually fine
• Board fully cleaned, liquid metal removed, no visible corrosion

Voltages during power-on
• RAM VDD/VDDQ: ~1.37–1.38 V
• APU core: ~0.90–0.91 V
• Rails appear briefly, then drop when console shuts off

UART 80801510 – RAM calibration / impedance fault on Chip 5

After a light reflow of chip 5 it briefly reported 80801F12 (RAM hardware fault), then reverted back to 80801510 consistently.

RAM work
• Chip 5 removed
• New same-model GDDR6 installed, same error (UART 80801510 – RAM calibration / impedance fault on Chip 5)
• Multiple reflows performed
• Original chip reinstalled for comparison, same error (UART 80801510 – RAM calibration / impedance fault on Chip 5)

Same UART error with original RAM, new RAM, and reflowed RAM

Measurements
• RAM VDD/VDDQ → GND: ~23–26 Ω, equal across all banks
• Diode-mode checks show one shared RAM reference-type net with significantly lower readings, present even with the RAM chip removed. A shared RAM reference-type net (many pins on each RAM + many nearby MLCCs; “Net72” on PHAT boardviews) measures ~0.040 V in diode mode to GND across multiple points, while other nearby nets read ~0.24–0.37 V.
• Thermal camera: all RAM chips heat evenly during the short startup window

Status
• RAM IC itself does not appear to be the cause
• Power rails are present during startup
• Suspecting shared RAM reference / termination rail or passive component issue

Looking for input
• Best next steps on PS5 Slim RAM calibration faults
• Anyone seen 80801510 persist after confirmed-good RAM replacement
• Tips to isolate faulty passives on RAM reference / termination rails
 

Attachments

  • mydiode.png
    mydiode.png
    1.1 MB · Views: 55
  • RAM_pcb_DiodeMode.png
    RAM_pcb_DiodeMode.png
    26.8 KB · Views: 39
  • 617374943_2291628811345547_8673640928527879209_n.jpg
    617374943_2291628811345547_8673640928527879209_n.jpg
    107.5 KB · Views: 34
Disclaimer : I'm not an expert at repairing PS5s but out of interest I took a look at the PS5 service manual.

Each DDR6 chips has 2 pads, ZQ_A and ZQ_B which are used for the impedance calibration.
DDR impedance calibration ensures signal integrity, reduces reflections, and matches driver impedance to PCB trace impedance to compensate for voltage and temperature variations.

https://www.levirepair.eu/infusions/forum/attachments/gddr6_gddr6x_2.jpg

On the EDM-010/020 each ZQ pad is connected to a 120 ohm 1% 1/16W resistor to ground.
On your board, I think those resistors are the 2 resistors next to each ram chip.

In the datasheet of the RAM chips, it is mentioned not to leave those pins floating.
So my assumption is that either the trace is bad, the resistor is not seated properly or the resistor got damaged by the heat of the ram.
 
Disclaimer : I'm not an expert at repairing PS5s but out of interest I took a look at the PS5 service manual.

Each DDR6 chips has 2 pads, ZQ_A and ZQ_B which are used for the impedance calibration.
DDR impedance calibration ensures signal integrity, reduces reflections, and matches driver impedance to PCB trace impedance to compensate for voltage and temperature variations.

https://www.levirepair.eu/infusions/forum/attachments/gddr6_gddr6x_2.jpg

On the EDM-010/020 each ZQ pad is connected to a 120 ohm 1% 1/16W resistor to ground.
On your board, I think those resistors are the 2 resistors next to each ram chip.

In the datasheet of the RAM chips, it is mentioned not to leave those pins floating.
So my assumption is that either the trace is bad, the resistor is not seated properly or the resistor got damaged by the heat of the ram.
Thanks for the input.

I’ve already checked the suspected ZQ termination network in detail:
• The two resistors next to each RAM chip measure ~120 Ω to GND (one side 0 Ω, the other ~120 Ω), identical to the same resistors on known-good RAM chips.
• Measurements were confirmed in-circuit and directly on the exposed pads/vias under the RAM footprint.
• Diode-mode and resistance readings on the ZQ-related nodes are consistent between the faulty chip (Chip 5) and a good chip.
• Visual inspection under microscope shows no lifted pads or poorly seated resistors.

Because of this, ZQ pins do not appear to be floating and the 120 Ω pull-downs themselves seem electrically intact.

The same UART error (80801510 – RAM calibration/impedance fault on Chip 5) occurs with both the original RAM IC and a replacement IC, which makes a bad ZQ resistor or missing termination less likely.

At this point I’m suspecting either:
• a footprint / via / inner-layer issue under that RAM chip, or
• another calibration-related signal (not the simple ZQ pull-down) triggering the error.

Open to further suggestions on what else to probe on that channel.
 
I did some reading up and it seems that the calibration is done inside the DDR6 module but it has no way of letting the APU know something is wrong.
The dram controller in PS5 APU does impedance testing on it's own and concludes that something is bad or out of the tolerated range.

Other things less likely are cracks in the traces or some issue with the vpp reference voltage which might lead the dram controller to conclude that something is wrong with the impedance.

The VPP reference voltage are pads V10, A10, V5 and A5.
They are connected together and connected with a capacitor to a voltage line.
The capacitor value is 1uF
The board you have probably has all those capacitors on the opposite side of the DDR module.
The small capacitors are all 1uf. The big ones are 4.7uF.
 
Thanks for the input, makes sense.

I checked the VPP side already: the 1 µF caps on the opposite side of the DDR are present, in continuity, and diode/resistance readings are consistent across multiple RAM banks, including the failing one. No anomaly there.

Same for ZQ_A / ZQ_B: both 120 Ω to GND, identical diode readings compared to other banks. Traces and surrounding passives also measure the same bank-to-bank.

At this point it looks less like a discrete RAM / VPP / ZQ issue and more like the APU’s own impedance training failing internally. Reworking RAM and verifying the external reference networks didn’t change behavior.

Open to ideas, but I’m starting to suspect IMC/APU rather than the memory side.
 
What about the K1 pad (VRefC) reading (in the RAM_pcb_reading.png)?
Compared to the other readings that seems to stand out.
It's a voltage divider to get a voltage like ~0.4v.
1770166046072.png


If I'm not mistaken, in diode mode that pad should read something like ~0.5–0.7 V

If it ain't that, then maybe something is wrong with the traces to the APU, pad under the APU or the APU itself.
Lifting the APU is a heck of job.
 
What about the K1 pad (VRefC) reading (in the RAM_pcb_reading.png)?
Compared to the other readings that seems to stand out.
It's a voltage divider to get a voltage like ~0.4v.
View attachment 554720

If I'm not mistaken, in diode mode that pad should read something like ~0.5–0.7 V

If it ain't that, then maybe something is wrong with the traces to the APU, pad under the APU or the APU itself.
Lifting the APU is a heck of job.
Thanks — I checked that. On my board the only net matching that VRefC divider behaviour is the one I measured at ~0.361V in diode mode and ~2.18kΩ to GND, and it’s consistent across multiple RAM banks (bank 5 and a known-good bank on the same board show the same readings).
The ZQ resistors (120Ω) also measure identical across banks.
So at least from DC/diode measurements, the VRefC divider and nearby passives look normal; the fault may be elsewhere (signal integrity / under-APU / internal APU memory controller domain).
 
from my side I can only say that I did not find obvious shorts or abnormal readings, but without a known-good board for comparison I cannot state with certainty that everything is within spec.
What I did verify is:
  • ZQ_A / ZQ_B termination resistors measure ~120 Ω on all banks, including the affected one.
  • Diode-mode and resistance readings on the backside components (caps and resistors opposite the RAM footprint) are consistent across all RAM banks, including the one reported by UART.
  • Replacing the RAM chip and reflowing does not change the error.
At this point, it would help a lot to know what values you normally read on a known-good board, specifically:
  • Diode-mode readings on the capacitors and resistors located on the backside of the board, opposite the RAM footprint
  • Typical resistance-to-ground values for VDD/VDDQ, VPP, and reference nets in that area
That would allow me to determine whether my readings are actually normal or only “consistent but wrong”
 

Attachments

  • IMG_5051.png
    IMG_5051.png
    45.3 KB · Views: 43
  • F9DF4926-16DE-4D4A-9C2F-91DCCF7396A7.jpeg
    F9DF4926-16DE-4D4A-9C2F-91DCCF7396A7.jpeg
    482.1 KB · Views: 32
  • B4273410-5A92-45FE-993C-484D0E4FE2E6.jpeg
    B4273410-5A92-45FE-993C-484D0E4FE2E6.jpeg
    336.2 KB · Views: 30
  • IMG_5057.jpeg
    IMG_5057.jpeg
    291.9 KB · Views: 27
  • IMG_5058.jpeg
    IMG_5058.jpeg
    243.7 KB · Views: 56
Sorry, I have no Slim boards at hand at the moment.

What is up with this capacitor?
Is it just dirt? It should look like the other ones.

1770697974987.png


Not sure if it helps but you can find the PS5 EDM-020 board service manual on Scribd.

All this stuff hasn't changed for the newer models.
 
Sorry, I have no Slim boards at hand at the moment.

What is up with this capacitor?
Is it just dirt? It should look like the other ones.

View attachment 555943

Not sure if it helps but you can find the PS5 EDM-020 board service manual on Scribd.

All this stuff hasn't changed for the newer models.
Thanks for checking and for the Scribd reference. I’m already using the EDM-020 service manual as a guide.

That capacitor was already inspected under the microscope and cleaned thoroughly with IPA.
What you’re seeing in the photo is just light staining, electrically it measures the same as the corresponding caps on the other RAM banks (same diode-mode and resistance readings), so it doesn’t appear damaged or leaky.

I’ve also verified:
• VRefC (K1) network is present and consistent across banks
• VPP rises to ~1.83 V at runtime on all banks
• ZQ resistors measure ~120 Ω on all banks

So far all reference and power rails look symmetric, even on the affected bank.

If you have any other specific spot you’d suggest probing or comparing (signal-side rather than power/reference), I’m open to ideas.
 
PS5 Slim (EDM-140) powers on with beep + blue LED for ~2 seconds, then shuts off. No second beep.

PSU / basics
• PSU tested standalone: 12.15 V stable
• 12 V remains stable on motherboard during shutdown
• All fuses OK, HDMI visually fine
• Board fully cleaned, liquid metal removed, no visible corrosion

Voltages during power-on
• RAM VDD/VDDQ: ~1.37–1.38 V
• APU core: ~0.90–0.91 V
• Rails appear briefly, then drop when console shuts off

UART 80801510 – RAM calibration / impedance fault on Chip 5

After a light reflow of chip 5 it briefly reported 80801F12 (RAM hardware fault), then reverted back to 80801510 consistently.

RAM work
• Chip 5 removed
• New same-model GDDR6 installed, same error (UART 80801510 – RAM calibration / impedance fault on Chip 5)
• Multiple reflows performed
• Original chip reinstalled for comparison, same error (UART 80801510 – RAM calibration / impedance fault on Chip 5)

Same UART error with original RAM, new RAM, and reflowed RAM

Measurements
• RAM VDD/VDDQ → GND: ~23–26 Ω, equal across all banks
• Diode-mode checks show one shared RAM reference-type net with significantly lower readings, present even with the RAM chip removed. A shared RAM reference-type net (many pins on each RAM + many nearby MLCCs; “Net72” on PHAT boardviews) measures ~0.040 V in diode mode to GND across multiple points, while other nearby nets read ~0.24–0.37 V.
• Thermal camera: all RAM chips heat evenly during the short startup window

Status
• RAM IC itself does not appear to be the cause
• Power rails are present during startup
• Suspecting shared RAM reference / termination rail or passive component issue

Looking for input
• Best next steps on PS5 Slim RAM calibration faults
• Anyone seen 80801510 persist after confirmed-good RAM replacement
• Tips to isolate faulty passives on RAM reference / termination rails
Hi Slimmy182, were you able to solve the problem?

I had one with the same UART fault, but on chip 1, and I fixed it by reballing the APU. I hope that helps.
 

Site & Scene News

Popular threads in this forum