Where did you buy your CPU cleaning supplies?

burningrave101 said:
Whats to substantiate?
Click to expand...
Your assertions that 70% isopropyl alcohol is not pure enough, that even a fingerprint will lead to higher temperatures, that using 70% alcohol to clean can result in slightly higher temperatures, and that these "microscopic particles" affect heat transfer in any meaningful way.

If the technique is to use the alcohol to soften the material, dissolve it, and then wipe it off, why does evaporation rate matter? If evaporation rate matters, at what point does a higher temperature or lower relative humidity overcome the lower concentration of isopropyl alcohol?

What is this residue composed of, and why does it not completely wipe off? How much do you think it affects the themal resistance of the interface between the spreader and the heatsink, quantitatively?

Maybe you're right, and I'm not challening that. I'm just trying to figure out if you really are, and why.

burningrave101 said:
I did a google search a minute ago and found several threads on different forums where users said the same thing.
Click to expand...
That's interesting, but it doesn't prove anything. It's a fallacy to think that, because something is common practice, it must be correct.
 
burningrave101 said:
Why would you be? 70% isn't pure enough so it leaves behind more of a residue. That residue will slow down the heat transfer potential more then if you had used 91% or 99%. If it wasn't true then everyone would just use 70% and the CPU cleaning guides on the net would recommend the usual 70% instead of the high percentage.
Click to expand...
that 1% of water in your 99% junk leaves residue anyways, and your wiping grease off a hunk of aluminium or ceramic, this isnt surgury. an extra 3/100ths of a degree from a nearly invisible residue isnt a catastrophy.
 
mikeblas said:
Your assertions that 70% isopropyl alcohol is not pure enough, that even a fingerprint will lead to higher temperatures, that using 70% alcohol to clean can result in slightly higher temperatures, and that these "microscopic particles" affect heat transfer in any meaningful way.

If the technique is to use the alcohol to soften the material, dissolve it, and then wipe it off, why does evaporation rate matter? If evaporation rate matters, at what point does a higher temperature or lower relative humidity overcome the lower concentration of isopropyl alcohol?
Click to expand...

Well for starters, when working with electronic components you want something that evaporates as quickly as possible. Thats why you clean something like your motherboard with high percentage alcohol and not just tap water. Water can damage electronic equipment. Dried water leaves mineral deposits and other contaminants behind which can cause a short. People wouldn't say that 70% Isopropyl left a residue if it didn't. But this isn't a question of whether or not 70% will Isopropyl will work. 50% Isopropyl will work. Its a question of which works best and which drys the fastest and leaves the less residue and offers the best possibility for maximum heat transfer that is not going to impeeded by contaminants left behind in the microscopic crevices of the heatsink and CPU IHS. Because the 70% is less pure then 91% it may not remove all of the previous thermal paste ither and leave a thin film. You usually have to clean more with the 70% then you do the high percentage stuff.

And also the comment about finger prints.

In addition, oils from your fingers can adversely affect the performance by preventing the micronized silver fill from directly contacting the metal surface. (Finger prints can be as thick as 0.005")
Click to expand...

http://www.eprom.com/home/Heatsink/ArcticSilver/instructions.html
 
burningrave101 said:
Well for starters, when working with electronic components you want something that evaporates as quickly as possible.
Click to expand...
It depends on what you're doing. If you're cleaning contacts, for example, you actually want a residue--you want to leave behind a lubricant and a coating to protect the surface.

burningrave101 said:
People wouldn't say that 70% Isopropyl left a residue if it didn't.
Click to expand...
People say things that aren't true all the time. Those statements don't hold up against critical questioning.

I hope you'll have time to answer the questions I posed. I'd like to come to understand your reasoning.
 
mikeblas said:
I hope you'll have time to answer the questions I posed. I'd like to come to understand your reasoning.
Click to expand...

I've already answered them the best that i can. You'll have to search more on the subject yourself if you desire more info. I can tell you from my own experience that 70% does leave more of a residue then 91% and it takes longer to evaporate. If you'll google it you'll find others that are experienced that say the same thing about 70% vs high percentage Isopropyl. Its not going to hurt anything to use 70% but it just doesn't work as well as 91% and considering the cost of them both i dont see why anyone would ever use 70% unless that was just all they had on hand at the time.
 
burningrave101 said:
I've already answered them the best that i can.
Click to expand...
I don't have answers to these; perhaps you overlooked them:

  • If the technique is to use the alcohol to soften the material, dissolve it, and then wipe it off, why does evaporation rate matter?
  • If evaporation rate matters, at what point does a higher temperature or lower relative humidity overcome the lower concentration of isopropyl alcohol?
  • What is this residue composed of, and why does it not completely wipe off?
  • How much do you think it affects the themal resistance of the interface between the spreader and the heatsink, quantitatively?

You referenced a post pointing out that a fingerprint might be 5 thousandths of an inch thick. That's interesting, but it isn't useful information because the thickness of the print doesn't tell us much about how it affects cooling. How does that thickness compare to the layer of interface material? What happens to heat as it passes through the fingerprint? Does the fingerprint conduct heat well, or does it act as an insulator? What is the area of a fingerprint? How much does it really impact the thermal conductivity of the interface?

I'd figure that the fingerprint is made up of oils from the skin. How do these compounds change when heated? Do they get hard? Bake off? React with the thermal grease?

Without any measurements, descriptions, or comparisons behind your claim, any accurate description of this "residue", or an explantion of why evaporation rate is relevant, I'm not ready to believe that using 70% alcohol is in any way worse than using higher concentrations, nor is it any worse than using ArctiClean.
 
mikeblas said:
Without any measurements, descriptions, or comparisons behind your claim, any accurate description of this "residue", or an explantion of why evaporation rate is relevant, I'm not ready to believe that using 70% alcohol is in any way worse than using higher concentrations, nor is it any worse than using ArctiClean.
Click to expand...

Well you're more then welcome to believe that. Most of the information that i've found tends to agree with what i've stated and most hardware experts seem to agree as well that 91% evaporates faster and does a better job of not leaving a residue so maybe that doesn't mean much to you and maybe there isn't much of a difference between the two but for those of us that want the best, and not just "what will work", its really no question. I really can't say for sure what kind of temp difference there would be if any but what we do know is that 91% is most definitely more pure then 70% which has a greater proportion of water and so less impurities are left behind on the heatsink and IHS. On Arctic Silver's website they also talk about the ArctiClean and state that micro molecules are left behind from the previous thermal paste in the base in the heatsink unless you use something good enough to clean it well. I can also tell you that the Isopropyl in its purest form will work the best at getting rid of a thermal pad or something that is really stuck to the heatsink. The more water that is in the Isopropyl the harder it will be to clean it. There really isn't any cost difference between the two so there isn't an argument over price vs performance.

It may prove beneficial to your own knowledge and to the rest of us in the thread if you do some research on your own to add to the mix instead of only questioning what others have stated and demanding substantial proof when you've yet to provide anything yourself to way the argument in your favor. Have you ever even used high percentage Isopropyl and compared it to using 70%? What kind of results did YOU see?
 
Jewel - Alc. Swabs and Keenex - very easy to use and great at removing just about any kind of thermal compound (unless its the pink chewing gum, oldtimers should know this one :) )

D
 
Well, when I've been using it, I've been cleaning and buffing and drying and w/e several times. Haven't noticed anything yet <shrugs>
 
burningrave101 said:
It may prove beneficial to your own knowledge and to the rest of us in the thread if you do some research on your own to add to the mix instead of only questioning what others have stated and demanding substantial proof when you've yet to provide anything yourself to way the argument in your favor. Have you ever even used high percentage Isopropyl and compared it to using 70%? What kind of results did YOU see?
Click to expand...

They were your assertions, so it only makes sense that you explain them. After all, there must be a reason you agree with them, right?

Just the same, I did a few searches to find an answer and didn't find anything that quantitatively meausres the difference, explains what phenomena might be at work, or what this "residue" is about. There are plenty of posts and pages which parrot the same advice, apparently also without any insight into what's really happening.

There's no argument -- I'm just asking for background information. In general, I'm skeptical of unbased claims. I'm curious about the physical world around me, so I'm always intereste in how things work and how they can be measured scientifically. That's why I'm asking.

I've used 70% alcohol and a higher percentage -- though I can't remember what specific dilution. I've also installed plenty of heatsinks on parts without any cleaning whatsoever; on smaller parts (like voltage regulators and transistors in TO-220, TO-218, and TO-3 cases) which dissipate less heat than modern processors, I don't believe it would make any difference. But I've never done a side-by-side comparison.

I have done back to back comparisons using Thermalcote and Arctic Silver on a proessor. I found that Thermalcote worked slightly better, though I didn't write-up my findings. IIRC, the Thermalcote heatsink was consistently 2 or 3 degrees centigrade cooler.

I continue to use Thermalcote as a result, though. It's a fraction of a cost of the Arctic Silver products, and the experience has made me skeptical of the company's claims.
 
I just picked up a bottle of Isopropyl Alcohol (91%) from Target, for a $1.24 it should do the job well enough for me.
 
You must log in or register to reply here.
Back
Top