Jump to content

980 Ti - Fixing a blown inductor (1R0)

Beneteau

Hi guys,

 

I've set out on the path of trying to fix my dead 980 Ti and figured posting a thread could possibly help others in the future so here goes.

 

So what happened?

 

In short, Furmark happened. Possibly in combination with a not-so-amazing PSU. My 980 Ti is a second-hand card so I can't say what happened to it before I bought it. I was able to snag it for 250 EUR or about 296 USD and it ran perfectly fine for about 100 hours of gaming. Not even any coil whine. 

Until I ran Furmark. Oddly enough the card had done the benchmark a couple of times before without any problems. Until a couple weeks back. As soon as the benchmark started I heard a sharp electrical crack. My system immediately shut down and a strong burning smell filled the room. The PSU's short circuit protection kicked in as soon as I tried to start the system. Removing the 980 Ti was the only way to boot the system.

 

The diagnosis

 

I knew beforehand that RMA'ing was going to be out of the question. The card was "imported" by the previous owner from some dealer in Poland and Gigabyte does not do RMA's on cards bought outside of the country. Poland is a 2 hour flight away and I don't speak Polish, so that option is closed.

 

That left me with only one option. Take the card apart and look for visible damage. I don't have a picture of my card, but the picture below is of the exact same failure but on a GTX TITAN which has a similar PCB to the 980 Ti. There's quite a few TITAN owners on the internet that have suffered the same failure related to excessive power-draw. It seems that this particular part of the board is made up of sub-par components (or Furmark is just crazy). Every post I've found on the subject was resolved by an RMA, unfortunately I'm not in that position.

 

The failure is a cracked inductor value 1UH (code 1R0)  north of the intake voltage controller circuit as depicted on the picture below.

 

The plan

 

The way I see it there's two different problems to overcome here.

 

  1. Replacing the inductor is going to be guessing work. There's no data sheets available so figuring out the exact specifications is beyond my knowledge. I've talked it over with my local electronics shop and we've narrowed it down based on inductance value (1R0) and physical dimensions (6.5mm by 6mm). My repair guy claims it probably was rated somewhere between 6A and 12A. So we went and ordered a 12A one to be on the safe side. Again, pure guesswork.
     
  2. We're hoping that the excessive power-draw by Furmark has caused the blown inductor to act as a fuse. But let's be honest, the chances of it blowing clean without anything else downstream getting fried is fairly small. This seems to concur with what I've read on other forums in topics on similar failures. RMA feedback in general was "Sorry, you've also blown your voltage controller / memory controller / core ( o_o ) and your card is unfixable".
     

the progress

Card is currently at the shop and will have the new inductor soldered this week. I'll post back with the result and if she posts...

 

5a402f67d3ee8_GTXTitan.jpg.c536d011add823f023292290703560bc.jpg

Link to comment
Share on other sites

Link to post
Share on other sites

18 minutes ago, Crunchy Dragon said:

I didn't know furmark was capable of killing Maxwell-based cards, just everything up to Kepler.

I heard about Furmark killing cards but thought it was urban legend. Well, I certainly got proven wrong on that...

Link to comment
Share on other sites

Link to post
Share on other sites

6 minutes ago, Beneteau said:

I heard about Furmark killing cards but thought it was urban legend. Well, I certainly got proven wrong on that...

Yeah, it was known as the GPU-killer back when Kepler was king.

Quote or tag me( @Crunchy Dragon) if you want me to see your reply

If a post solved your problem/answered your question, please consider marking it as "solved"

Community Standards // Join Floatplane!

Link to comment
Share on other sites

Link to post
Share on other sites

Good luck on the repair and let me know if it works.

 

In general, inductors are very robust componets and can handle huge but short overloads without damage. I assume the driver FET was unhappy / shorted out. Or the inductor was just undersised by design.

Mineral oil and 40 kg aluminium heat sinks are a perfect combination: 73 cores and a Titan X, Twenty Thousand Leagues Under the Oil

Link to comment
Share on other sites

Link to post
Share on other sites

6 hours ago, Stefan1024 said:

Good luck on the repair and let me know if it works.

 

In general, inductors are very robust componets and can handle huge but short overloads without damage. I assume the driver FET was unhappy / shorted out. Or the inductor was just undersised by design.

I tried Googling but can't find the answer: What is a driver FET?

Link to comment
Share on other sites

Link to post
Share on other sites

12 minutes ago, Beneteau said:

I tried Googling but can't find the answer: What is a driver FET?

FET = field effect transistor: https://en.wikipedia.org/wiki/Field-effect_transistor

 

Basicly the switch that is used in a step down converter: https://en.wikipedia.org/wiki/Buck_converter

 

Mineral oil and 40 kg aluminium heat sinks are a perfect combination: 73 cores and a Titan X, Twenty Thousand Leagues Under the Oil

Link to comment
Share on other sites

Link to post
Share on other sites

10 minutes ago, Stefan1024 said:

FET = field effect transistor: https://en.wikipedia.org/wiki/Field-effect_transistor

 

Basicly the switch that is used in a step down converter: https://en.wikipedia.org/wiki/Buck_converter

 

Thanks for the information. Given you mentioned inductors are generally pretty sturdy I don't have high hopes for the card. There's no other physical damage visible, but that doesn't really mean anything.

 

Anyway, I'll post back when the card is soldered!

Link to comment
Share on other sites

Link to post
Share on other sites

6 hours ago, Beneteau said:

Thanks for the information. Given you mentioned inductors are generally pretty sturdy I don't have high hopes for the card. There's no other physical damage visible, but that doesn't really mean anything.

 

Anyway, I'll post back when the card is soldered!

The FET can be replaced easely as well, they cost about 5$. I repeared a board that was worth 1000$ by replaising a single FET (sadly it was at work and mnot my own).

 

They are usually "normaly open", meaning when you measure the resistance between all pinn you should have > 10k Ohm. They usually fail short and create a short circuit between drain and source (< 1 ohm). This makes it relatively easy to finde the faulty one.

Mineral oil and 40 kg aluminium heat sinks are a perfect combination: 73 cores and a Titan X, Twenty Thousand Leagues Under the Oil

Link to comment
Share on other sites

Link to post
Share on other sites

woah good luck for the repair of your gpu, hope it will be revived. 

 

and about furmark, yes, it is a gpu killer, i ran furmark just for stress test because i saw it on LTT and other tech channels, then it killed my 7950 ?

Link to comment
Share on other sites

Link to post
Share on other sites

Guest
This topic is now closed to further replies.

×