Hello all, 
 
I purchased this 2080Ti back in November of last year. In February 2019 I purchased a full custom EK water loop focused on keeping this thing cool in a Small form factor case. When I originally put the custom water block together I was getting over heating issues with the GPU which made it stick to 1350 Mhz and perform terribly. After conversing with EVGA I re-applied the GPU water block and used the thermal pads that came with the card, rather than the ones that came with the EK water block and for the last 5 months my computer has been running well. 
 
However sometime in the last couple weeks something has changed, my computer now black screens with a [No signal] message shortly after doing ANY GPU intensive tasks. Path of exile, Stardew Valley, Heaven Benchmark, PCMark benchmarks, anything that puts load on the GPU seems to cause this black screen.
 
When it happens; my screens both shut off and the default monitor [Displayport No Signal] appears. All audio continues working, and I can perform actions and hear myself doing so in games. After 30 seconds the audio pauses for about 30 seconds and then recovers, sometimes with a Nvidia drivers crashed and recovered sound, sometimes not. But the displays never turn back on. I have let them sit for 3 hours to be sure and they never come back.
 
My temperatures are hardly ever above 65 degrees C on the GPU, CPU or fluid when this crash happens. 
 
The issue start:

• Windows 10 did not recieve an update that started the issue
• There were no changes to the nvidia control pannel
• There was no Nvidia graphics update that happened before the issue started.
  • I did try updating the graphics drivers to 430.86 AFTER the issue started as an attempt to fix.

 
Steps I have taken so far, NONE of these have solved the issue:

• I have checked EVGA precision for any new firmware
• I have set the power limit to max in EVGA precision to ensure it has enough power
• I have done a clean install of the Nvidia drivers, as well as removing EVGA precision and removing all precision profiles before the driver clean install.
• I have booted into safe mode, used the techguru DDU program to completely remove the Nvidia drivers and restart my system. Then installed the newest drivers from the Nvidia website 
• I have tried using EVGA Precision X1 to reduce the clock speed of my GPU by up to 500
• I have tried using only a single monitor on all displayport ports, as well as 2, and 3 monitors.
• I have tried using only an HDMI monitor.
• I tried doing a clean install of windows.
  • Did a complete format of my 500GB M.2 SSD and installed with a USB boot disk.
  • Previous Version was: Windows 10 version 1809 (Not sure of build)
  • New Version was: Windows 10 version 1809 (OS Build 17763.557) 
• I have tried cleaning the GPU block and replacing the liquid in the loop to ensure there wasn't buildup preventing flow for the GPU.
• I have tried 100% fan curves to keep temperatures closer to 55 degrees C, but the crashes still happen in the same timeframe. (My cooling setup is 3x120mm radiators for the single 2080Ti and 9900k, both stock clocked. with Noctua Industrial 2000 RPM high flow rate fans and a EK D5 Pump-res combo.)
• I have tried putting the stock cooler back on the card.
• I have tried re-seating the EK GPU block and adding more Thermal Pads to VRM and Ram modules.
• I have purchased a more powerful PSU and removed the overclock on my CPU to ensure it has enough power.
  • The PSU has 2 separate 8 pin power cords going to the GPU
  • The PSU uses Custom Cablemod created cables that were used by both the 600 watt, and the 750 watt PSU. The cables are the same cables that were working fine before, and I confirmed with Cablemod that the cables are the same pin outs for the PCIe cables for both PSU's

 
Why I think it's the GPU issue:

• I have plugged my monitor directly into the On board Intel graphics and this issue goes away. I only run at 7FPS, but I can play for an hour as just tested without the screen crashing to black. 

 
 
Link to 2 HWInfo logs before the black screens. The second tab (..._10-23) was run with a -300 clock on the GPU. (EDIT: Can't link to my google sheets to show info, and can't attach CSV or xlsx files so I'm not sure how to get the logs here... If a moderator could let me know how to share this info that would be appreciated.)
 
 
My Hardware Setup

• Motherboard : ASUS ROG Strix Z390-I Gaming LGA1151
• Graphics Card Model : 11G-P4-2382-KR
  • 2 separate 8 pin PSU cables going directly from PSU to GPU.
• CPU : Intel Core i9-9900k
• Memory : 16GB Corsair Dominator Platinum 3200 Mhz DDR4 x 2
• Power Supply : Corsair SFX 600 upgraded to a Corsair SFX 750
• Operating System : Windows 10
• Monitor :ViewSonic VP3268-4K, ASUS ROG Swift PG279QZ 27" WQHD 1440P IPS 165Hz
• Other: NZXT Hue 2, Schiit DAC USB audio
• Water cooling: EK 1x240 coolstream rad, EK 2x240 coolstream Rad, EK D5 Pump/res combo, EK 2080Ti GPU block, EK Intel CPU block, 
• Fans: Noctua Industrial 2000 RPM PWM High Pressure Fans, 2 x 120's 2 x 140's 
• Case Phantex Shift X
• Windows: Windows 10 version 1809 (OS Build 17763.557)

Thank you for the link to the starter thread, I apologize I did not see that before posting.
 
I have updated my post with all the requested information and will provide it here as well for your ease of viewing.
 
Windows Update

• Previous Windows: Windows 10 version 1809 (Not sure of build)
• Windows: Windows 10 version 1809 (OS Build 17763.557)
• Did a complete format of my 500GB M.2 SSD and installed with a USB boot disk.

PSU cable questions

• The PSU has 2 separate 8 pin power cords going to the GPU
• The PSU uses Custom Cablemod created cables that were used by both the 600 watt, and the 750 watt PSU. The cables are the same cables that were working fine before, and I confirmed with Cablemod that the cables are the same pin outs for the PCIe cables for both PSU's

Nvidia driver questions

• There was no Nvidia graphics update that happened before the issue started.
  • I did try updating the graphics drivers to 430.86 AFTER the issue started as an attempt to fix.
• The fresh installs I performed were done to the 430.86 driver version.
• There were no changes to the Nvidia control panel before the issue started.

 
Edit: Adding additional edited requested info
 
DP cables:

• My cables are certified
• These are also cables I have used successfully for well over a year since I had my 1080 previous to my 2080 Ti.
• Both display port and HDMI fails at the same time in the same way, leading me to believe it is not a cable issue
• I have just tested with other certified DP cables as well, same issue.

Monitors:

• ASUS ROG Swift PG279QZ 27" WQHD 1440P IPS
• ASUS ROG Swift PG278Q 27" WQHD 1440P IPS
• The mointors are showing up as Generic PnP Monitor in Device Manager
  • They show up as Anchor Communication... (Gsync Compatible) in nvidia control panel. 
• The issue persists with both monitors limited to 60 Hz.

 
Edit 2: Why I think it's the GPU issue:

• I have plugged my monitor directly into the On board Intel graphics via the same DP cable that was used in the GPU and this issue goes away. I only run at 7FPS, but I can play for an hour as just tested without the screen crashing to black. 

What happens is you remove power to your Monitor, for 30 sec & then plug it back in ?

Nvidia - Try Clean install of Only the GPU driver & PhysX
1) test older driver 419.67
2) and your newer driver 430.86


This is the best black screen guide I've found - good read & reference

How to fix black screen problems on Windows 10
"When your Windows 10 device is displaying a black screen, use this guide to troubleshoot and resolve this issue."
 
(excerpt 1)
On Windows 10, a black screen can happen for a number of reasons. It could be related to an issue with a recent graphics driver update or physical connection between the display and video adapter. Also, you may experience this behavior while upgrading to a new feature update or applying a new cumulative update. Or it can just happen out of the blue.
 
(excerpt 2)
you'll see a black screen on your device, because Windows 10 lost its connection to the display. When this happens, you can use the Windows key + Ctrl + Shift + B keyboard shortcut, which restarts the video driver and refreshes the connection with the monitor.

Windows key + ctrl + shift + B:

• Audible beep, but still no signal to the GPU.

Unplug Monitor for 30 seconds and plug back in: 

• No signal upon plugging back in
• No signal if unplugged for 30 seconds along with the display port cable and then both plugged back in.

Cablemod vs Stock cables:

• Tried using the PSU cables that came with the SFX 750, still same issue
• Tried using the SFX 600 separately with the stock cables that came with the 600, and running the graphics card off it's own PSU, same issue

Clean install of Drivers:

• Tried a clean install using DDU of driver 419.67, Same issue
• Tried a another clean install using DDU of driver 430.86, Same issue

Thanks everyone for your time on a Sunday.
 
I am going to start the RMA process, if anyone has any other ideas I'd still be happy to give them a try.

The RMA has been stress testing in my custom loop for the last 20 hours now. Temps are down ~5-8C compared to the previous card and I have had zero stability issues of any kind so far.
 
I was able to raise the power target and give it a bit of an overclock with no issues in any games or stress tests in the limited testing I did yesterday as well. 
 
Thanks for everyone's help, so far everything is looking good with the new card.