The Age-Old Question of Break-In Time for Thermal Paste: A Response to Comments
As we dive into this video about thermal paste, I want to address the concerns that have been raised by some of my viewers. Specifically, they've mentioned criticisms of the methods used to test these thermal compounds, and one comment caught my attention in particular from Cory Philger. He pointed out that Arctic Silver 5, according to their instructions, has a break-in period of 200 hours for both AMD and Intel processors.
This got me thinking about the theory behind break-in time when testing thermal compounds. The idea is that there's a prescribed amount of time required until the paste reaches maximum effectiveness, which in turn nullifies the validity of tests whose procedures include swapping out different thermal compounds in succession without giving them a chance to "break in." In other words, if you're not allowing the thermal compound to reach its full potential before testing it, you might be getting inaccurate results.
The manufacturer of Arctic Silver 5, Arctic Materials, has addressed this issue directly on their website. They provide an important reminder that outlines the phenomenon of break-in time for their conductive matrix. According to this warning, it will take up to 200 hours and several thermal cycles for the particles in the Arctic Silver 5's conductive matrix to achieve maximum particle-to-particle thermal conduction and for the heatsink to CPU interface to reach maximum conductivity.
Arctic Materials also notes that during this break-in period, the measured temperature will often drop by 2C to 5C. This is a crucial point to consider when testing thermal compounds, as it highlights the importance of allowing the paste to "break in" before drawing conclusions about its effectiveness.
This brings us full circle back to the issue at hand: how do we test thermal compounds accurately? In this video, we'll be exploring ways to test thermal paste that take into account the break-in period and ensure that our results are reliable. Whether you're a seasoned enthusiast or just starting out on your journey, I invite you to join me as we delve into the world of thermal paste and explore the latest developments in this fascinating field.
Important Reminder: Arctic Silver 5 Break-In Period
As mentioned earlier, the manufacturer of Arctic Silver 5 has provided an important reminder about the break-in period for their conductive matrix. This warning is essential reading for anyone looking to understand how this thermal compound works and how to test it accurately.
According to Arctic Materials, "Due to the unique shape and sizes of the particles in Arctic Silver 5's conductive matrix, it will take up to 200 hours and several thermal cycles to achieve maximum particle-to-particle thermal conduction and for the heatsink to CPU interface to reach maximum conductivity." This warning highlights the importance of allowing the paste to "break in" before testing its effectiveness.
It also mentions that during this break-in period, the measured temperature will often drop by 2C to 5C. This is a critical point to consider when testing thermal compounds, as it emphasizes the need for patience and persistence when evaluating their performance. By understanding the break-in period and how it affects the results, we can ensure that our tests are reliable and accurate.
In the next section, we'll explore ways to test thermal paste that take into account the break-in period and provide guidance on how to evaluate its effectiveness in a way that's both accurate and reliable.
"WEBVTTKind: captionsLanguage: enAlright, It’s time for another thermal paste video and this is in response to several comments on my videos about thermal paste which are criticizing the methods in which I and others like myself use to test these thermal compounds. Such as this comment from Cory Philger, He says... “Arctic silver 5, per their instructions has a break in period of 200 hrs for both AMD and intel. This being done by using the computer normally and turning it off when done, then back on again etc. Supposedly this drops temps from 1-5c after break in time.” so, it’s finally time to tackle the age-old question of break in time for thermal paste. The theory is, when testing thermal compounds there is a prescribed amount of time that is required until the paste reaches maximum effectiveness. Which in theory nullifies the validity of tests whose procedures include swapping out different thermal compounds in succession without giving them a change to “Break in”.In this case we are specifically looking at Arctic Silver 5 and conveniently you can find directly on the manufacturers website an important reminder that outlines this phenomenonImportant Reminder: “Due to the unique shape and sizes of the particles in Arctic Silver 5's conductive matrix, it will take up to 200 hours and several thermal cycles to achieve maximum particle to particle thermal conduction and for the heatsink to CPU interface to reach maximum conductivity.” It then goes on to say… “the measured temperature will often drop 2C to 5C over this \"break-in\" period. This break-in will occur during the normal use of the computer as long as the computer is turned off from time to time and the interface is allowed to cool to room temperature. Once the break-in is complete, the computer can be left on if desired.” So, at least with arctic silver 5 we have relatively clear parameters on how we should test this theory out. And what we want to see exactly is how the temperature will change over time, specifically the time that the PC is actually in use. So, About 2 months ago I put a fresh dollop of arctic silver 5 on my Ryzen 3700X and over the course of 2 months I painstakingly took notes of how long I used my PC everyday, turning it off when not in use. I tested once a week with the first test being performed immediately after applying the thermal paste. Before each test I fired up the machine and let it idle for 10 minutes so that the machine could warm up.The tests include Cinebench R23 so we can get a look at the thermal limits and get an idea if the CPU is performing better over time. For a gaming Benchmark I have come to love the simplicity of F1s Benchmark mode, which I have chosen to use this time. The F1 2020 version because F1 2021 benchmark is also grood but at the time I started this test that game wasn’t out yet so we’re sticking with F1 2020 All temps as usual captured with HWiNFO using the CPU CCD1 (Tdie) reading So without further delay let’s have a look at the results Cinebench R23 Starting with hour 0 we have an average temp of 75.5, Which happens to be the lowest temperature in the series of tests. As we move along roughly 30-40 hours of PC usage time a week we see the temps rise up to its peak on the 3rd test @ 77.8 deg c and then settle down for the remaining weeks with surprisingly consistent results. As for the Cinebench multi core score they all came within 150 points of each other with no strong correlation other than the possibility that the slightly warmer temps resulted in a higher score though there could be some other underlying factor causing these differences. Either way I find the differences to be negligible.Coming to the results from our F1 2020 benchmark On Ultra 1440p 5 laps in Spa which is about a 12 minute test. We see a very similar result to that of cinebench, at least in the progression of temperature changes over time. The 0 hour once again the lowest by just .2 degrees @ 56.8 deg c while the highest temp reached during the 7th test which comes right after the paste has passed the 200 hour mark. Again, no strong difference, only a marginal upward trend from beginning to end. With that I think it’s pretty clear by these results that at least in my case there has been no significant change in the conductive performance. If anything it got a bit warmer. But even there, between our highest and lowest temperature we have a difference of 2.5 c for F1 and 2.3c for cinebench. Nothing to worry about in my opinion because it looks like the paste stabilized and will not change much over the coming months. How long should you leave it on there before changing? That’s a different question for a different video. thats my take on the results, and I need to say that this experiment is inherently flawed because I don’t have a control. I have a single PC, It would be nice if I had the resources to set up a series of PCs to test this with but that just not practical at this moment. I’m using an AIO over an air cooler, I have a notoriously hot GPU that may be affecting the temps, Maybe my thermal paste application wasn’t the most ideal. Maybe there’s some background processes that came with a windows update that I’m not aware about that’s changing the power usage. However, at the end of the day, I have a fairly typical PC setup and I believe that these results are valid enough as a practical example of whether or not break in time is actually going to affect your temps in any meaningful way. At least in the case of Arctic silver 5. But as always I urge you to see what others have found with similar tests because it very well may be different and it very well may be different with different thermal pastes that’s just the way it is. So with that, this has been tech illiterate, My name is Nic, Thanks for watching. Unique new york Unique new york. Come on Mr. Teleprompter show me the words ♪ All temps as usual are captures with F (Brain Fart) I feel like this shirt is just a bit too large.\n"
