Repair amd sapphire nitro r9 390

Mat2786

n00b
Joined
Apr 27, 2021
Messages
30
Hello all i have problème with my r9 390 sapphire nitro.
Can you help me?
I have Black screen and no boot.
I have tested all Voltage and i have no Voltage to vmem.
And two capacitor.
After other test i connect power supply 2 mosfet 4c10n have 12v
And 2 mosfet 4c10n 0v fuse 0v capacitor 0v see picture.
Thank you
 

Attachments

  • 20210427_194455.jpg
    20210427_194455.jpg
    344.5 KB · Views: 0
  • 20210427_230849.jpg
    20210427_230849.jpg
    311.7 KB · Views: 0
Gah, stop moving this post around!

First, I need you post actual measurements, and a picture marked with exactly where you're taking them from. Resistance first:

1. VCore to ground
2. Vmem to ground
3. VAux to ground
4. Resistance across each fuse (four by the power connectors, one by the slot connector)

Get those measurements, and we can talk about the operating voltages later.
 
Sorry.
Ok
All Vcore to ground 2.0 ohm
Vmem to ground 120 ohm
Vaux to ground 57.3 ohm
All fuse 0.1 ohm

Thank you for reply
 

Attachments

  • 20210428_230230.jpg
    20210428_230230.jpg
    344.4 KB · Views: 0
  • 20210428_230242.jpg
    20210428_230242.jpg
    314.5 KB · Views: 0
  • 20210428_230251.jpg
    20210428_230251.jpg
    339.5 KB · Views: 0
Ok, those are all normal.

What voltage are you seeing on:
1. VCore?
2. Vmem?
3. Vaux?
 
What do you get if you probe the gate pins on the 4C10Ns in frequency mode?
 
Whatever you do, DO NOT put it in the microwave!

That being said, there are enough complaints on these 390's to conclude that its possible that the solder has microfractures in it. You could try to Re-Ball it, in the oven or use a heatgun, but you must completely strip it down.
 
I have not fréquence to my multimeter
 

Attachments

  • 20210428_233401.jpg
    20210428_233401.jpg
    329 KB · Views: 0
No microwave🤣 i have bga
 

Attachments

  • 16196458836151716122877764220709.jpg
    16196458836151716122877764220709.jpg
    357.8 KB · Views: 0
Whatever you do, DO NOT put it in the microwave!

That being said, there are enough complaints on these 390's to conclude that its possible that the solder has microfractures in it. You could try to Re-Ball it, in the oven or use a heatgun, but you must completely strip it down.
This most likely isn't a BGA problem. He has no power on the memory power rail, which means there's likely a dead controller or something.

sorry i have frequency mode .
how can i test?
thank you for your times
Check for a frequency on the gate pin of all four FETs. On a working card, you'd have a value of 200Khz-500Khz.

If you have nothing, then you need to figure out why the controller isn't sending a gate signal, so you'd check for the proper input voltages on it. If you have a value, then the problem is likely the FETs, which are super easy to replace if you have a BGA machine.
 
i have nothing.
i have another cards for parts .
you think i do change two mosfet 4c10n with 0v?
 
i don t know if i use good frequency multimeter i never use it
i do check ground to fets for frequency?
 
To check for a frequency, place the black probe on ground, and the red probe on the gate pin. If you're getting nothing, then the problem most likely isn't the FETs themselves, but the control IC, so your next step would be to troubleshoot that.

On the 4C10N, the gate pin is the one all the way to the right when you look at the side with four pins.
 
I have nothing with all pin
 

Attachments

  • 16196478545142800888258051346687.jpg
    16196478545142800888258051346687.jpg
    187 KB · Views: 0
Ok but i have nothing to all 4 4c10n
 

Attachments

  • 16196481452358585129770009665580.jpg
    16196481452358585129770009665580.jpg
    230.7 KB · Views: 0
Ok but i have nothing to all 4 4c10n

The pin you want is at the other end of the row. On the right as you look at that picture.

If you have nothing on the gate pins, then your next step is to troubleshoot the controller. As I recall, it's on the back of the card. Find the datasheet and look up which pins are inputs, and check for a voltage at each one. If you have a voltage at each input, then the problem is likely the controller. If you're missing one of the inputs, then you need to figure out where it comes from.
 
Can you help for found contrôler?
 

Attachments

  • 20210429_003241.jpg
    20210429_003241.jpg
    489.5 KB · Views: 0
  • 20210429_003252.jpg
    20210429_003252.jpg
    500.1 KB · Views: 0
I have test and have 12v in
vcc (15)
Vccp (1)
Boot (4)
No other
 

Attachments

  • 16196518808963615954548521667295.jpg
    16196518808963615954548521667295.jpg
    715.3 KB · Views: 0
I have test and have 12v in
vcc (15)
Vccp (1)
Boot (4)
No other
You should double check that, just to be sure, but the problem is that you're missing the enable signal.

Your next step is to figure out why. You can go about this in a couple of different ways. One is to follow the traces around the board until you figure out what the enable pin is hooked up to. The other way is to just hunt for other circuits that aren't working and fix them until it starts working again. My guess is that you have another minor voltage rail that isn't working. My first suspect would be the 5V regulator (the big fat transistor looking thing on the back), but there is also a 3.3V rail, which is supplied by the PCI-E bus and may be fused, and maybe also a 7V rail.
 
You should double check that, just to be sure, but the problem is that you're missing the enable signal.

Your next step is to figure out why. You can go about this in a couple of different ways. One is to follow the traces around the board until you figure out what the enable pin is hooked up to. The other way is to just hunt for other circuits that aren't working and fix them until it starts working again. My guess is that you have another minor voltage rail that isn't working. My first suspect would be the 5V regulator (the big fat transistor looking thing on the back), but there is also a 3.3V rail, which is supplied by the PCI-E bus and may be fused, and maybe also a 7V rail.
Thank you i check to night and reply.
What is this in my picture 1?
Can i test that?
The big transistor in second picture i have test and in my memory i have 12v and 3v i check to night.
 

Attachments

  • 20210429_074529.jpg
    20210429_074529.jpg
    410.6 KB · Views: 0
  • 20210429_003252.jpg
    20210429_003252.jpg
    502.6 KB · Views: 0
You should double check that, just to be sure, but the problem is that you're missing the enable signal.

Your next step is to figure out why. You can go about this in a couple of different ways. One is to follow the traces around the board until you figure out what the enable pin is hooked up to. The other way is to just hunt for other circuits that aren't working and fix them until it starts working again. My guess is that you have another minor voltage rail that isn't working. My first suspect would be the 5V regulator (the big fat transistor looking thing on the back), but there is also a 3.3V rail, which is supplied by the PCI-E bus and may be fused, and maybe also a 7V rail.
I have test regulator and 2 fuse and it s ok.
Where is 3.3v rail and 7v rail?
Thank you
 

Attachments

  • 20210429_220048.jpg
    20210429_220048.jpg
    354.9 KB · Views: 0
I have test regulator and 2 fuse and it s ok.
Where is 3.3v rail and 7v rail?
Thank you
The 3.3V comes from the slot connector. There are a total of five pins, as I recall. You can look this up in the pinout, but it looks like you found the shunt for that rail, and it's working, so that's probably not your problem.

The next thing you should look for is additional power regulators. These are frequently SOP-8 packages, although the BIOS chips on this card are SOP-8 as well, so you'll need to identify what each one does and then check for proper function.
 
I have one in rear it is bios
And two on front
I have dual bios in this card it is possible 2 chip same référence is bios 1 front and bios 2 rear? ref: pm25ld010
Ce1445
Hr1n5cg
Other chip is ref: apl5930
Vr396
 

Attachments

  • 20210429_223757.jpg
    20210429_223757.jpg
    390.1 KB · Views: 0
  • 20210429_223807.jpg
    20210429_223807.jpg
    494.8 KB · Views: 0
I think have problem this i not ground
 

Attachments

  • 20210429_225416.jpg
    20210429_225416.jpg
    391.2 KB · Views: 0
  • 20210429_225219.jpg
    20210429_225219.jpg
    433.8 KB · Views: 0
I think have problem this i not ground
I'm pretty sure that's a BIOS chip. See if the resistance to ground changes when you switch the position of the BIOS switch.

The APL5930 is a regulator. What voltage are you seeing out of that?
 
Back
Top