ฝัง
- เผยแพร่เมื่อ 11 ก.ย. 2024
- There is one piece of important information i left out: it is very important that the 1.8V rail does turn on when attaching an E-Power to the memory like i did because as per GDDR5X documentation the chips can die if the 1.8V VPP rail is lower than the VMem rail.
Thanks for watching!
Comments aswell as Likes/Dislikes are appreciated.
====================================================================
Most of what i do was inspired by Buildzoid and his Channel Actually Hardcore Overclocking
/ @actuallyhardcoreoverc...
====================================================================
Music Used in my videos includes Destiny OST as well as Satisfactory OST and tracks from the TH-cam Audio Library
![Asus ROG Strix 1080ti graphics card Repair [Computer won't power on] #nvidia #gpurepair](/img/n.gif)
great video, also love the shout-out for Tech Cemetery, his videos are very helpful
Waww what a video great content there, Whens the Funeral? id like to attend
youre one of the best, thats for sure, the experience youve gain is, i can say is awsome, ahve saw just like 10 videos but is enough to tell that if someone needs advices, can watch some of your videos... just needed to see when you in 03:30 talk bout resistance on Vcore power stages... gonna keep following your videos
hope you the best!!!
Thank you for taking the time to rerecord this video.
Haha thanks you helped me a lot. Got a gtx 1070 with the same read errors on every memory module, thought it's due to memory not getting power. Now I know that the memory controller is dead
It's odd, but comforting, to see my exact card PCB used in this. Card is Gigabyte GAMING GeForce GTX 1080 Ti 11 GB, by the way.
My 6-pin is shorted, looks like I had a burn out when my PSU blew. Component 0504 blew (anyone know a way to get a replacement??), which is other side of the board as the upside down "MAGIC 1RO" module on the lower right hand corner of the board, which was tested at 6:20.
Q504 is one of the P-Channel current steering mosfets. the one next to it is identical. it seems to be an excelliance MOS Corporation EMB20P03V P‐Channel Logic Level Enhancement Mode Field Effect Transistor. should be relatively easy to replace with hot air. funnily enough the card might actually work with that thing shorted since all it does is balance the 12V inputs. so worst case it just shorts together all your 12V which isn't ideal for power balancing but it should be enough for making sure nothing else on the card is damaged (do this at your own risk of course)
i have rx 560. it wont boot after restarting windows. and makes pc freeze when idle (sleep monitor) any ideas?
quite interesting how company with such good vlsi design and manufacturing divisions have subpar reference pcb engineering, and how they approve of designs that cut corners even from their borderline passable ones, no?
on related note, if your mosfet driver and/or controller ic does not provide good over current protection (ie. immediately drives low stage when detecting shorted high stage) there is enough charge stored in typical power stage that no psu can save your core
just a small comment. I have 1080ti mini a mats showed same write errors on all chips. Which is weird. And i found out it is basically because of the version. I tried version 400 of mats and that one showed issue with 2chips. Which after checking was problem of wrong resistance values on the memory side. And card was fully back.
I get nothing on the 3.3v rail on my gt 730. Like, no reading at all, not a low reading, nothing. I reversed the probes and went out of and into continuity mode several times. In resistance mode I get 6.5 k ohms measuring to ground.
got a gigabyte 1080ti aorus eith 0 power no damage only source of heat i see is on a 3 pin transistor lableled u551 on pcb and the ic is labled aa6k cant find a pdf got an idea what i can replace it wort
Thanks, bro
Where do I send my card. Error 43 but gpz z reads all data. I think it's a blown resistor somewhere that is causing my issue
Hello. I have a STRIX 1080 Ti AC origins ed. that crashes. I noticed in afterburner that the gpu and memory ramps up and goes down almost immidiately. The Hz goes from like 400MHz to 5500MHz on the memory. Got picture and all but all of the sudden it crashes. Any idea on what it could be?
I have replaced all of the pads and paste without any success. Also reflashed the vBIOS. No luck.
Zoras Tech, I just plugged an 8 Pin CPU Cable into my EVGA GTX 1070 and it smoked. Is this repairable? I am not sure how it fit in plug?
depending on what made the smoke, yes or no. Not sure exactly on the pin layouts but you might have reversed ground and +12 (or whatever cpu voltage is) and fried some of the components. Though I think the mosfet drivers might have been cooked and not the GPU itself. Just measure the resistances on the mosfets and replace them if necessary. but other things could also be damaged too...
I have a 980TI that displays but with artifacts. All the voltage rails are solid except for 1.8V. I actually get 1.59V on the memory phase, which I dont' think is right. I need to run MAPS but not sure I'll get a reliable result until all the correct voltages are present. Without correct voltages, especially to VRAM, it's going to give errors, which is misleading. Also, I have some RAM modules that run much hotter than the rest (can a shorted RAM pull down the 1.8V rail?) so maybe a short. I'm going to remove the suspect RAM and see if I get an improvement. I really wish I knew if I should be replacing a voltage IC...or if there even is one.
GDDR5 has an internal 1.8V supporting rail. Your 1.8V rail is just for the bios chip(s). GDDR5 runs on 1.6V VMEM. The Memory Module thar gets hot is likely faulty and will error out in the MATS test. Run that test to confirm which one it is then replace it
@@CmdrSoyo Thanks for the advice. I verified with Tech Cemetary that 900 series doesn't have a 1.8V rail, so between the two of you I have narrowed it down to I have vram that maybe toast...like all of it. I've never seen a gpu with all bad vram and look this good. And read-only errors is even more confusing.
Hy bro maybe you can help me i have rtx 3090ti waterforce gigabyte and I do one mistake my socket fan i put opus of pins and fan is stuck 100% load
i have rx480 nitro+ i got 86 ohm on 3.3volt rail and gpu don't get hot , no power in memory supply and controller and gpu
I have Zotac 1080Ti Amp Extreme with burned components. If anyone can tell me name/model number of those components I would be highly thankful.