Basic: 5 minutes GPU Service | NVIDIA GT630

Project Dark Mist | PC Mod & Reborn

HY510, K4-PRO & K5-PRO Thermal Comparison

HY510, K4-PRO & K5-PRO Thermal Comparison

In today article I will discuss 2 different brand of thermal paste I bought, their claimed performance and product property, finally my test bench result and final conclusion. Before I started the science experiment, I have to say I totally regret to purchase the “K” series product and I will tell you why in the following paragraph.

  2 different thermal paste I purchased at the different time, both purchased on ebay and their shipping and delivery time around 30days arrived my door with the good packaging. The product property and thermal conductive performance of K series thermal paste is mention on their website. K4-PRO has thermal conductivity K>4.6 W/m.K and K5-PRO has thermal conductivity K>5.3 W/m.K both are no electrical conductivity, they also claimed the thermal conductivity is 3 times higher than common thermal pastes and pads.
http://www.computer-systems.gr/content/product/k5-pro-k4-pro-set

http://www.computer-systems.gr/content/product/k5-pro-k4-pro-set

HY510 has thermal conductivity of K>1.93 W/m.K but did not mention any information about electric conductivity, however, tested the result is non conductivity as well.
Product property stated at ebay as well. http://www.ebay.co.uk/itm/HY510-30g-Grey-Thermal-Conductive-Grease-Paste-For-CPU-GPU-Chipset-Cooling-UK-/192120683326?hash=item2cbb48fb3e:g:4ccAAOSwCU1Yu3NT

  After I introduced the product information, this is the test bench information I should give for better reference. CPU = E2140 1.6GHz, stock heatsink crew mounted method. Motherboard = Gigabyte G31M-S2L. RAM = 1GB 667MHz. No PCIe and No PCI in use.

  Started the testing at room temperature around 28-29°C with well air flow condition during summer night. Stress tested the CPU by using HeavyLoad application to run about 10minutes for each test. First, without unscrew anything on my test bench, the current thermal paste at the time was HY510, so I ran HeavyLoad at the first time, ended my first test of HY510 = 63°C.

 Unscrewed the CPU heatsink, had a looked at the CPU and heatsink surface, the paste was spread out very well covered, greasy and easy to remove thanks to its silicon based. I cleaned up everything by using Polylab contains alcohol 95%.

 

  Second test, I applied K4-Pro to the CPU, ran the same test again. Ended my second test of K4-Pro = 69°C.

  Unscrewed the CPU heatsink, had a looked at the CPU and heatsink surface, the paste was spread out very well covered and this time it was so thin thanks to the low viscosity but harder. After that I cleaned up everything by using Polylab contains alcohol 95%.

  Test No.3, I applied K5-Pro to the CPU, ran the same test again. Ended my test of K5-Pro = 72°C.

  Unscrewed the CPU heatsink, had a looked at the CPU and heatsink surface, the paste was spread out very well covered but it was thick due to the high viscosity for thermal pad replacement and it is harder than HY510 as well. After that I cleaned up everything by using Polylab contains alcohol 95%.

  Finally, I ended up with all the disappointment.

I repeated the same cleaning process, applied HY510 and repeated the same stress test for a few times. Eventually still got me the same result HY510 = 63°C at the same ambient temperature with every factor stayed controlled variable but thermal paste!

So, why am I regret? I paid another RM40 for a lower performance item! I spent 1.5 hour for this experiment just to make sure if there was other factor cause the unsatisfied result, in these tests, the HY510 the lower cost item completely stand out. Why HY510 has lower thermal conductivity than K4 and K5 but has a better performance?

So why HY510 has lower thermal conductivity than K4 and K5 but has a better performance?