Overclock.net › Forums › AMD › AMD CPUs › Lapped CPU and waterblock = no difference
New Posts  All Forums:Forum Nav:

Lapped CPU and waterblock = no difference

post #1 of 3
Thread Starter 
Last night I spent well over 3 hours lapping my cpu and waterblock. I started with 1500 and then 2000 grit, being very careful. After a long time and a very sore wrist/arm I managed to get both the waterblock and CPU to a mirror finish.

This is on my 965BE running at 4.2ghz @ 1.56v. Shin-itsu TIM was used before and after. P95 blend running for 10 mins gave the load temps.

Before:
Ambient air temp = 17.0C
idle temp =28C
Load temp = 49C

After
Ambient air temp = 17.4C
idle temp =29C
Load temp = 49C

So anyway, what's the deal? I ran it overnight just to see if the TIM needed to set/cure (which they say it does not) and noted identical results this morning. So it LOOKS as if lapping made no difference for me.

Thoughts/questions:

Could it be the shin-itsu is just that good with rough/irregular surfaces that you won't see a difference?

Maybe I would have noticed a difference if using a cheaper compound like AS5?

Maybe the waterblock is limited by the waterflow or cooling capacity of the radiator, so better thermal transfer at the waterblock does not result in an increase in cooling performance?
    
CPUMotherboardGraphicsRAM
965BE @ 4.24Ghz (223x19) 1.56v Crosshair IV Formula MSI GTX465 unlocked to 470 G.skill 8gb DDR3 1600mhz cas 9 @ 7-8-7-15 
Hard DriveOptical DriveOSMonitor
Intel SSD X-25M 80gb Hitachi 7200RPM 500GB Win7 home premium 64 bit ASUS 21.5" LED 2ms GTG 
KeyboardPowerCaseMouse
Cheapo Antec Truepower New 750w None Logitech Performance MX (awesome) 
  hide details  
Reply
    
CPUMotherboardGraphicsRAM
965BE @ 4.24Ghz (223x19) 1.56v Crosshair IV Formula MSI GTX465 unlocked to 470 G.skill 8gb DDR3 1600mhz cas 9 @ 7-8-7-15 
Hard DriveOptical DriveOSMonitor
Intel SSD X-25M 80gb Hitachi 7200RPM 500GB Win7 home premium 64 bit ASUS 21.5" LED 2ms GTG 
KeyboardPowerCaseMouse
Cheapo Antec Truepower New 750w None Logitech Performance MX (awesome) 
  hide details  
Reply
post #2 of 3
Quote:
Originally Posted by LocoJason View Post
Last night I spent well over 3 hours lapping my cpu and waterblock. I started with 1500 and then 2000 grit, being very careful. After a long time and a very sore wrist/arm I managed to get both the waterblock and CPU to a mirror finish.

This is on my 965BE running at 4.2ghz @ 1.56v. Shin-itsu TIM was used before and after. P95 blend running for 10 mins gave the load temps.

Before:
Ambient air temp = 17.0C
idle temp =28C
Load temp = 49C

After
Ambient air temp = 17.4C
idle temp =29C
Load temp = 49C

So anyway, what's the deal? I ran it overnight just to see if the TIM needed to set/cure (which they say it does not) and noted identical results this morning. So it LOOKS as if lapping made no difference for me.

Thoughts/questions:

Could it be the shin-itsu is just that good with rough/irregular surfaces that you won't see a difference?
Yes
Quote:

Maybe I would have noticed a difference if using a cheaper compound like AS5?
Maybe, probably
Quote:
Maybe the waterblock is limited by the waterflow or cooling capacity of the radiator, so better thermal transfer at the waterblock does not result in an increase in cooling performance?
If the bottleleck isn't with the thermal transfer rate between the CPU and the waterblock, then you may be limited by waterflow, or airflow thru the rad, or static pressure of the pump.

Side question, has the water temperature increased after lapping? If yes it would mean you're probably waterflow/airflow limited, if no then it means you simply didn't add a significant improvement to the thermal transfer rate between the CPU and the waterblock.
Centurion
(14 items)
 
  
CPUMotherboardGraphicsRAM
Phenom II 940 @3.3GHz 1.25V Gigabyte 790X-DS4 XFX 5850 8GB OCZ DDR2-800 
Hard DriveCoolingOSMonitor
Corsair Force 3 TRUE Win 7 Pro x64 ACER 23" TN  
PowerCase
400W Silverstone Strider Antec P182 
  hide details  
Reply
Centurion
(14 items)
 
  
CPUMotherboardGraphicsRAM
Phenom II 940 @3.3GHz 1.25V Gigabyte 790X-DS4 XFX 5850 8GB OCZ DDR2-800 
Hard DriveCoolingOSMonitor
Corsair Force 3 TRUE Win 7 Pro x64 ACER 23" TN  
PowerCase
400W Silverstone Strider Antec P182 
  hide details  
Reply
post #3 of 3
Thread Starter 
Quote:
Originally Posted by dragosmp View Post
If the bottleleck isn't with the thermal transfer rate between the CPU and the waterblock, then you may be limited by waterflow, or airflow thru the rad, or static pressure of the pump.

Side question, has the water temperature increased after lapping? If yes it would mean you're probably waterflow/airflow limited, if no then it means you simply didn't add a significant improvement to the thermal transfer rate between the CPU and the waterblock.
I did not measure the water temp before and after, though in hindsight I probably should have thought about measuring it beforehand. All it would take is a lone RTD in the res or in the hose right after the cpu.

After reading the thread and other comments on lapping the cpu, it got my hopes up for being able to drop temps even more. Now I feel kind of like lapping is really overrated and even a complete waste of time. Maybe this can aid someone else because I wish I had read feedback from someone who had done the same test, as opposed to "advice" from people who thought it made a huge difference.

I bought some foam insulation and tonight I plan to add some PG coolant to the system and put the rad/pc outside in a "cold box". It's -6C outside so I should have cpu temps just above freezing. This will give me an idea what I can hit with a small TEC chiller in the system (should I deem it a worthwhile project).

Next step after that is high speed push/pull fan testing and then I'll eventually try a MCP350 pump. People also seem to say the XSPC 750 pump isn't that great. With colder water, the flow rate/pressure might fall off even more.
    
CPUMotherboardGraphicsRAM
965BE @ 4.24Ghz (223x19) 1.56v Crosshair IV Formula MSI GTX465 unlocked to 470 G.skill 8gb DDR3 1600mhz cas 9 @ 7-8-7-15 
Hard DriveOptical DriveOSMonitor
Intel SSD X-25M 80gb Hitachi 7200RPM 500GB Win7 home premium 64 bit ASUS 21.5" LED 2ms GTG 
KeyboardPowerCaseMouse
Cheapo Antec Truepower New 750w None Logitech Performance MX (awesome) 
  hide details  
Reply
    
CPUMotherboardGraphicsRAM
965BE @ 4.24Ghz (223x19) 1.56v Crosshair IV Formula MSI GTX465 unlocked to 470 G.skill 8gb DDR3 1600mhz cas 9 @ 7-8-7-15 
Hard DriveOptical DriveOSMonitor
Intel SSD X-25M 80gb Hitachi 7200RPM 500GB Win7 home premium 64 bit ASUS 21.5" LED 2ms GTG 
KeyboardPowerCaseMouse
Cheapo Antec Truepower New 750w None Logitech Performance MX (awesome) 
  hide details  
Reply
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: AMD CPUs
Overclock.net › Forums › AMD › AMD CPUs › Lapped CPU and waterblock = no difference