The truth about AS-5 settling in

[I.M.O.G.]

Originally noted on http://www.arcticsilver.com/as5.htm#s5rev

Important Reminder:
Due to the unique shape and sizes of the particles in Arctic Silver 5's conductive matrix, it will take a 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 period will be longer in a system without a fan on the heatsink or with a low speed fan on the heatsink.) On systems measuring actual internal core temperatures via the CPU's internal diode, 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.

I noticed people are worrying about thermal cycling and vibrations and all different sorts of things in order to get their paste to settle... All it takes is normal use while occasionally turning the computer off. No need for the rituals.

 

[Korndog]

hm.. don't remember reading that before, is that an updated version from AS??
btw, thanks for sharing that with us IMOG
this would make a nice cooling sticky

 

[archilochus]

Dog, I tell ppl that all the time..200 hours of thermal cycling.Thanks for bringing it to lite, IMOG.THE FANMAN

LINKAGE

 

[Sk8erfreak986]

Getting stuff like that in its own thread is very useful cause most people don't read the AS site to in-depth and it would be hard to pick something like that out of a thread on something else...Thanks

 

[k3nshin]

What rituals?

Doing the as5 settiling dance? The Thermal cycle shuffle?

 

[I.M.O.G.]

What rituals?

Doing the as5 settiling dance? The Thermal cycle shuffle?

Well, attaching a vibrator to the waterblock has been suggested though I don't personally recommend it... I know some people have ran several cycles of Prime or other loading program and raised their vcores "to help the thermal cycling" also.

Haven't seen any reports of people dancing naked under the moon yet, but we probably just haven't heard about it because they are still in jail.

As far as I know that excerpt I posted is not an amendment to the AS site, but that doesn't mean it isn't. The AS guys really have their fingers on the pulse of the community so it might have been a reaction to recent questions and such.

 

[Joe Camel]

i just set up a pelt / water system and was wondering if this info holds true for the cold side of things. everything above the pelt will be hot or room temp, but below the pelt is room or LOWER. (CPU idle 8C, load 22C) i used Ceramique, but it has a set in time just like AS5.

 

[mccoyn]

i just set up a pelt / water system and was wondering if this info holds true for the cold side of things. everything above the pelt will be hot or room temp, but below the pelt is room or LOWER. (CPU idle 8C, load 22C) i used Ceramique, but it has a set in time just like AS5.

My guess is you still have to do the thermal cyclying to get the best effect. You probally can do that by leaving the pelt off and do thermal cyclying without overclocking.

 

[I.M.O.G.]

The pelt goes between the waterblock and the CPU (CPU>ceramique>coldplate>pelt>waterblock) - so he can't leave it off or he will fry his chip. A peltier is a good insulator when it is not powered and actively pumping heat.

I would guess that your settling in time would be slightly longer than the average stated time by AS. At idle you are probably roughly 15C below ambient, and most users are probably typically about 15C or so above ambient so the absolute temperature difference is comparable, but I would guess the cold would slow the settling process.

I PM'ed Colin to see if he has a comment on this. He will know better what to tell you.

It's nice to see another NE Ohioan around here Joe.

 

[Silversinksam]

Twisting or gently Wiggling the heatsink helps break in the compound. This is especially helpful to people with waterblocks, being that the processors die gets little disturbance when using a waterblock. A fan on a heatsink helps the paste settle, so no fan = recommended gentle jiggle

 

[Ritalin]

good to know

 

[Big_Baller]

Has anyone witnessed a 2c-5c drop in temps after thermal cycling is done? Mabye its time to lap my water block and add some as5 now.

 

[Colin]

Thanks for the head's up IMOG.

Joe - I would do just what IMOG recommends at the beginning of this thread. Just use your computer as you normally would. As your temps drop with break in, you may want to tweek your overclock.

BTW, Ceramique works really well at low temps, it’s rated to -180C.

 

[Sniper.nkc]

I highly suggest not "wiggling" your AS set-up during the cure time. You may break the paste elastic; hence, you will have a discontinuity on thermal disapation between CPU core and hs.

The water block cooling set-up will take longer for AS to cure. The 200 hour cure is average time with average temps. Water block cooling is better than average, thus, longer time to cure.

 

[Korndog]

Dog, I tell ppl that all the time..200 hours of thermal cycling.Thanks for bringing it to lite, IMOG.THE FANMAN

LINKAGE

i know, the red highlighted part is new to me though..

 

[mfjonny]

I highly suggest not "wiggling" your AS set-up during the cure time. You may break the paste elastic; hence, you will have a discontinuity on thermal disapation between CPU core and hs.

I'm sorry, i'm a little confused... What do you mean by "paste elastic"? Are you talking about the particles? The elasticity of the microscopic particles?

I really doubt that wiggling would break any particles designed to fit within microscopic valleys in the contact point. Please explain....

 

[Silversinksam]

I highly suggest not "wiggling" your AS set-up during the cure time. You may break the paste elastic; hence, you will have a discontinuity on thermal disapation between CPU core and hs.

The water block cooling set-up will take longer for AS to cure. The 200 hour cure is average time with average temps. Water block cooling is better than average, thus, longer time to cure.

I'd like to see you back up that statement with proof, because the 'Jiggling' method is an age old trick. Until you back up your comment with proof I will Jiggle, and most 'old school' Overclockers know, and use this recommended method. BTW I am recommending jiggling during the first few hours, and not after 200 or so hours.

As far as your comments about why Waterblocks take longer, you are only half correct, its only partly true due to the cooler temps, but it has to do with the block having no 'disturbance' as I mentioned earlier. Its also important to allow Thermal cycles rather than leaving your pc on 24/7 initially.

I'm sorry, i'm a little confused... What do you mean by "paste elastic"? Are you talking about the particles? The elasticity of the microscopic particles?

'Paste elastic' is generally a term in regards to Thermal stresses, Its a term commonly used in the Concrete industry. Unfortunatly his mentioning not wiggling the block is ridiculous advise. I'd like to see him prove his statement, but he won't, because its simply not true.

 

[Sniper.nkc]

I deal with many type of "sealants" in all types of applications. I have experience and I back-up what I say. I have thrown technical terms here you are not familiar with and let me explain so that some of you can understand.

There are serveral stages a sealant goes through to acheive performance/function. First stage is mix (self explanitory). Second stage is application time. This is the time to apply the sealant. Third stage is cure. There is a fourth stage which is hardness, but we are not concerned with that. The critical stage of all is cure. You don't want to disturb the cure of sealant or you will inherit problems with it. One problem I mentioned was discontinuity. This is a break or void in the material. The elastic is sometime like rubber, how much flexiblity and durability the material has. The AS uses the elastic for optimum design, especially when adding a vibration source (fan) on top of it.

What I said on my previous post was meant as a precaution. If you wiggle it during cure, you will inherit a risk of a tear with sealant, which by definition is a discontinuity. This will be unacceptable if you wanting to maximize heat disapation via heat transfer with AS.

 

[DayUSeX]

ah thnx for the true nfo, im tired of people telling all this BS on how to get AS to work the best.

 

[Silversinksam]

I deal with many type of "sealants" in all types of applications. I have experience and I back-up what I say. I have thrown technical terms here you are not familiar with and let me explain so that some of you can understand.

There are serveral stages a sealant goes through to acheive performance/function. First stage is mix (self explanitory). Second stage is application time. This is the time to apply the sealant. Third stage is cure. There is a fourth stage which is hardness, but we are not concerned with that. The critical stage of all is cure. You don't want to disturb the cure of sealant or you will inherit problems with it. One problem I mentioned was discontinuity. This is a break or void in the material. The elastic is sometime like rubber, how much flexiblity and durability the material has. The AS uses the elastic for optimum design, especially when adding a vibration source (fan) on top of it.

What I said on my previous post was meant as a precaution. If you wiggle it during cure, you will inherit a risk of a tear with sealant, which by definition is a discontinuity. This will be unacceptable if you wanting to maximize heat disapation via heat transfer with AS.

I'm familiar with the terms, as I was in the aviation field before I retired. I still am waiting on you to prove what you stated. Your precautionary statement said, you dont recommend jiggling it during the initial curing. I'd like you to Prove it with data not conjecture and 'guess-timation'

BTW Arctic Silver pastes are GREASES and NOT SEALANTS AND THEY DO NOT CURE, THEY THICKEN