Hello All,

I have a Gainward 1060GTX 6GB which works normal in Windows but the core clock still stays in idle mode.
I checked the phases of the card with an oszilloskope and there are 2 of the 3 GPU-phases dead.
I changed the 2 responsible Mosfets SIC632CD and also the Controller NCP81174 but the phases are still missing.
Due to the fact I have a second working Gainward 1060GTX6GB card I tried to measure the differencies around the NCP81174 Controller.
I found a 10kOhm-resistor below a transistor(connected to Base) which normally should have 36kOhms to Ground but on the defect card it has only around 4Ohms. This seems to be a short.
I desoldered the 10kOhm resistor and checked the resistance of the right leg to ground -> still 4 Ohms, also with desoldered NCP81174.
So I tried to find the way the trace takes to find the next part but after 2 vias I goes to maybe into the middle layer. Please see pictures attached.

Does anybody have a shematic of the Gainward 1060 or maybe another shematic with NCP81174 as controller to check from where are the transistor is controlled or connected

Thank you in advance.

Hello All,

I take up the topic again.
I now found the reaseon why only 1 phase works on the card. The NCP81174 has only PSI-Level of 1,35V which is below 1,4V and so the NCP81174 goes into 1-phase power saving mode.
This happen because the gate of the upper Transistor(name on the PCB: "SQ30") is pulled low by the line which is also connected to 5V pullup and to the DSBL#-pin of the non working 2 MOSFETs(SIC632). If I look at the Datasheet of the SIC632, the DSBL# pin will disable the MOSFET when DSBL# is low. But I cannot find any connected part to this DSBL# pin, except the Gate of the upper MOSFET, which is also pulled low.
SIC 632 Measurements:
EN: 5V
VCIN:5V
PWM: ~300KHz
DSBL#: Starts with 1,4V and falls down to 0,4V within a few seconds
Vswh: Starts at around 200KHz als falls down to 0KHz within a few seconds
Does anybody know a shematic of a GPU with SIC632 MOSFET, where I can look where the DSBL#-pin is also connected and who is pulling 5V to low? I don't think that it is the GPU because of the high voltage.

Ok, I measured again and desoldered some Resistors to get a better understanding of the circuit.
It is defintely the PSI-Circuit.
On a working card it has 36kOhm to GND but in my case it has 15KOhm. Maybe this is the reason why it iss pulled low or below 1,4V.
I desoldered a 0Ohm resistor and on the other side a 10KOhm Resistor. Now it has no connection to any IC/Transistor but still has 15KOhm to GND. but also 1KOhm to the GPIO16_FAN_PWM and GPIO13_FAN_TACH.
It seems that the 36KOHM -> 15KOhm short is coming from the GPU but this is hard to believe because it is not 0Ohm and also the GPU is woking with one phase in 139MHz-mode.
I cannot find any Mosfet Capacitor who is shorting the PSI because it is da very different PCB-Layout to the already known shematics.
I saw that Palit GTX 1060 has also the same PCB but I didnt found it here in the forum.
Does anybody have this Layout for Gainward or Palit GTX 1060?

---

Хорошо, я снова измерил и распаял несколько резисторов, чтобы лучше понять схему.
Это определенно PSI-Circuit.
На рабочей карте он имеет 36 кОм на массу, но в моем случае - 15 кОм. Может быть, это причина, по которой он опускается ниже 1,4 В.
Я распаял резистор 0 Ом, а с другой стороны - резистор 10 кОм. Теперь он не имеет связи с какой-либо микросхемой / транзистором, но все еще имеет 15 кОм на GND. но также 1 кОм для GPIO16_FAN_PWM и GPIO13_FAN_TACH.
Кажется, что короткое замыкание 36 кОм -> 15 кОм исходит от графического процессора, но в это трудно поверить, потому что это не 0 Ом, а также графический процессор работает с одной фазой в режиме 139 МГц.
Я не могу найти какой-либо конденсатор Mosfet, который закорачивает PSI, потому что это очень отличается от уже известной схемы.
Я видел, что у Palit GTX 1060 такая же печатная плата, но я не нашел ее здесь, на форуме.
У кого-нибудь есть такая раскладка для Gainward или Palit GTX 1060?