Overclock.net › Forums › AMD › AMD Motherboards › AMD Motherboards - VRM info database
New Posts  All Forums:Forum Nav:

AMD Motherboards - VRM info database - Page 113

post #1121 of 1209
Quote:
Originally Posted by perche View Post

ga-790x-ds4 + amdx61090t @ 4ghz + intel burn test. = a bit of flame and bad smell but the board still working.

Eeek!

What CPU cooler are you running, and what sort of airflow is there over the MOSFETS and associated circuitry?
post #1122 of 1209
Quote:
Originally Posted by Geonerd View Post

7.5 mOhm is not all that fantastic. The dreaded NIKOS on my 870A-G54 are rated* to operate in the ballpark of 6~9 mOhm, depending on actual current flow and voltage, and gate voltage.

They die not because of the actual FET, but because the driver part overloads, starts to improperly drive the FET part of the package, as such the FET doesn't turn on or off quickly enough, generates a crapload of heat, and explodes.

I've had this happen when driving high power FETs with a 555 at high frequencies. I'm talking about huge ass 65A @700V, properly heatsinked FETs burning with only 0.1A @300V going through.
   
AGP bencher
(14 items)
 
CPUMotherboardGraphicsRAM
Phenom X6 1055T ASUS M4A89GTD Pro/USB3 Sapphire HD 6950 2GB Kingston HyperX DDR3-2000 
Hard DriveHard DriveCoolingCooling
Seagate 7200.12 Samsung SpinPoint F3 EK Supreme HF EK CoolStream XT360 
CoolingCoolingOSMonitor
Alpahcool Coolplex Pro 10LT Phobya DC12-400 Windows 7 Ultimate & Fedora 17 LG IPS235V 
MonitorKeyboardPowerCase
LG E2250V Logitech G11 NOX Sonar 780W (Kingwin 750W 80+) Lian Li PC-A05NB 
MouseAudio
Logitech G9 ASUS Xonar DX 
CPUMotherboardGraphicsRAM
AMD A10-5700 Gigabyte F2A75M-HD2 Radeon 7660D G.SKILL Ares 2133 CL9 
Hard DriveCoolingOSOS
Hitachi 5K750 Noctua NH-L9a Fedora 18 Windows 7 x64 
MonitorKeyboardPowerCase
AUO B156HW01 Logitech MK220  PicoPSU-80-WI-25V AIO Aluminium Handmade 
Other
TP-LINK 150 N/G/B 
CPUCPUCPUMotherboard
Core2Duo E6400 Core2Quad Q6600 Pentium Dual Core E5200 AsRock 4COREDUAL-SATA2 R2.0 
GraphicsRAMHard DriveOptical Drive
A dumpload of ancient AGP cards Kingston Value DDR2-667 CL4 2T @CL3 1T Seagate 160GB 7200.10 LG IDE DVD-ROM 
CoolingCoolingOSMonitor
Ghettomade CPU waterblock 49cc 2stroke engine copper radiator WinXP SP2 32bit ProView 17" 
PowerCase
Tacens Radix V 550W Ghetto aluminium bench 
  hide details  
Reply
   
AGP bencher
(14 items)
 
CPUMotherboardGraphicsRAM
Phenom X6 1055T ASUS M4A89GTD Pro/USB3 Sapphire HD 6950 2GB Kingston HyperX DDR3-2000 
Hard DriveHard DriveCoolingCooling
Seagate 7200.12 Samsung SpinPoint F3 EK Supreme HF EK CoolStream XT360 
CoolingCoolingOSMonitor
Alpahcool Coolplex Pro 10LT Phobya DC12-400 Windows 7 Ultimate & Fedora 17 LG IPS235V 
MonitorKeyboardPowerCase
LG E2250V Logitech G11 NOX Sonar 780W (Kingwin 750W 80+) Lian Li PC-A05NB 
MouseAudio
Logitech G9 ASUS Xonar DX 
CPUMotherboardGraphicsRAM
AMD A10-5700 Gigabyte F2A75M-HD2 Radeon 7660D G.SKILL Ares 2133 CL9 
Hard DriveCoolingOSOS
Hitachi 5K750 Noctua NH-L9a Fedora 18 Windows 7 x64 
MonitorKeyboardPowerCase
AUO B156HW01 Logitech MK220  PicoPSU-80-WI-25V AIO Aluminium Handmade 
Other
TP-LINK 150 N/G/B 
CPUCPUCPUMotherboard
Core2Duo E6400 Core2Quad Q6600 Pentium Dual Core E5200 AsRock 4COREDUAL-SATA2 R2.0 
GraphicsRAMHard DriveOptical Drive
A dumpload of ancient AGP cards Kingston Value DDR2-667 CL4 2T @CL3 1T Seagate 160GB 7200.10 LG IDE DVD-ROM 
CoolingCoolingOSMonitor
Ghettomade CPU waterblock 49cc 2stroke engine copper radiator WinXP SP2 32bit ProView 17" 
PowerCase
Tacens Radix V 550W Ghetto aluminium bench 
  hide details  
Reply
post #1123 of 1209
Quote:
Originally Posted by Artikbot View Post

They die not because of the actual FET, but because the driver part overloads, starts to improperly drive the FET part of the package, as such the FET doesn't turn on or off quickly enough, generates a crapload of heat, and explodes.

I've had this happen when driving high power FETs with a 555 at high frequencies. I'm talking about huge ass 65A @700V, properly heatsinked FETs burning with only 0.1A @300V going through.

What causes the driver to overload? Perhaps the FETs response time or input capacitance starts to increase as it heats up and/or ages? None of the product datasheets I've seen mention this, but it seems possible. (?)
post #1124 of 1209
Just a little note for the Gigabyte GA-78LMT-USB3, the revision that I got new was 4.1, and it came with the F3 BIOS, in case someone which BIOS one could get off the shelf. Thought it was important since it is that BIOS that begins to support 125W CPUs.
post #1125 of 1209


cpu cooler is a scyte mugen 2, and 4 scythe gentle thypon at 1850 rpm


and yes still working, i have now in it my old phenom2 940 be and i bought a 990xa-ud3 for my 1090t
post #1126 of 1209
Quote:
Originally Posted by Artikbot View Post

They die not because of the actual FET, but because the driver part overloads, starts to improperly drive the FET part of the package, as such the FET doesn't turn on or off quickly enough, generates a crapload of heat, and explodes.

I've had this happen when driving high power FETs with a 555 at high frequencies. I'm talking about huge ass 65A @700V, properly heatsinked FETs burning with only 0.1A @300V going through.

I thought so. In most pictures, the failed FET is one of the third FETs replacing the driver. It is mostly these three-transistor configs that are failure prone - even if there are quality ICs. The GA-MA790X-DS4 is a good example. They are newer low RDS (on) ICs but the third transistor layout (as well as the low phase count) causes easy overheating.

 

The person who recently posted with the MA-790X-DS4 failure should have properly cooled the VRMs (the setup with the mugen 2 provides no VRM cooling, so you're trying to push an x6 with a low-quality, 4+1 phase setup.... that's failure prone in the same way as the MSI 870A boards).

 

Quality drivers/transistor layout is key in creating a VRM system that doesn't explode.

Hexagonal
(16 items)
 
Voyager [XT875]
(11 items)
 
 
CPUMotherboardGraphicsRAM
ARM Cortex A9 dual core [1.2Ghz]  TI OMAP 4430 PowerVR SGX 540 1GB low power DDR2 
Hard DriveHard DriveOSMonitor
16GB on board storage 32GB Class 10 MicroSDHC card Android 4.0.4 "Ice Cream Sandwich" (LiquidSmoot... 4.3" 960x540 TFT capacitive multi-touch screen 
KeyboardPowerCase
Swiftkey 3 Battery Motorola DROID BIONIC 
  hide details  
Reply
Hexagonal
(16 items)
 
Voyager [XT875]
(11 items)
 
 
CPUMotherboardGraphicsRAM
ARM Cortex A9 dual core [1.2Ghz]  TI OMAP 4430 PowerVR SGX 540 1GB low power DDR2 
Hard DriveHard DriveOSMonitor
16GB on board storage 32GB Class 10 MicroSDHC card Android 4.0.4 "Ice Cream Sandwich" (LiquidSmoot... 4.3" 960x540 TFT capacitive multi-touch screen 
KeyboardPowerCase
Swiftkey 3 Battery Motorola DROID BIONIC 
  hide details  
Reply
post #1127 of 1209
Quote:
Originally Posted by Geonerd View Post

What causes the driver to overload? Perhaps the FETs response time or input capacitance starts to increase as it heats up and/or ages? None of the product datasheets I've seen mention this, but it seems possible. (?)

Excess of temperature, most likely. As a small temperature increase makes a cascade effect.

The MOSFET becomes a bit too hot, the driver section starts to decrease output power to compensate for this excessive temperature (due to thermal protection of the own package), leading to an improperly driven MOSFET, which in turn heats up even more, heating up even more the driver, which literally fails to reach the logic level of the FET gate.
Then the FET stays in a half-on or half-off state, generating horrible amounts of heat due to not being fully on or off, heating the whole package until it blows up.

That's the main issue with DrMOS packages. Also, if one of the MOSFETs shuts down entirely due to excessive heat, the remaining ones risk becoming overloaded, and if the thermal shutdown isn't designed for the whole MOSFET array, there's a very high chance of the remaining powered array exploding due to overheat/overcurrent.

Also note that the resistance of the entire assembly is higher when temperatures rise.
   
AGP bencher
(14 items)
 
CPUMotherboardGraphicsRAM
Phenom X6 1055T ASUS M4A89GTD Pro/USB3 Sapphire HD 6950 2GB Kingston HyperX DDR3-2000 
Hard DriveHard DriveCoolingCooling
Seagate 7200.12 Samsung SpinPoint F3 EK Supreme HF EK CoolStream XT360 
CoolingCoolingOSMonitor
Alpahcool Coolplex Pro 10LT Phobya DC12-400 Windows 7 Ultimate & Fedora 17 LG IPS235V 
MonitorKeyboardPowerCase
LG E2250V Logitech G11 NOX Sonar 780W (Kingwin 750W 80+) Lian Li PC-A05NB 
MouseAudio
Logitech G9 ASUS Xonar DX 
CPUMotherboardGraphicsRAM
AMD A10-5700 Gigabyte F2A75M-HD2 Radeon 7660D G.SKILL Ares 2133 CL9 
Hard DriveCoolingOSOS
Hitachi 5K750 Noctua NH-L9a Fedora 18 Windows 7 x64 
MonitorKeyboardPowerCase
AUO B156HW01 Logitech MK220  PicoPSU-80-WI-25V AIO Aluminium Handmade 
Other
TP-LINK 150 N/G/B 
CPUCPUCPUMotherboard
Core2Duo E6400 Core2Quad Q6600 Pentium Dual Core E5200 AsRock 4COREDUAL-SATA2 R2.0 
GraphicsRAMHard DriveOptical Drive
A dumpload of ancient AGP cards Kingston Value DDR2-667 CL4 2T @CL3 1T Seagate 160GB 7200.10 LG IDE DVD-ROM 
CoolingCoolingOSMonitor
Ghettomade CPU waterblock 49cc 2stroke engine copper radiator WinXP SP2 32bit ProView 17" 
PowerCase
Tacens Radix V 550W Ghetto aluminium bench 
  hide details  
Reply
   
AGP bencher
(14 items)
 
CPUMotherboardGraphicsRAM
Phenom X6 1055T ASUS M4A89GTD Pro/USB3 Sapphire HD 6950 2GB Kingston HyperX DDR3-2000 
Hard DriveHard DriveCoolingCooling
Seagate 7200.12 Samsung SpinPoint F3 EK Supreme HF EK CoolStream XT360 
CoolingCoolingOSMonitor
Alpahcool Coolplex Pro 10LT Phobya DC12-400 Windows 7 Ultimate & Fedora 17 LG IPS235V 
MonitorKeyboardPowerCase
LG E2250V Logitech G11 NOX Sonar 780W (Kingwin 750W 80+) Lian Li PC-A05NB 
MouseAudio
Logitech G9 ASUS Xonar DX 
CPUMotherboardGraphicsRAM
AMD A10-5700 Gigabyte F2A75M-HD2 Radeon 7660D G.SKILL Ares 2133 CL9 
Hard DriveCoolingOSOS
Hitachi 5K750 Noctua NH-L9a Fedora 18 Windows 7 x64 
MonitorKeyboardPowerCase
AUO B156HW01 Logitech MK220  PicoPSU-80-WI-25V AIO Aluminium Handmade 
Other
TP-LINK 150 N/G/B 
CPUCPUCPUMotherboard
Core2Duo E6400 Core2Quad Q6600 Pentium Dual Core E5200 AsRock 4COREDUAL-SATA2 R2.0 
GraphicsRAMHard DriveOptical Drive
A dumpload of ancient AGP cards Kingston Value DDR2-667 CL4 2T @CL3 1T Seagate 160GB 7200.10 LG IDE DVD-ROM 
CoolingCoolingOSMonitor
Ghettomade CPU waterblock 49cc 2stroke engine copper radiator WinXP SP2 32bit ProView 17" 
PowerCase
Tacens Radix V 550W Ghetto aluminium bench 
  hide details  
Reply
post #1128 of 1209
Can someone explain to me why the gigabyte 78lmts2p rev 5 has a black no instead of the red no under heatsinked, I'm interested seeing as how i have it myself.
post #1129 of 1209
Gigabyte 78LMT-S2 rev 1.1 (besides being 3+1,lol) is reported to have Chinese capacitors. Wish I had known, though purchased to replace dead 880 motherboard with Athlon II X2 250.
Aparently cheap is becoming bad even from name brands.

I would guess color of numbers is either a cosmetic update by Gigabyte, or by the motherboard producer.
Edited by brasslad - 1/19/13 at 4:09am
post #1130 of 1209
Quote:
Originally Posted by brasslad View Post

Gigabyte 78LMT-S2 rev 1.1 (besides being 3+1,lol) is reported to have Chinese capacitors. Wish I had known, though purchased to replace dead 880 motherboard with Athlon II X2 250.
Aparently cheap is becoming bad even from name brands.

I would guess color of numbers is either a cosmetic update by Gigabyte, or by the motherboard producer.

Perhaps, can anyone confirm this who actually knows specifically what it is?
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: AMD Motherboards
Overclock.net › Forums › AMD › AMD Motherboards › AMD Motherboards - VRM info database