Great Video, I will add time stamps with some comments: 02:20 disassembly and visual inspection 07:55 get a multimeter (use resistance mode) 08:50 why not to use continuity mode Vcore measurement (ground to Vcore capacitors) 10:12 check 12 V power lanes (ground to first three connector pins should show several kOhms, 4th is ground; power connectors on pins should show very high resistance of k-MOhms to mass) 13:48 check 3,3 V power lane (last three pins should show very high resistance of MOhms, fourth pins from the right 400 Ohms, fith from right is ground) 14:14 power lanes to VRMs (voltage regulator modules) output Vcore ["+" side of capactiors to corresponding 12V pin (usually PCIe slot for Vmem and 6/8pins for Vcore, find out by going from the slot or pins to the input inductor or big capacitors) or input inductor or a big capacitor in front of the Vcore/Vmem; several kOhms expected] 16:55 power lanes to VRMs (voltage regulator modules) output Vmem (same strategy as for Vcore) 17:45 AMD VDDCI (memory controller power) VRM 19:42 VRMs to ground (very low resistance of down to 600 mOhms to Vcore "end" capacitor, little higher of around 20-200 Ohms to Vmem) 22:12 no short detected - power on card (2 possibilities: no picture or artifacts (both usually BGA or silicon damage); check voltages when card is powered on; "bake" card using heat gun but not in the oven!)
because these are stuff you only find out when enrolling to electronics courses and engineering programs. these are not some basic stuff that you could easily understand and apply, there are a lot of pre-requisites before you can understand the basics (an example is understanding diodes, what are the physics behind it?, etc, like that). Kudos to Zoras Tech for making videos like this that are not very simple.
@@reinerabad3482it's not that hard to learn, just casually watching videos about these topics will get you knowledge, you don't need to be an expert to tell if something is not how it's supposed to be
Please never give up. i got in 1 video so much information RESPEKT before i was reading in the internet for Days to figured the Stuff out. Bin wirklich beeindruckt
Hey man, you got a sub. I do electronics repair, but these cards are alien technology compared to what I normally work on. Thank you for breaking it down, saves me a ton of time!
I love ze German accent. Awesome video. I also watched the video where you explain the components of a GPU PCB. What I'm missing and would love for you to make is the missing link between those two videos. What points do you need to jab with your multimeter to test what component and what resistance are you expecting to see on a good card? If you find no flaws with resistance testing and power the card, what points are you then testing to test the various components and what voltages are you expecting?
Hey man, it's very nice to see another small tech channel which is actually that good. Your content is on pair if not better than Buildzoid's, you should have 1000 times the subs you have. I especially liked the washing suggestion, even though I guess it was obvious seen my name and my channel lol. Keep it up!
@13:10 It looks like one or more capacitors are getting charged when you measure, therefore the resistance rises. Measuring applies a small voltage across the probes.
Yup, as they get charged the current through them drops and so the resistance reading increases (cause they are reaching the point of being an open switch for dc voltage)
Wow. Perfect. You're the first to really hit the nail on the head with common BGA damage. I've been fixing (attempting) 980TIs and many of them have the same memory issue and sometimes VRM issues. I think the root-cause is bga failure, which leads to other failures upstream, offen showing up as blown inductors and mosfets. You fix the short upstream but the real issue downstream on the GPU Core or VRAM itself. I think cards with large cpu cores and big power rails that support them just get too hot and thermal cycling is what kills the card, starting with BGA failure. So many "repaired" cards stil show artifacts or will show them within days after the so-called repair. Lead-free solder is just not good for this type of cycle as it's too brittle and get worse over time. I used to fix Plasma TVs and solder failure from thermal cycles is the #1 failure point. Most repairs involve refreshing solder joints with good quality leaded solder and then they run forever. Now to plan to do some serious reballing with leaded solder and hope resurrect a few decent GPUs.
How about random black artefacts that don’t take up whole space on screen and disappear quickly last 2 seconds. There are no green lines. Some times they take 20% of screen space. Like black areas (large pixel effect) is this GPU issue? I have RTX 3090
Great video. I have a 1080ti that shorted something from a water fitting leak on the back of the PCB. I tried briefly to measure the voltage rails but the whole card practically looks like a short but that's probably due to the low core resistance as mentioned here. Trying to bone up on some GPU knowledge with videos like this so I can take another look.
When you are testing for a short on a mosfet or a pad under a removed mosfet,what is the lowest reading In ohms that you would receive and still say it is not shorted? This is without considering the masters shorted reading? Thanks for shooting videos from start to end with explanations.
Depends what part you are measuring. Newer GPUs like 2080Tis have practically 0 ohms of core resistance (older cards sit around 0.3-5 ohms). Memory VRM usually have 20-100 ohms. 12V power planes should be several kiloohms and so on
Hey thanks for the video, really appreciate it. Would it make sense to apply some pressure to the GPU while heatgunning so that the BGA gets better contact?
i've never put any pressure on the GPU while reflowing. sounds more like an easy way to create shorts by shifting the GPU on accident. the weight of the GPU itself should be enough. plus the solder also has surface tension when liquid
@@CmdrSoyo Wish I saw this earlier my chip got shifted and now it won't start, not even the fans spin but the led on the shroud lights up. Is my video card done for good? Amazing channel by the way keep uploading great informative content. Subscribed
what if i inspect the card visually and i find something missing on the back? i find pictures of the PCB with the component on , it looks like a capacitor (very small) but it has no code next to it , how do i find a replacement ? or what steps should i take to fix my problem ?
Awesome video! Information on this topic is so hard to find. So I’ve got a card that doesn’t do anything. RGB does come out but fans don’t move, no video and no detection in Windows. The PC will boot. I’ve inspected the entire board, tried reflowing / resoldering joints where possible (capacitors and 8pin power) and used an 858D knock off to heat up the vram and BGA but so far no dice. I’m trying to learn how to apply the multimeter to test capacitors and mosfets because it seems to be a power issue. Any thoughts on how I could continue diagnosing? It’s a 5700 XT that I bought in late 2019, shame it died so early. Trying my best to find what limited videos are available out there on the topic.
Hi. I can say that you are very experienced technician re gpu repair but you still lecture as if we are all kids here watching your video and that is very helpfull. Could you pls explain in another video how a rtx 3070 could become shorted if a riser it was attatched to shorted one of its capacitor. It has two 12v power connector and one of them is shorted. If i detach the shorted one the system would turn on but the gpu is still not detected. Pls explain what to check and how. Im in the philippines and i am willing to acquire equipment to fix this gpu because i also have rtx3080 with same issue and exactly same cause of failure.
You mentioned "i did not heat gun it long enough" grabbed my kids old gtx 650 ti and dropped 375 degrees on it for about 15 minutes, you can see the top of the solder go from a dark meatal color to a bight silver sheen on top, that's when I turned it off and let it set. it works now!!! Do you still work on GPU's I got a HD 7950 I wanted to send to you. it had a fan controller issue and I fried something on it and cant figure it out.
I have tried but have failed!! Could you get a cheap video card (10-20) on ebay, and show the steps and show what you could do that would be wrong or get wrong readings--like checking for shorts, ohms setting or diode, black on ground or red, checking mosfets. I have ruined several cards just trying to get started!!!!! Thanks
Thanks for a great video. Good lighting, super well explained and actually useful. My 1080Ti died and I'm waiting on the RMA to come back at the moment. In the mean time I put in my old GTX 580, which worked for about 2 weeks before it also died. That's got me worried my PSU or motherboard could be killing GPUs. The 1080Ti wouldn't boot at all, no display output. The GTX580 does have display output but the PC will reboot when I try to install the drivers. Sounds like a VRM issue on the 1080Ti and a BGA issue with the 580 right? Do you think it's just coincidence both cards failed close together considering the 580 is about 8 years old by now? Both the PSU and Motherboard were top of the range back in 2011 but they're also very old by now. What do you think?
It's unlikely the defects are related. You could try the heatgun method on the 580 and see if it starts working again. I had a 580 once that did the same and it was fine after i heatgunned it. Your PSU is likely fine. After all those yreas maybe a but full of dust but that's fixable by just cleaning it. The only thing i can think of that could degrade over time is the capacitors which might cause some higher coltage ripple but that shouldn't kill components. May cause some crashes from time to time if it gets really bad at worst i think
@@CmdrSoyo Sweet as, I figured it was probably just bad luck for one and old age for the other. Puts my mind at ease a little so I can wait a little longer for the Ryzen 5600 non-x if it ever comes out. Didn’t have any dust or anything either. I think the PSU was so over specced the fan never even runs when it’s set to semi passive cooling. Thanks for the help!
Wow you are really knowledgeable. Help me please. I have a vega 64 that I know something has shorted my psu cuts power when I start up games in 4k resolution. The strange thing is that benchmarks complete. But a few games trigger it every time. Can you point me in the right direction of where you might check first or maybe help me please? Or would you possibly like the opportunity to fix it for your channel? That would be quite cool actually.
@@Swc_Leader I think that would be easier for you to get help from youtube videos. I did it watching videos and worked for me. Just search "Flashing AMD Radeon bios" and you wont believe how easy it is
@@dearestgonzo ok I have a few questions before you flashed the bios where you getting black screen? Or werw you already able to see picture with your GPU it was just the artifcating/freezing issue you have to fix?🤔
I am trying to diagnose a GPU, (980 ti) which does not display any output despite the power planes being fine. The LED's and fan turn on just fine however no video output is detected. There are no signs of visible corrosion or signs of shorts using the techniques in this video. My assumption is that the core itself is dead, how should I go about identifying if this is true? Any suggestions?
Hey awesome videos, love it. So i went and measured resistances on the major power plane and nothing came up short for what I can tell, I also just started this today. So I powered on the card and checked if the GPU die was getting warm, it was not. The BIOS chips on the back was getting hot. I think my VBIOS chip might be dead. Can I replace that? Also, I measured voltage on the PCIe pins and was only getting 0.3V when powered, so maybe I'm not measuring correctly. Maybe I need to check the minor power components? Thanks.
Where can I find out more about the "12V plane"? Where do you find this info about card layouts? I can't find specs and what measurements to really expect. I've got a general idea but, I'm troubleshooting a 1080 GTX that has no obvious burns ... but still not sure. I'd like to know what resistance values to expect on that specific card if that exists? Thanks for the video either way.
Hope you can answer my question, I have a r9 270x that powers on, but I doesn’t display, checkin the resistors to core they give me 9-12 ohms, that would be a short? Or it’s normal? It’s a xfx model
Hello! I'm trying to research if my water-damaged RTX 3070 is repairable. The card runs stable but is power throttling. GPU-Z shows high power draw from one 12v and virtually none from the other, TDP at 190%. The back of the PCB showed corrosion around the power connectors and some nearby components before cleaning with alcohol and QD cleaner. Does this indicate any specific problem? Is it salvageable? Thank you.
Thank you, this is a great video. I have a radeon r9 390 I used for many years, but it developed this silicon / BGA damage as you describe. I'm hope to get it working again.
I have a evga 980ti with the blower fan style I love those looks just not a good setup .. it turns on shows no video and fan sounds like its on high the chip gets hot the board shows no sign of anything getting hot it was put in a oven buy the signs on it im getting what looks like a short on the pcie 3 volt power can u go over proper diag of a card from pcie
No microscope zoom, you only touch on the power rails or what components are involved with a circuit, and we can't see the contact points. We have to take your word for where the multimeter is set and where your stabbing, while you admit mistakes in the first video. I do not want to discourage you, but electronics is a subject that does not lend itself to poor equipment, or brief explanations. The students need to see, to hear, and to understand to have confidence. How about drawing it on paper? I believe that would help. We can see that.
30:23 really ? Put the gpu in the water : P Btwn I had some issues with my gtx 660 ti it freezes and displays a white color on the hole screen or pink or other color and before 15 days it stopped working what do you Think should Wash it in water or use a heat gun or something ? Pls answer
My RTX 2080ti, the fan spin but it doesn’t have power the lights are off..My PC is water cooled so the radiator dripped water on it and that’s how I found, no power. I took it apart cleaned it. I bought a new thermal paste do you think I should replace the thermal pads? Hopefully it will turn back on.
Very nice and helpful video.... I had a card that was artifacting tried the heat gun method but when I plug it back in it the mb starts to beep constantly.. No end... Any ideas?
Hi , I am learning. I have a HD7950 card with black screen. The core was getting hot so I had a heat sink on it and did not leave it on too long. I was cleaning up the controller chip solder joints and now when I plug it in there is no heat and non of the power stages have a voltage or a signal with the oscilloscope. At least I had .9 volts before. Any thoughts? Thanks
bro can you plz guide me i have an sapphire rx 570 its light blinks on startup abd then goes off nothing happens no video no power would it be short or main chip
Hello, my PC toppled over and the motherboard and GPU seem damaged. The GPU sometimes take signal sometimes doesn't. Is it just simple PCI-E connector issues or more serious? Motherboard only decides to boot with 2 but not 4 ram sticks. What to do? 😔
I just wish at least the high end cards would be constructed in a more industrial manner, you know like replaceable modules of the VRM, memory sticks that can be switched for new ones, like a motherboard. I hate the "this is consumer electronics" way of engineering of these.
at 10:54 your measuring the resistance on the pci express? what if you dont get a resistance on that side of the pci where you poked and only get resistance on the far right side on the long rail... sorry if my questions stupid lol
0 ohms on 12V PCI-E means it's shorted. Make sure you are actually probing the 12V pins though there are some ground pins in there which will always show 0 ohms (look up a pin out on the internet if you're not sure)
Hey, i have a question, the first fuse that you mention, what amperage does it has? i need to replace it but i'm not sure about it's amperage, thanks for ur video!
i don't know the exact value but i know the value from another card. there it used a 20A (2x10A in parallel) fuse for an 8pin and a 10A fuse for the PCI-E slot. it should work fine if you just copy that
My 970 recently fluked after some sparks on some small resistors/capacitors at the back of the board. Since then it keeps hard resetting at the windows logo i also notice temperature in the vrm/inductor area is shitass hot. So i open it up and there's only a little bit of tearing in the painting on the pcb between vrm mosfets and 2 specific inductors. No blow marks, smoke cracking on vrms/inductors or anything tho. Resistance readings seem fine, i've also checked major spots he mentions in the video. Any ideas on what i'm missing?
I have a card, it's video output ports are not working, but it works fine as laptop eGPU using internal display (with nvidia optimus), what could be wrong with it?
i have red devil rx5700xt .. when i turn on my pc , all component are working but my gpu fan is not spinning , and no display (gpu rgb on) ... is there a chance that my gpu shortt?
I have a short in the power connector showing. MOSFETs are good from the readings I get. My fuse is ok too. What would be the most likely issue for my card? Gigabyte GTX 1060 6GB. Lightning strike on my house took it and some other electronics out.
Short on a power connector usually means 12V is shorted to some output. But if you bring the lightning strike into the picture it might also be that there was some very high voltage sent into the Card and may have popped some capacitors or other components
I have a gtx 980 ti when I checked over for shorts and everything looked fine when I installed and turn on it powered up the lettering on it lights up fan sounds like its on high and no screen output I get nothing
Hey thanks for the vid!l so helpful! I’ve done one attempt to use the heat gun on my 780 ti which is artifacting. But when I started to heat it up there seemed to be liquid coming from under the chips and core? Is that right? as I backed off and tested the card but didn’t make it any worse but still the same should I keep the heat on it longer?
Help! the one of the "towers" close to the VRMs is detached from the VGA board! It happened because i accidently snapped it while trying to dissamble the VGA. The VGA now runs but it shuts down if i put loads like games and stuff... Please help.. Cant afford another VGA while the prices are a bit crazy
Hi thank you so much for your videos firstly! I have a video card (gtx 1060 6gb) that was plugged in and my house was struck by lightning and I know the card was shorted because no it has no display, BUT the fan runs at max speed. Any idea if it is fixable? Thank you so much in advance! I am German also by the way, but live in the states greetings!
Please help! My RTX 2070 just started showing artifacts. Does same thing on 2nd computer, so appears it is a hardware issue. When measuring resistance from the capacitors to the pin on the 8pin slot, I get a measurement from all capacitors except 1, which is an open line. Does this capacitor need to be replaced? All of the other capacitors start at 0 and increase quickly increase by hundreds of Ohms.
Bro I have a the Same Gpu. When i Start it, its fan works. But when i Install any gpu drivers, it shows black screen. I cant use my pc. Only if i enter with safe mode and delete those drivers, my pc works. Can I get solution? Mine GPU amd R9 270x
So, when you find a card, and all the info that is given in the description is "Not working" and you ask for more info about it and their reply is "read the description", Do you skip it?
i would skip it yeah. i usually only go for cards that artifact so i can just heatgun them for a bit and forget. i may try some proper repairs one day but for now i just get cards to heatgun
Hi, I have a 2060 that seems to be dead. Motherboard doesn't even pick it up, just shows an empty slot. This happened while I was experimenting with risers for the first time, the 1x adapter sparked at the motherboard and the card hasn't worked since. There's no visible damage to the pcb. I've tried poking around with the fluke, but it is my first time and is a bit daunting. Any ideas for specific areas of the pcb to check?
Since you saw a spark on the PCI-E slot you should look there. If the card has a 12V fuse it is likely blown. It could also have damaged a VRM connected to it. It could also be that the motherboard was damaged so try the card in a different slot or in a different motherboard.
@@CmdrSoyo Thank you for the comment! I have ruled out the motherboard by testing with other cards. I’ll keep testing the pcb based on your suggestions.
Have a Galax GTX980, was using this as egpu from my laptop, I remember last time it was working before I unplugged the PSU from Socket and I turned the laptop on and it even POSTED, and had quickly pulled out my laptop. Now I managed to buy a i5 9th gen system and the graphic card shows a display and distorts at after installing drivers, but weird I managed to open GPU-Z application to see my GPU details here, but RAM was showing as 0MB, can share you the details, would need help over here with a direction, what Shall I do next..
0MB always shows when the drivers are not running that's normal. Crash on driver install usually indicates the same thing you fix by the heatgun method. Could also be a damaged memory trace on the PCB or a knocked off capacitor. So look closely dor that before you expose the card to perhaps unneeded heat
@@CmdrSoyo fixed the missing capacitors did reflow applying a heatgun, problem still exists, what is the next thing I should look for? Also missing capacitors weren't a problem it's not been there for more than a year, I used to play games on a low clock speed as the system restarts when there is a load, so what do you prefer me to look at next??
@@michaelsantosh1610 honestly at that point it is likely a problem only a certified technician can fix. could always try to heatgun it again for a bit longer just in case but if it doesn't help it could be something you can only fix by actually replacing the entire GPU for example.
I got an old club3d hd7750,one time pc shuts off while playing and turns on only without video card I located a burnt and shorted ceramic cap, removed it and nothing changed. I took a cap that looked the same size from a more populated area of the board and soldered it where was the old one, now pc fans take speed like normal bootup but still no post/nothing on screen, gpu core gets hot and i cannot locate anymore shorts. From here how can i proceed? I really need my pc to be working, right now it's impossible to buy gpus even old and used ones
After 1 week of measurements i realized that two 0 ohm resistors in the same rail as the shorted cap were measuring 150 ohm each. For now i just bridged everything because i don't have any 0 ohm smd resistors and it works again.
Little update, it died after 30 seconds of furmark. Same symptoms as before. I even soldered new two 0 ohm resistors back in place, and this time they're fine. I'm wondering what could've gone wrong this time
Ok. Yesterday i couldn't work anymore on it because i was too tired. Today after some random testing i found a 2.2 ohm resistor split in half, obviously was measuring open circuit. I replaced it with a bigger smd one from a broken fluorescent lamp driver. I fitted it with a solder blob on one end. Working again for now... And stress tested it for an hour. I still don't trust it though. Let's see how much will it last
hi can you help me find my shorted part my fuse in the pcie slot is broken i tried to replace it but it still blow , i cant find where is the shorted part
The fuse will likely measure 0ohms or close to it to one of the VRM outputs. If you find that VRM you can look into the Highside mosfets of that VRM as at least one of it is likely shorted. Sometimes you can see the mosfet bulge a bit or it is slightly discolored. If not you will have to just remove all of them that are connected to the powerplane if the fuse. You can just check which highside mosfet measures 0 ohms from its drain to the output side of the blown fuse
I'm hoping someone can help me. I'm desperate. I built a new system (Ryzen 5 5600X and Asus TUF X-570 Pro), but salvaged the graphics card, a RX 5700 XT, because new models are impossible to find right now. I had everything hooked up and cable managed and then realized I had to add one more hard drive. I had the system running and went to loosen a screw on the panel covering the drive bay and my screwdriver tapped a heat-pipe on the 5700 XT, I heard a "tick" sound and then the lights went out...so to speak. The graphics card just died. I tested the card in another system, hoping it would boot up and mean that maybe my power supply or motherboard had the issue. But, it didn't work. So, I know it's the graphic card. Can anyone help me find someone to inspect it and see if there is an easy solution? I tried the oven trick, and had no luck at all. I'm going to watch the many videos I have found today and learn how to diagnose the problem. But, if anyone has experience with this card and may have some tips, please let me know. Thank you.
Hello I have an MSI R9 380 it works just fine, never crashed, no overheat fans spin normally BUT only outputs video through DVI I and DVI D ports no displayport nor hdmi port. How could I fix it or what may be the cause of this?
@@alexisxyz7531 the chip shouldn't be more than 5€. to replace it you would need some solder flux, fresh solder, maybe sodler wick and a hot air station. if you don't have that already i think it would be better to ask a local repair shop to repalce it
My GPU won't post when it's above ~42c, however, it can easily handle 80+c in Windows. Once it does cool down I can get a POST, but my BIOS throws a 'The VGA card is not supported by the UEFI driver, CSM has been enabled for compatibility'. I can then enter BIOS, disable CSM and 80% of the time it'll boot right into Windows just fine. It's even overclocked and renders like a champ, no overheating issues either. I've stripped the card and taken isopropyl and a toothbrush to it with no results. Any ideas what may be wrong with it? It's so strange that the BIOS says the driver isn't compatible, because the BIOS recognizes the card fine (shows Nvidia GPU 16X) For reference, it's an EVGA 1080Ti FTW3. Based on my limited knowledge, I feel like, since the card renders perfectly fine that it might be the BIOS VRM?, or the PXE\PLL VRM, or something related to either of those. If you or anyone could please help me I would really appreciate it. I've only had this card for 2 years and I'm disabled with no income, I can't let this thing die because it's all I have for a long time.
That is a very weird issue. Never heard of anything like that happening. Almost sounds to me like the GPU is fine and it's actually a motherboard issue. Maybe try to test the card in another system if you can
I'll give it a go in my streaming PC and let you know what happens. I'm 99% certain it's the card though. I tested the card at high temps at my benchmark of choice garbled up and the PC froze at 73c. I stripped the card again and scrubbed it for even longer and it's now able to reach 80c+ with no problems. I wish I knew more about PCBs so I could put a multimeter to it. There was no condensation on anything in the case besides the GPU, I'm assuming it's because the fans are smaller and move faster which made a perfect environment for the liquid to buildup. It almost feels like the bios vrm is getting inadequate power at high temps. When the card is below 40c I can restart as many times as I like with no error. I'm just blown away by how odd this is. I've been doing this shit for over 20 years and I've never encountered something like this.
I tried the other 16x PCI slot on my motherboard and got the same error. I've also flashed the vbios and the motherboards bios. Motherboard is an Asus Crosshair VIII Hero Wifi, barley over a year old, no liquid was on it so I'm confident it isn't the board.
This video is completely useless if you don't already know how to diagnose a GPU. The video is wayy too zoomed out and everything you say makes no sense if you don't know about all the electronic components
Great Video, I will add time stamps with some comments:
02:20 disassembly and visual inspection
07:55 get a multimeter (use resistance mode)
08:50 why not to use continuity mode Vcore measurement (ground to Vcore capacitors)
10:12 check 12 V power lanes (ground to first three connector pins should show several kOhms, 4th is ground; power connectors on pins should show very high resistance of k-MOhms to mass)
13:48 check 3,3 V power lane (last three pins should show very high resistance of MOhms, fourth pins from the right 400 Ohms, fith from right is ground)
14:14 power lanes to VRMs (voltage regulator modules) output Vcore ["+" side of capactiors to corresponding 12V pin (usually PCIe slot for Vmem and 6/8pins for Vcore, find out by going from the slot or pins to the input inductor or big capacitors) or input inductor or a big capacitor in front of the Vcore/Vmem; several kOhms expected]
16:55 power lanes to VRMs (voltage regulator modules) output Vmem (same strategy as for Vcore)
17:45 AMD VDDCI (memory controller power) VRM
19:42 VRMs to ground (very low resistance of down to 600 mOhms to Vcore "end" capacitor, little higher of around 20-200 Ohms to Vmem)
22:12 no short detected - power on card (2 possibilities: no picture or artifacts (both usually BGA or silicon damage); check voltages when card is powered on; "bake" card using heat gun but not in the oven!)
thanks very helpful
these videos are exactly what people need! What to measure and why, what could be causes...there is not much stuff on the internet about this
because these are stuff you only find out when enrolling to electronics courses and engineering programs. these are not some basic stuff that you could easily understand and apply, there are a lot of pre-requisites before you can understand the basics (an example is understanding diodes, what are the physics behind it?, etc, like that).
Kudos to Zoras Tech for making videos like this that are not very simple.
@@reinerabad3482it's not that hard to learn, just casually watching videos about these topics will get you knowledge, you don't need to be an expert to tell if something is not how it's supposed to be
Please never give up. i got in 1 video so much information RESPEKT before i was reading in the internet for Days to figured the Stuff out. Bin wirklich beeindruckt
I've done the oven trick to a iMac vga card, worked for several years without a single issue! Amazing trick!
Hey man, you got a sub. I do electronics repair, but these cards are alien technology compared to what I normally work on. Thank you for breaking it down, saves me a ton of time!
I love ze German accent. Awesome video. I also watched the video where you explain the components of a GPU PCB. What I'm missing and would love for you to make is the missing link between those two videos. What points do you need to jab with your multimeter to test what component and what resistance are you expecting to see on a good card? If you find no flaws with resistance testing and power the card, what points are you then testing to test the various components and what voltages are you expecting?
I’ve been wanting to get into this as a hobby. Thanks for the info! Hard to find good stuff on this
Hey man, it's very nice to see another small tech channel which is actually that good. Your content is on pair if not better than Buildzoid's, you should have 1000 times the subs you have. I especially liked the washing suggestion, even though I guess it was obvious seen my name and my channel lol. Keep it up!
agree
way more coherent, organized
@13:10 It looks like one or more capacitors are getting charged when you measure, therefore the resistance rises. Measuring applies a small voltage across the probes.
Yup, as they get charged the current through them drops and so the resistance reading increases (cause they are reaching the point of being an open switch for dc voltage)
Wow. Perfect. You're the first to really hit the nail on the head with common BGA damage. I've been fixing (attempting) 980TIs and many of them have the same memory issue and sometimes VRM issues. I think the root-cause is bga failure, which leads to other failures upstream, offen showing up as blown inductors and mosfets. You fix the short upstream but the real issue downstream on the GPU Core or VRAM itself. I think cards with large cpu cores and big power rails that support them just get too hot and thermal cycling is what kills the card, starting with BGA failure. So many "repaired" cards stil show artifacts or will show them within days after the so-called repair. Lead-free solder is just not good for this type of cycle as it's too brittle and get worse over time. I used to fix Plasma TVs and solder failure from thermal cycles is the #1 failure point. Most repairs involve refreshing solder joints with good quality leaded solder and then they run forever. Now to plan to do some serious reballing with leaded solder and hope resurrect a few decent GPUs.
How about random black artefacts that don’t take up whole space on screen and disappear quickly last 2 seconds. There are no green lines. Some times they take 20% of screen space. Like black areas (large pixel effect) is this GPU issue? I have RTX 3090
I wonder if you're usally a really really kind person, love the wholesome outro. And also love your videos!! :')
Great video. I have a 1080ti that shorted something from a water fitting leak on the back of the PCB. I tried briefly to measure the voltage rails but the whole card practically looks like a short but that's probably due to the low core resistance as mentioned here. Trying to bone up on some GPU knowledge with videos like this so I can take another look.
I dont trust your Benning multitool. If my machine didnt change from when I touched something I would be very suspect.
2022 iam a builder but due to cronic illness i cant work my dream is elctronics i see your video verry help full tu buddy
Where is the part one ? Video... I can't find it ...
Wow I think I remember those "Twin Frozr" cards back then, or maybe it's a deja vu 😂
Where is the other video you mentioned at the start as I need to know what to measure on my dead 1080 ti thank you
When you are testing for a short on a mosfet or a pad under a removed mosfet,what is the lowest reading In ohms that you would receive and still say it is not shorted? This is without considering the masters shorted reading? Thanks for shooting videos from start to end with explanations.
Depends what part you are measuring. Newer GPUs like 2080Tis have practically 0 ohms of core resistance (older cards sit around 0.3-5 ohms). Memory VRM usually have 20-100 ohms. 12V power planes should be several kiloohms and so on
@@CmdrSoyo thanks
Hey thanks for the video, really appreciate it. Would it make sense to apply some pressure to the GPU while heatgunning so that the BGA gets better contact?
i've never put any pressure on the GPU while reflowing. sounds more like an easy way to create shorts by shifting the GPU on accident. the weight of the GPU itself should be enough. plus the solder also has surface tension when liquid
@@CmdrSoyo Wish I saw this earlier my chip got shifted and now it won't start, not even the fans spin but the led on the shroud lights up. Is my video card done for good? Amazing channel by the way keep uploading great informative content. Subscribed
probably fixable with reballing but if it's not an expensive card it's probably not worth it.
what if i inspect the card visually and i find something missing on the back? i find pictures of the PCB with the component on , it looks like a capacitor (very small) but it has no code next to it , how do i find a replacement ? or what steps should i take to fix my problem ?
Awesome video! Information on this topic is so hard to find.
So I’ve got a card that doesn’t do anything. RGB does come out but fans don’t move, no video and no detection in Windows. The PC will boot. I’ve inspected the entire board, tried reflowing / resoldering joints where possible (capacitors and 8pin power) and used an 858D knock off to heat up the vram and BGA but so far no dice. I’m trying to learn how to apply the multimeter to test capacitors and mosfets because it seems to be a power issue.
Any thoughts on how I could continue diagnosing? It’s a 5700 XT that I bought in late 2019, shame it died so early. Trying my best to find what limited videos are available out there on the topic.
Most likely a power sequencing issue. Test all the VRMs. Also don't "reflow" things it doesn't work
Hi. I can say that you are very experienced technician re gpu repair but you still lecture as if we are all kids here watching your video and that is very helpfull. Could you pls explain in another video how a rtx 3070 could become shorted if a riser it was attatched to shorted one of its capacitor. It has two 12v power connector and one of them is shorted. If i detach the shorted one the system would turn on but the gpu is still not detected. Pls explain what to check and how. Im in the philippines and i am willing to acquire equipment to fix this gpu because i also have rtx3080 with same issue and exactly same cause of failure.
Insanly good video! Thanks a lot for spending your time.
Finally, ive been waiting for this since u said youd remake it
Great info and clear explanations. Very useful video.
VERY THANKFUL FOR THIS VIDEO
Where can I contact you about a problem with a GPU (R9 280X Windforce, no image, no fan spinning, no shorts found but also no 12V on VCore Coils)?
You mentioned "i did not heat gun it long enough" grabbed my kids old gtx 650 ti and dropped 375 degrees on it for about 15 minutes, you can see the top of the solder go from a dark meatal color to a bight silver sheen on top, that's when I turned it off and let it set. it works now!!! Do you still work on GPU's I got a HD 7950 I wanted to send to you. it had a fan controller issue and I fried something on it and cant figure it out.
I have tried but have failed!! Could you get a cheap video card (10-20) on ebay, and show the steps and show what you could do that would be wrong or get wrong readings--like checking for shorts, ohms setting or diode, black on ground or red, checking mosfets. I have ruined several cards just trying to get started!!!!! Thanks
Thanks for a great video. Good lighting, super well explained and actually useful.
My 1080Ti died and I'm waiting on the RMA to come back at the moment.
In the mean time I put in my old GTX 580, which worked for about 2 weeks before it also died.
That's got me worried my PSU or motherboard could be killing GPUs.
The 1080Ti wouldn't boot at all, no display output.
The GTX580 does have display output but the PC will reboot when I try to install the drivers.
Sounds like a VRM issue on the 1080Ti and a BGA issue with the 580 right?
Do you think it's just coincidence both cards failed close together considering the 580 is about 8 years old by now?
Both the PSU and Motherboard were top of the range back in 2011 but they're also very old by now.
What do you think?
It's unlikely the defects are related. You could try the heatgun method on the 580 and see if it starts working again. I had a 580 once that did the same and it was fine after i heatgunned it.
Your PSU is likely fine. After all those yreas maybe a but full of dust but that's fixable by just cleaning it. The only thing i can think of that could degrade over time is the capacitors which might cause some higher coltage ripple but that shouldn't kill components. May cause some crashes from time to time if it gets really bad at worst i think
@@CmdrSoyo Sweet as, I figured it was probably just bad luck for one and old age for the other. Puts my mind at ease a little so I can wait a little longer for the Ryzen 5600 non-x if it ever comes out.
Didn’t have any dust or anything either. I think the PSU was so over specced the fan never even runs when it’s set to semi passive cooling.
Thanks for the help!
omg dude i love this channel
Wow you are really knowledgeable. Help me please. I have a vega 64 that I know something has shorted my psu cuts power when I start up games in 4k resolution. The strange thing is that benchmarks complete. But a few games trigger it every time. Can you point me in the right direction of where you might check first or maybe help me please? Or would you possibly like the opportunity to fix it for your channel? That would be quite cool actually.
Thanks for this guide and your first part! I will use for testing my GPU, will not boot but the RGB will turn on.
Have the same issue with my GPU what were your results?🤔
If this card is AMD try to flash the bios
@@dearestgonzo could we possibly get into contact with each other so I could get guidance with how to fix this?🤔
@@Swc_Leader I think that would be easier for you to get help from youtube videos. I did it watching videos and worked for me. Just search "Flashing AMD Radeon bios" and you wont believe how easy it is
@@dearestgonzo ok I have a few questions before you flashed the bios where you getting black screen? Or werw you already able to see picture with your GPU it was just the artifcating/freezing issue you have to fix?🤔
where te F can i find a roos rs2 c145 ?
I am trying to diagnose a GPU, (980 ti) which does not display any output despite the power planes being fine. The LED's and fan turn on just fine however no video output is detected. There are no signs of visible corrosion or signs of shorts using the techniques in this video. My assumption is that the core itself is dead, how should I go about identifying if this is true? Any suggestions?
Hey awesome videos, love it. So i went and measured resistances on the major power plane and nothing came up short for what I can tell, I also just started this today. So I powered on the card and checked if the GPU die was getting warm, it was not. The BIOS chips on the back was getting hot. I think my VBIOS chip might be dead. Can I replace that? Also, I measured voltage on the PCIe pins and was only getting 0.3V when powered, so maybe I'm not measuring correctly. Maybe I need to check the minor power components? Thanks.
How do e power boards vary the voltage as needed? Do they provide a constant unchanging voltage to the core?
How complicated is it for a professional to make a graphic card diagnostic?
(From-to, in case it's easy and in case it's complicated (
Where can I find out more about the "12V plane"? Where do you find this info about card layouts? I can't find specs and what measurements to really expect. I've got a general idea but, I'm troubleshooting a 1080 GTX that has no obvious burns ... but still not sure. I'd like to know what resistance values to expect on that specific card if that exists? Thanks for the video either way.
Hey I have a resistance of 60 K Ohms from the third pin on pci express to ground what could be the issue?
Hope you can answer my question, I have a r9 270x that powers on, but I doesn’t display, checkin the resistors to core they give me 9-12 ohms, that would be a short? Or it’s normal?
It’s a xfx model
Hello! I'm trying to research if my water-damaged RTX 3070 is repairable. The card runs stable but is power throttling. GPU-Z shows high power draw from one 12v and virtually none from the other, TDP at 190%. The back of the PCB showed corrosion around the power connectors and some nearby components before cleaning with alcohol and QD cleaner. Does this indicate any specific problem? Is it salvageable? Thank you.
I just pull the old thermal pads and apply new ones.
Thank you, this is a great video. I have a radeon r9 390 I used for many years, but it developed this silicon / BGA damage as you describe. I'm hope to get it working again.
I have a evga 980ti with the blower fan style I love those looks just not a good setup .. it turns on shows no video and fan sounds like its on high the chip gets hot the board shows no sign of anything getting hot it was put in a oven buy the signs on it im getting what looks like a short on the pcie 3 volt power can u go over proper diag of a card from pcie
No microscope zoom, you only touch on the power rails or what components are involved with a circuit, and we can't see the contact points. We have to take your word for where the multimeter is set and where your stabbing, while you admit mistakes in the first video. I do not want to discourage you, but electronics is a subject that does not lend itself to poor equipment, or brief explanations. The students need to see, to hear, and to understand to have confidence. How about drawing it on paper? I believe that would help. We can see that.
Please do u accept mail in repairs? I have a faulty RTX 3090 due to high voltage damage
Excellent! Thank you.
30:23 really ? Put the gpu in the water : P
Btwn I had some issues with my gtx 660 ti it freezes and displays a white color on the hole screen or pink or other color and before 15 days it stopped working what do you Think should Wash it in water or use a heat gun or something ?
Pls answer
My RTX 2080ti, the fan spin but it doesn’t have power the lights are off..My PC is water cooled so the radiator dripped water on it and that’s how I found, no power.
I took it apart cleaned it. I bought a new thermal paste do you think I should replace the thermal pads? Hopefully it will turn back on.
Very nice and helpful video.... I had a card that was artifacting tried the heat gun method but when I plug it back in it the mb starts to beep
constantly.. No end... Any ideas?
OK it's not beeping anymore.. Just no display.. The computer comes on an boot... Just no display
The card is not getting detected / doesn't power on
@@CmdrSoyo the fans and leds comes on and the die it self gets warm
@@imranpaul1162 VCore works then. Could still be a dead minor rail or a damaged BIOS or just straight up a dead GPU
@@CmdrSoyo ok... So what do you recommend I do?
Hi , I am learning. I have a HD7950 card with black screen. The core was getting hot so I had a heat sink on it and did not leave it on too long. I was cleaning up the controller chip solder joints and now when I plug it in there is no heat and non of the power stages have a voltage or a signal with the oscilloscope. At least I had .9 volts before. Any thoughts? Thanks
Zoras Tech what is your multimeter?
@Zoras Tech thanks a lot, can you make a video for nvidia new cards rtx series.
bro can you plz guide me i have an sapphire rx 570 its light blinks on startup abd then goes off nothing happens no video no power would it be short or main chip
Hello, my PC toppled over and the motherboard and GPU seem damaged.
The GPU sometimes take signal sometimes doesn't. Is it just simple PCI-E connector issues or more serious?
Motherboard only decides to boot with 2 but not 4 ram sticks. What to do? 😔
guys i need help i have gtx980ti 6gb ichill (if i start game he got crash )
i got 2 xfx rx 580s that doesnt give display...the fan spins die gets hot but no display....can anyone tell me whats the issue
I just wish at least the high end cards would be constructed in a more industrial manner, you know like replaceable modules of the VRM, memory sticks that can be switched for new ones, like a motherboard. I hate the "this is consumer electronics" way of engineering of these.
modular parts like ram and vrm modules would ruin efficiency and signal integrity as well as transient response. just not practical on GPUs
at 10:54 your measuring the resistance on the pci express? what if you dont get a resistance on that side of the pci where you poked and only get resistance on the far right side on the long rail... sorry if my questions stupid lol
0 ohms on 12V PCI-E means it's shorted. Make sure you are actually probing the 12V pins though there are some ground pins in there which will always show 0 ohms (look up a pin out on the internet if you're not sure)
@@CmdrSoyo thank you!!!
Hey, i have a question, the first fuse that you mention, what amperage does it has?
i need to replace it but i'm not sure about it's amperage, thanks for ur video!
i don't know the exact value but i know the value from another card. there it used a 20A (2x10A in parallel) fuse for an 8pin and a 10A fuse for the PCI-E slot. it should work fine if you just copy that
@@CmdrSoyo thanks mate, u are really cool 😎 thanks for everything
My 970 recently fluked after some sparks on some small resistors/capacitors at the back of the board. Since then it keeps hard resetting at the windows logo i also notice temperature in the vrm/inductor area is shitass hot.
So i open it up and there's only a little bit of tearing in the painting on the pcb between vrm mosfets and 2 specific inductors. No blow marks, smoke cracking on vrms/inductors or anything tho. Resistance readings seem fine, i've also checked major spots he mentions in the video.
Any ideas on what i'm missing?
I have a card, it's video output ports are not working, but it works fine as laptop eGPU using internal display (with nvidia optimus), what could be wrong with it?
i have red devil rx5700xt .. when i turn on my pc , all component are working but my gpu fan is not spinning , and no display (gpu rgb on) ... is there a chance that my gpu shortt?
I have a short in the power connector showing. MOSFETs are good from the readings I get. My fuse is ok too. What would be the most likely issue for my card? Gigabyte GTX 1060 6GB. Lightning strike on my house took it and some other electronics out.
Short on a power connector usually means 12V is shorted to some output. But if you bring the lightning strike into the picture it might also be that there was some very high voltage sent into the Card and may have popped some capacitors or other components
Do you tape anything off with something like Kapton tape before using the heat gun?
it's good practice to cover sensitive componentes but to be honest i only cover can type capacitors with aluminum foil
I have a gtx 980 ti when I checked over for shorts and everything looked fine when I installed and turn on it powered up the lettering on it lights up fan sounds like its on high and no screen output I get nothing
I have some cards I would like to have fixed can I pay you to trouble shoot them?
Hey thanks for the vid!l so helpful! I’ve done one attempt to use the heat gun on my 780 ti which is artifacting. But when I started to heat it up there seemed to be liquid coming from under the chips and core? Is that right? as I backed off and tested the card but didn’t make it any worse but still the same should I keep the heat on it longer?
sometimes there is solder flux under the chips which does become liquid if heated. so it is "normal" you might say but i never had it happen myself.
@@CmdrSoyo that makes sense! Thankyou for the reply I’ll give it another go for a bit longer this time, see if it sorts it out
Artifacts come from vram issues mainly and if it is a gpu (silicon) issue your chance of fixing it is exactly 0%
Hey are u need specific Multitester for Gpu ?
my secondhand rx 480 have 213.5 ohm on 3.3v pcie power, is that normal?
my gpu is just spinning it fans and my pc is not booting whenever i plugged it in, what do you think the problem?
Help! the one of the "towers" close to the VRMs is detached from the VGA board! It happened because i accidently snapped it while trying to dissamble the VGA. The VGA now runs but it shuts down if i put loads like games and stuff... Please help.. Cant afford another VGA while the prices are a bit crazy
Hi thank you so much for your videos firstly! I have a video card (gtx 1060 6gb) that was plugged in and my house was struck by lightning and I know the card was shorted because no it has no display, BUT the fan runs at max speed. Any idea if it is fixable? Thank you so much in advance! I am German also by the way, but live in the states greetings!
got the same problem with a 1060, can't get it to work
I would heat it with heat gun xD
hey bro i need your opinion on that video on the link i shared please.
rx 560 4gb main chip not heating and pc not turning on when plugged in any solutions
Please help! My RTX 2070 just started showing artifacts. Does same thing on 2nd computer, so appears it is a hardware issue. When measuring resistance from the capacitors to the pin on the 8pin slot, I get a measurement from all capacitors except 1, which is an open line. Does this capacitor need to be replaced? All of the other capacitors start at 0 and increase quickly increase by hundreds of Ohms.
About to try my luck on 2 borked 580 8GBs wish me luck
Bro I have a the Same Gpu. When i Start it, its fan works. But when i Install any gpu drivers, it shows black screen. I cant use my pc. Only if i enter with safe mode and delete those drivers, my pc works.
Can I get solution? Mine GPU amd R9 270x
So, when you find a card, and all the info that is given in the description is "Not working" and you ask for more info about it and their reply is "read the description", Do you skip it?
i would skip it yeah. i usually only go for cards that artifact so i can just heatgun them for a bit and forget. i may try some proper repairs one day but for now i just get cards to heatgun
My card turns off when playing
Nice
Hi, I have a 2060 that seems to be dead. Motherboard doesn't even pick it up, just shows an empty slot. This happened while I was experimenting with risers for the first time, the 1x adapter sparked at the motherboard and the card hasn't worked since. There's no visible damage to the pcb. I've tried poking around with the fluke, but it is my first time and is a bit daunting. Any ideas for specific areas of the pcb to check?
Since you saw a spark on the PCI-E slot you should look there. If the card has a 12V fuse it is likely blown. It could also have damaged a VRM connected to it. It could also be that the motherboard was damaged so try the card in a different slot or in a different motherboard.
@@CmdrSoyo Thank you for the comment! I have ruled out the motherboard by testing with other cards. I’ll keep testing the pcb based on your suggestions.
Have a Galax GTX980, was using this as egpu from my laptop, I remember last time it was working before I unplugged the PSU from Socket and I turned the laptop on and it even POSTED, and had quickly pulled out my laptop. Now I managed to buy a i5 9th gen system and the graphic card shows a display and distorts at after installing drivers, but weird I managed to open GPU-Z application to see my GPU details here, but RAM was showing as 0MB, can share you the details, would need help over here with a direction, what Shall I do next..
Had to probe to find 1.5v at Vmem and 0.9 x 4 at Vcore, any kind of help would be appreciated!
0MB always shows when the drivers are not running that's normal. Crash on driver install usually indicates the same thing you fix by the heatgun method. Could also be a damaged memory trace on the PCB or a knocked off capacitor. So look closely dor that before you expose the card to perhaps unneeded heat
@@CmdrSoyo fixed the missing capacitors did reflow applying a heatgun, problem still exists, what is the next thing I should look for? Also missing capacitors weren't a problem it's not been there for more than a year, I used to play games on a low clock speed as the system restarts when there is a load, so what do you prefer me to look at next??
@@michaelsantosh1610 honestly at that point it is likely a problem only a certified technician can fix. could always try to heatgun it again for a bit longer just in case but if it doesn't help it could be something you can only fix by actually replacing the entire GPU for example.
hello ny 1060 has very low resistance( about below 0.0 ohms) between ground and 1.8 v inductor.. is this a defect..?
of your multimeter likely
What if graphic card shut down PC ?
you have a short circuit
I got an old club3d hd7750,one time pc shuts off while playing and turns on only without video card
I located a burnt and shorted ceramic cap, removed it and nothing changed. I took a cap that looked the same size from a more populated area of the board and soldered it where was the old one, now pc fans take speed like normal bootup but still no post/nothing on screen, gpu core gets hot and i cannot locate anymore shorts. From here how can i proceed? I really need my pc to be working, right now it's impossible to buy gpus even old and used ones
After 1 week of measurements i realized that two 0 ohm resistors in the same rail as the shorted cap were measuring 150 ohm each. For now i just bridged everything because i don't have any 0 ohm smd resistors and it works again.
Little update, it died after 30 seconds of furmark. Same symptoms as before. I even soldered new two 0 ohm resistors back in place, and this time they're fine. I'm wondering what could've gone wrong this time
Ok. Yesterday i couldn't work anymore on it because i was too tired. Today after some random testing i found a 2.2 ohm resistor split in half, obviously was measuring open circuit. I replaced it with a bigger smd one from a broken fluorescent lamp driver. I fitted it with a solder blob on one end. Working again for now... And stress tested it for an hour. I still don't trust it though. Let's see how much will it last
hi can you help me find my shorted part my fuse in the pcie slot is broken i tried to replace it but it still blow , i cant find where is the shorted part
The fuse will likely measure 0ohms or close to it to one of the VRM outputs. If you find that VRM you can look into the Highside mosfets of that VRM as at least one of it is likely shorted. Sometimes you can see the mosfet bulge a bit or it is slightly discolored. If not you will have to just remove all of them that are connected to the powerplane if the fuse. You can just check which highside mosfet measures 0 ohms from its drain to the output side of the blown fuse
@@CmdrSoyo thanks a lot bro your kind for helping me , but i already surrender hahahah so hard to fix my rx 580 or im just newbie thanks again
@@CmdrSoyo clever thinking
🥰
I'm hoping someone can help me. I'm desperate. I built a new system
(Ryzen 5 5600X and Asus TUF X-570 Pro), but salvaged the graphics card, a
RX 5700 XT, because new models are impossible to find right now. I had
everything hooked up and cable managed and then realized I had to add
one more hard drive. I had the system running and went to loosen a screw
on the panel covering the drive bay and my screwdriver tapped a
heat-pipe on the 5700 XT, I heard a "tick" sound and then the lights
went out...so to speak. The graphics card just died. I tested the card
in another system, hoping it would boot up and mean that maybe my power
supply or motherboard had the issue. But, it didn't work. So, I know
it's the graphic card. Can anyone help me find someone to inspect it and
see if there is an easy solution? I tried the oven trick, and had no
luck at all. I'm going to watch the many videos I have found today and learn how to diagnose the problem. But, if anyone has experience with this card and may have some tips, please let me know. Thank you.
Hello I have an MSI R9 380 it works just fine, never crashed, no overheat fans spin normally BUT only outputs video through DVI I and DVI D ports no displayport nor hdmi port. How could I fix it or what may be the cause of this?
Well the ports are dead. There is likely a chip that controls HDMI and DP outputs that is defectice and has to be replaced
@@CmdrSoyo Oh no, do you know if it is expensive?
@@alexisxyz7531 the chip shouldn't be more than 5€. to replace it you would need some solder flux, fresh solder, maybe sodler wick and a hot air station. if you don't have that already i think it would be better to ask a local repair shop to repalce it
@@CmdrSoyo I see, thank you so much for your attention and patience to reply!!
Where have you learned all that ?
Btw are you German ? :)
yes he is
Give me moahhhhh
Washing does nothing.
Don't clean your graphics card with water as it was said in the video. Use isopropyl alcohol instead.
man why I didn't find you earlier.
My GPU won't post when it's above ~42c, however, it can easily handle 80+c in Windows. Once it does cool down I can get a POST, but my BIOS throws a 'The VGA card is not supported by the UEFI driver, CSM has been enabled for compatibility'. I can then enter BIOS, disable CSM and 80% of the time it'll boot right into Windows just fine. It's even overclocked and renders like a champ, no overheating issues either.
I've stripped the card and taken isopropyl and a toothbrush to it with no results. Any ideas what may be wrong with it?
It's so strange that the BIOS says the driver isn't compatible, because the BIOS recognizes the card fine (shows Nvidia GPU 16X)
For reference, it's an EVGA 1080Ti FTW3.
Based on my limited knowledge, I feel like, since the card renders perfectly fine that it might be the BIOS VRM?, or the PXE\PLL VRM, or something related to either of those. If you or anyone could please help me I would really appreciate it. I've only had this card for 2 years and I'm disabled with no income, I can't let this thing die because it's all I have for a long time.
That is a very weird issue. Never heard of anything like that happening. Almost sounds to me like the GPU is fine and it's actually a motherboard issue. Maybe try to test the card in another system if you can
I'll give it a go in my streaming PC and let you know what happens. I'm 99% certain it's the card though. I tested the card at high temps at my benchmark of choice garbled up and the PC froze at 73c. I stripped the card again and scrubbed it for even longer and it's now able to reach 80c+ with no problems. I wish I knew more about PCBs so I could put a multimeter to it. There was no condensation on anything in the case besides the GPU, I'm assuming it's because the fans are smaller and move faster which made a perfect environment for the liquid to buildup.
It almost feels like the bios vrm is getting inadequate power at high temps. When the card is below 40c I can restart as many times as I like with no error. I'm just blown away by how odd this is. I've been doing this shit for over 20 years and I've never encountered something like this.
I tried the other 16x PCI slot on my motherboard and got the same error. I've also flashed the vbios and the motherboards bios.
Motherboard is an Asus Crosshair VIII Hero Wifi, barley over a year old, no liquid was on it so I'm confident it isn't the board.
This video is completely useless if you don't already know how to diagnose a GPU. The video is wayy too zoomed out and everything you say makes no sense if you don't know about all the electronic components
Alter, wieso babbelst Du nicht einfach Deutsch? Das ist doch ne Farce ^^