Ben's Bargains Serving Fresh Deals 24/7

A high quality silicone grease has a relatively narrow molecular weight distribution, thus there is no oil to "pump out". It should not degrade at temperatures on a CPU package. I agree that there are other polymers having somewhat higher thermal conductivity. Fine on the trade market jargon, I do not pretend to know or be able to sling that about.

Thanks to the other poster for the pointer to a better and bargain supplier. Grim chances for finding any Bargains on Ben's these days.

<table width="90%" cellspacing="1" cellpadding="3" border="0" align="center"><tr> <td><span class="genmed"><b>dave_c wrote:</td> </tr> <tr> <td class="quote"><table width="90%" cellspacing="1" cellpadding="3" border="0" align="center"><tr> <td><span class="genmed"><b>danpi wrote:</td> </tr> <tr> <td class="quote">Sorry #6, your chemistry is woefully lacking. Silicone vacuum grease is pure polymethylsiloxane, a pure material, or sometimes a polyaryl siloxane derivative. The thickness/viscosity depends on the degree of polymerization.

There can be no separation because there is nothing to separate and the polysiloxanes *are* in fact synthetic.

To get you started look at Wikipedia http://en.wikipedia.org/wiki/Silicone

Sorry Charlie, we're looking for good taste, not taste good.</td> </tr></table><span class="postbody">

Too bad you don't know WTF you're talking about. The separation is between the oil and the solids, not the silicone oil itself, and over years of use anyone who had dealt with both paste formulations knows this. A grease like AS5 using a polyol ester blend is superior to polysiloxane alone because thermal cycling does not pump out as much oil long term.

Synthetic is used as it was in the marke... [Truncated]

<table width="90%" cellspacing="1" cellpadding="3" border="0" align="center"><tr> <td><span class="genmed"><b>danpi wrote:</td> </tr> <tr> <td class="quote">A high quality silicone grease has a relatively narrow molecular weight distribution, thus there is no oil to "pump out". It should not degrade at temperatures on a CPU package. I agree that there are other polymers having somewhat higher thermal conductivity. Fine on the trade market jargon, I do not pretend to know or be able to sling that about.</td> </tr></table><span class="postbody">

Whether it is high or low quality silicone grease, ALL silicone greases pump out the silicone oil far sooner than AS5, Alumina, or Ceramique does.

It's not that the oil itself degrades, again you do not understand. The oil viscosity relative to temperature has a higher change and thus, after several thermal cyclings there are dryed out solids causing poor thermal conductivity.

I have personally pulled heatsinks off CPUs that had brown scorch marks on their pin-side because of this problem with silicone oil based greases.

Again, silicone oil based greases are not appropriate for long term use on high heat density parts. 100W per sq. CM is excessive for a silicone oil based grease. It WILL, ALWAYS, be a problem within the lifetime of a system with these parameters. There isn't any particularly high or low quality silicone oil that makes a difference, silicone oil was a near-enough perfect science 30 years ago if not longer.

It is not a matter of conductivity at all, it is a matter of what happens to conductivity when the liquid and solid portions separate, the solids then leave elevated islands of compound that are poor at heat conduction and even that only in areas that these solid deposits remain.

To put it simply, going beyond theory I have years of observation of this effect to go on. Once CP... [Truncated]

Dave, I am puzzled a bit because some things you say sound as if you do understand silicone chemistry and other things appear woefully contradictory.

A monodisperse silicon grease cannot separate because it is a pure compound and completely uniform. A specific example is Dow Corning "High Vacuum Grease Silicone Lubricant" sold in 2oz containers. It is stable from -40 to +400F and has a less than 1 micron vapor pressure over this range. No "oil" will ever come from this material unless you alter it chemically (for example by chemical attack or thermal degradation).

Party me for being politically incorrect, but you statement is, simply put, wrong.

As I explained, you are correct in that I do not know all of the slick and deceptive marketing terms used with highly marked up goo that is sold to relatively unsophiscated (unsophisticated as to materials, heat transfer, and chemistry) users and home computer builders.

As for experience, I am an expert in materials *and* electronics and have been doing this stuff for substantially more time than you mentioned.

<table width="90%" cellspacing="1" cellpadding="3" border="0" align="center"><tr> <td><span class="genmed"><b>dave_c wrote:</td> </tr> <tr> <td class="quote"><table width="90%" cellspacing="1" cellpadding="3" border="0" align="center"><tr> <td><span class="genmed"><b>danpi wrote:</td> </tr> <tr> <td class="quote">A high quality silicone grease has a relatively narrow molecular weight distribution, thus there is no oil to "pump out". It should not degrade at temperatures on a CPU package. I agree that there are other polymers having somewhat higher thermal conductivity. Fine on the trade market jargon, I do not pretend to know or be able to sling that about.</td> </tr></table><span class="postbody">

Whether it is high or low quality silicone grease, ALL silicone gre... [Truncated]