I am in the process of building my first computer and before I installed my CPU and heatsink I decided to do some research into the thermal compound... boy am I glad I did! I discovered the practice of "lapping" as well as this interesting review of thermal compounds: http://www.xtremesystems.org/forums/showthread.php?t=23...
which seems to strongly endorse "Indigo Xtreme"
so my question is basically this: since the Indigo Xtreme actually melts during installation, is there any point to lapping my i7 or Vigor Monsoon III LT, if I plan on getting the Indigo Xtreme?
Prolimatech does not condone any type of lapping done to the CPU or to heatsink base. Every Prolimatech's heatsink base is designed on a pin-point scale of how the base is to be flat and/or curved where it's needed to be. We have programed our machines to machine the surface in a very calculated way. Any after-manufacture lapping or modding done to the base will alter the design, hence negating its performance factor as well as its warranty.
Lapping doesnt always work. Too many factors to consider. Quality of heat sink, and smoothness/flatness of the heatsink.
For low quality heat sinks, lapping most definitely helps most of the time.
In the process there's several more factors after the resolution of whether a heat sink needs lapping or not (most resolutions can be taken to Frostytech.com for they're analysis on the smoothness/flatness of the heat sink. If no review exists, either venture into the unknown or suck it up:
1. Heat sink surface's design. Is it 4 pips together with spacing (can create uneven surfaces)
2. Even pressure upon the heatink/grit paper not to mention speed as too fast of a grind could end up messing up the heat sink, and too slow would end up being less effective.
3. Are you starting with the correct size grit.
4. Using the grit paper for an adequate amount of time before going to a smaller size grit.
5. Proper work environment. It takes a serious amount of time to lap a heat sink and or processor or both. Make sure it's a comfortable area to work with, with as little moisture/dust as possible. Also make sure to take breaks, as tiring your arms can produce uneven pressure/thus producing an curved or uneven surface.
6. Unknown variables. Try to factor in as many variables/and set them up in a controlled environment (dust/moisture=bad idea in basement or near A/C vents.
The reason you want to be careful is because when dealing with processors, a bad lapping session could end up costing you the CPU since warranty would be plain gone.
I have not lapped any heat sink or processor, nor will I simply because I don't need it, not to mention I doubt I have the patience for such a length job for a few degrees Celsius and no security if I mess up it will okay. If I require even further thermal dissipation I would just go with water cooling.
I had an E6600 that needed a lap job because the IHS was horribly machined. It couldn't even contact the heat sink base properly. This is pretty rare though, and is on the extreme end of not-so-flat surfaces. Most of the time lapping could shave off a few degrees. In this case, I was looking at a 10-15C reduction in temp.
Flowing into gaps causes a larger distance that heat must travel before reaching the heatsink, which is not ideal. Ideally (assuming you still use TIM in the ideal scenario), you want two perfectly flat surfaces with a layer of TIM no thicker than a single particle of TIM at any point in between.
Randomizer, you are correct wrt to the ideal, but of course the real world doesn't give us ideal, and as a practical matter we can approximate ideal only somewhat. Lapping absolutely will improve performance, but in the case of Indigo Xtreme the improvement is so small as to be virtually unmeasurable.
Why? The reason has to do with the difference between the two kinds of thermal resistance characteristic of TIMs. (I apologize in advance if, as a newcomer here, I now belabor by this explanation something already well known to most people here.) Bulk thermal resistance refers to the resistance of a material to conduct heat within itself, while contact (or interface) resistance refers to the resistance of heat being conducted between the two surfaces of adjacent materials.
The bulk conductivity (the inverse of bulk resistance) of Indigo is ~20WmK, anywhere between 2-10x better than conventional grease-based TIMs. Thus thinner is generally better for the bond line thickness; however, as the bulk conductivity increases (i.e. as resistance decreases), overall thermal resistance is dominated by the contact resistance component.
Contact resistance depends of many factors and the theory is complicated. As a practical matter low contact resistance is achieved by high surface wetting, i.e. the ability to force the TIM to directly contact the surface (whether chip/chip package or heat sink/water block.) High wetting and low contact resistance are achieved in many ways, including the rheology of the TIM but also, significantly, the displacement of trapped air, a great thermal insulator.
As an aside, this is why high clamping forces are often used to lower thermal resistance. In part it is to create a thinner bond line, but the biggest component is squeezing out trapped air and compressing trapped air in the surface asperities; both increase the wetting, i.e. the direct contact of the TIM with the surface. (This is also why Indigo Xtreme does not require a high clamping force and why its performance does not improve with higher pressure.)
To the extent you can perfectly lap a surface you can eliminate the air trapped in surface asperities, but you cannot eliminate the air trapped between the TIM and the surface that is created by the simple application of the TIM directly to the surface.
Indigo Xtreme is not applied directly but instead reflows across the mated surfaces, filling asperities and displacing air as it goes. The result is an interface largely free of trapped air, and hence high wetting and low contact resistance. That, combined with the high bulk conductivity, is what gives Indigo Xtreme its high performance.
Lapping has greater effect on conventional TIMs because it lessens the trapped air. With Indigo Xtreme the effect is small because the difference in trapped air is miniscule, and the high bulk conductivity makes the bond line thickness reduction (more real in theory than practice) rather moot.
Lap if you want, but expermentally it makes little difference with Indigo Xtreme.
I hope this explanation is useful. If you use Indigo Xtreme and have found lapping provides a measurable performance increase, please post details; I would be very interested in learning about such a result.
Randomizer, I absolutely love your cat avatar. Bitchin'!!
Indigo does so well because it flows removing the chance of bubbles (NOT AN ISSUE WITH PROPER TIM APPLICATION TO START WITH) and it's incerdible high heat conductivity. It's thermal resistance is as close as we can get to zero unless you work at NASA and have an 'unobtainium sample' to work with.
You might get a decrease with lapping but it will be within the 1C tolerance of the sensors on the CPU. So why bother? Unless you got a ballon like CPU and a bottle bottom like HS bottom.
So, when i go to apply my indigo extreme on friday should i have my waterblock cranked down as tight as possible, then crank it down even more once the correct temp is reached while on a level surface?
im sure the product comes with a long list of directions, BUT I WANNA KNOW NOW!