Author Topic: Gigabyte GTX 1070 8gb Founders edition missing to SMD ceramic capacitos (Read 4062 times)

wraper · « **Reply #25 on:** October 12, 2021, 08:40:48 am »

Quote from: Koletsa on October 12, 2021, 06:21:00 am

I knew about the short, but believed that the reason was one of the two capacitors we talked about. Then Wrapper said it can't be a cap, because they are parallel to the rail. Can you share the 5 v rail scheme so I can start looking for the culprit as well?

Cap can short the rail, although it happens not very often. Failing semiconductors are way more common. Short likely is an initial failure, however removing the short by itself will not fix the card. Since there must be also something blown in series with the rail.

Johnny10 · « **Reply #26 on:** October 12, 2021, 11:12:13 am »

If you will notice this schematic shows C1100 and C1102.

Best of Luck

Koletsa · « **Reply #27 on:** October 12, 2021, 05:31:12 pm »

Thank you very much! Here on that schematics I see resistor... which I should find on the board

. Anyhow - very helpful aid. Thanks a lot!

Koletsa · « **Reply #28 on:** October 12, 2021, 05:34:53 pm »

I guess you are right. But need some kind of plan and at this point is to fix the 12v to the Vin of the converter. What will happen from then on, we will see. Thanks again for the info! If you have any other suggestions, pls feel free to share them.

Johnny10 · « **Reply #29 on:** October 12, 2021, 06:05:41 pm »

Read through the Wiki Repair Article on the Pascal 1.8V line, it has a very good analysis of problems.
From bad Resistor which I guess is common to bad controller.

There is a lot of good information on the site you chose.

You do have 3.3 coming into controller.
Its interesting that you have zero volts on Pin 2. I would think if something were dragging it down there would be something there?
Not simply zero.

I get .3168 Ohms from Pin 2 on PCIe to U2 Pin2. If I remember correctly?
I am in the middle of another project at the moment but I will check later.

Well the adventure begins, anew.

Koletsa · « **Reply #30 on:** October 12, 2021, 07:16:26 pm »

Hello! Yes, I have 3.200-3.300 v on pin 6. As for your second question - from pin 2 (also 1 and 3) on the PCIe to pin 2 on the convertor I have 6+ mega Ohms, decreasing steadily. The strange thing (at least for me) is that the resistance is dropping constantly. I started to write my answer at 6+ mOhmes, then i checked again few mins later - it was at around 3, not it is at 1,5 mOhms.
Take care of your projects. I have a lot of reading to do, so I'm, not in a hurry.

Koletsa · « **Reply #31 on:** October 14, 2021, 09:45:33 am »

UPDATE: I managed to fix the short to ground. The culprit was c1100 - after removing it I no longer have short on pin 2 (Vin) on the buck converter. I even have some voltage there - little less that 2 v. Now I have to find where are my other 10 v that I need on that pin.

Johnny10 · « **Reply #32 on:** October 14, 2021, 01:13:15 pm »

After removing did you check the value of C1100 ?

Koletsa · « **Reply #33 on:** October 14, 2021, 04:09:00 pm »

Yes, I got O.L. It was obviously dead. I hope it was 22uf. Now I have to find out what makes the voltage to drop. By measuring I found out that one of the resistors on the rail to pin 2 (Vin) on the buck convertor - R987 - gets on one of his legs 12 v, but on the other my multimeter measured 1.6-2 v. Exactly the same voltage that gets to pin 2. The resistor is 0 or 000, which makes it most logical that he is faulty as well, because 0 omh resistor shouldn't lose 10 v and drop the voltage from 12.10 to below 2 v. But I'd like to do a little bit more reading before dismounting it, because the mount process with those smd-s is pain in the ...

Johnny10 · « **Reply #34 on:** October 14, 2021, 05:18:43 pm »

It said in the troubleshooting description that resistor failure is common.
Maybe it was acting as fuse?

Jump around it and see.

Koletsa · « **Reply #35 on:** October 14, 2021, 07:52:54 pm »

The first thing I have in mind is to desolder it and replace it by a single, simple wire, because I read that 0 ohms are basically that - a single wire in a body. What do you mean by "jump around it" exactly?

Johnny10 · « **Reply #36 on:** October 14, 2021, 08:30:36 pm »

I'm laughing right now!!

Look what I just found today at 4:30PM

A video about missing 5V on 1070 card.
And he had C1100 shorted.
And R987 0 ohm resister showing high resistance.

Take a look.
Same as what we determined already.

Hope this is your cards only problem.

Johnny10 · « **Reply #37 on:** October 14, 2021, 09:05:34 pm »

I believe you could just measure Ohms across the resistor in place.
Should be low resistance. I get 0.027 Ohms on R987.

Koletsa · « **Reply #38 on:** October 14, 2021, 09:13:55 pm »

Ha, Tech cemetery. I know the channel. Now, after I looked at the video, I think I have watched before, but totally forgot about it. I hope I have the same problem. On mine r987 I have 600+ mOhms now. I measured another random o Ohms resistor - 2 ohm, which I believe is ok. Tomorrow I will try to find 0603 0 Ohms resistor and see what will happen! Will keep you posted.

Johnny10 · « **Reply #39 on:** October 14, 2021, 09:20:38 pm »

Yes, To measure this low resistance you need to zero your meter first.

You could solder a thin wire across the top of the resistor to quickly check if 12 volts is present and controller starts..

Koletsa · « **Reply #40 on:** October 15, 2021, 04:12:50 pm »

Hey, I have a good news and a bad one. The good new is that I managed to replace the 0 Ohms resistor and now I have my 5v, 1.8v and 1v PEX. I have the 0.7 on the Vcore. The bad news is that I don't have any voltage on my Vram. While I'm looking at the schematics, could you please do me a favor and measure the resistance of the R33 inductor. Mine currently show 0.1 Ohms. Also my 1.8 v inductor shows 7.5 Ohms resistance instead of the 800+ Ohms stated in the Wiki repair, but I get the voltage I need there. Should I be worried about its resistance? Could you also check Q 13 (right next to the Vram choke). I have 12 v on the top 2 right legs, 0.200 v on the right bottom one, on no voltage on the lest legs. Is the same case back to you. Tnahks in advance!

Johnny10 · « **Reply #41 on:** October 15, 2021, 04:29:58 pm »

Yes, Sorry about that I was reading it wrong.

You have 1.8V

RS3 Yes measures quite low 0.02 or lower.
Inductors are supposed to be very low to zero.

Koletsa · « **Reply #42 on:** October 15, 2021, 04:45:50 pm »

I read that as well. Not sure yet. At this point the Voltage I get to the Vcor is 0.7v as it should. Also my Chip get hot on all its surface, not only in the bottom left corner. And also the is a small change that (I guess 5%) that the bios chip is dead and replacing it may fix the issue. But first I need to get some voltage to the Vmem. After that I will figure out the next step.

Update: I managed to f%%% things up - I guess I shorted something while measuring the voltages across the boards, because I got a spark, followed by smoke. Now my PEX rail has 2v instead of 1v and gets very hot. All my choke inductors are with messed up resistance and once chip (right above the PEX inductor) got e fracture. At that point I'm about to sell the cart as non working and for parts.
Thank you a lot for all the help. Much appreciated!!!

Johnny10 · « **Reply #43 on:** October 16, 2021, 04:31:25 pm »

That is disappointing.


EEVblog Main Site	EEVblog on Youtube	EEVblog on Twitter	EEVblog on Facebook	EEVblog on Odysee

Author Topic: Gigabyte GTX 1070 8gb Founders edition missing to SMD ceramic capacitos (Read 4062 times)

Share me