new TIM king???

Hey,

I recently got myself a sample of the ESGN Nanotherm EXP now know as Nanotherm PCM+. I was very hessitant to try this TIM because it was extremely watery. But when i finnaly tried it I was amazed.

here are my results, results recored after one week of use on a epox 8k9a2+, with a 1800 t'bred at 1.7v @ 1900Mhz. Volcano 7+ on low was the heat sink. All temps are load.

AS3 - 43
Nanotherm blue II - 44
Nanotherm Silver XTC - 40
Nanotherm PCM+ - 37

Thats a whopping 6 degree drop, from AS3. i re tested the AS3 and got the same temps.

I was very surprised at these results. Just thought you might be interested in this stuff. Of it is xtremely easy to apply as well.

where can i get that stuff from?!?! :slobber:

well there website is

http://www.esgn.com

they have a list of resellers on the main page i think.

Are you sure those temps weren't half creditted to the difference in ambient temps?

Ambients were the same, 22 degrees. Well thats according to the Air condinitioner

Well, I'm sorry, but I can't believe that you got a 6°C drop in CPU temp, just by switching your TIM. There must have been some other changes, perhaps in the way of apllication of the TIM or the way you mounted the HSF.;) :rolleyes:

maybe the stuffs just really good:D!

Good info ! I'll have to try it out.

Well ESGN is coming out with Innovative Products. I think this thermal compound is good for phase-change pcs. You can buy it here

http://www.3dcool.com/?module=product&sku=nanotherm

Hmm, 6c is a very big drop, for jsut a TIM change.

Still, worth a try all the same!

Edit: Typical, no UK stockists!!

i wonder how this compares to Shin-Etsu 751

Quote:

---

Originally posted by gobbo
Edit: Typical, no UK stockists!!

---

yahahahaha thats what i thought earlier

I also find it very hard to believe that any TIM could give you a 6C drop. I've seen reviews that claim about the same for Shin-Etsu 751 over AS3. Makes you wonder why anybody uses AS3 at all...:rolleyes:

I'd be very happy to be proved wrong of course. :D

I dont beleve a 6*c drop is possible.
In fact im 100% sure its impossible.

8k9a2+ uses the insocket thermistor to measure temps so your temps are totaly innacurate in any case.

The real question is can you overclock any further ?

It is possible if your application of it is not good.

AS3 is thickish remembering that you only need a micro thin
spread of it, if you over do it and do not have enough heatsink
pressure to squeeze out the excess you can end up with too
great a remaining layer of paste which can act as a barrier
preventing all heat to escape.

I believe this is why many see improvements with Artic Alumina
and products like nanotherm due to there more liquid texture
more is squeezed out in the initial application stage of the
heatsink.

I alway add a little extra pressure and try to twist my heatsinks
side to side slightly a few time to help budge the excess, seems
to work for me as I don't see too much of a difference in temps
from using one to another and temps are alway in gen good.

When ever I take a heatsink off there is basically nill compound
on the chip surface just micro traces from the fine scratches on
heat sink bases.

On my Prommy it looks like everything is squeezed out all that is
normal left is a picture of the text off the CPU die on the base
of the evap.

Quote:

---

Originally posted by PMM
I believe this is why many see improvements with Artic Alumina
and products like nanotherm due to there more liquid texture
more is squeezed out in the initial application stage of the
heatsink.

---

Very good point.

Quote:

---

Originally posted by Johnny Knoxvill
i wonder how this compares to Shin-Etsu 751

---

My thoughts exactly.

I'll take watery over the Shin Etsu putty any day.

Still, I think I'll wait until there's more of a consensus. One informal review is not quite enough to persuade me.

Not a fan of Shin Etsu at all, goes on like cement and its awful to spread, give me AS3 anyday!

I've been using Artic Silver Ceramique recently.... this stuff pwns. I'd have to say it's my favorite so far.

Hey,

I spose with testing all TIM there are a heap of variables which can alter the results.

Here are some results from other sites, no one else scored 6 degree difference, but it was clearly the winner in most cases.



http://www.8ballshardware.com/articl...ts_celcius.gif
Thats from 8 Ball hardware

http://www.techangel.co.uk/reviews/nanotherm/graph.jpg

From tech angel

and another graph here

http://www.pcreviewspot.com/reviews/...rm/results.gif

Looking good, but wonder how accurate your results are, in-socket termistor is not an good way to measure the temperature....

Anyway, i can't get a hold of it here in sweden. I've been useing both Nanotherm Blue II and AS3 before that though... I've would like to test something else...

I always thought that the insocket termistor were badly calibrated to the actually CPU temp, but not in reading the change.

eg it may read 45, but the CPU is actually on 55 idel. and then the load temps are 55, where the CPU is actually 65. The reading maybe wrong but the difference of 10 degrees isn't.

What i am trying to get at is the Delta T wrong as well.

socket thermistors are absolute crap for accuracy. they can be waaaaaaaaay off for several reasons. BIOS settings, etc can make a big difference in how they read and there is no point in trying to compare temps from one mobo to the next if all you base your temps on is the thermistor readings. 12 C differences have been fairly well documented.


that being said, I realize that is somewhat irrelevant as we are talking about temps read from the same mobo here. still, there are problems even with that...
the most obvious problems are with the socket thermistors that do not make any contact with the back of the CPU --all you are doing is measuring the air temp under the CPU with those.

any airflow aimed at the mobo that causes airflow around the socket will make a difference unless you seal the socket with conformal coating.

man I just bought 3 tubes of thermal compound within the last like.. couple months.. sheez. Oh well, maybe I'll test it out.

Quote:

---

Originally posted by Dexter
What i am trying to get at is the Delta T wrong as well.

---

It depends on the testing procedure, which is usually not explained in enough detail. E.g. I assume that the cpu is well cleaned between each test. Is it removed from the motherboard to do this? If yes then it is possible that the contact between the thermistor and cpu varies between tests...

If I was testing TIMs I would at least;

1. Get a mobo that reads the XP diode
2. Do the whole test more than once to check results (test all the compounds, then go back and start with the first one again).

Still, I wouldn't mind trying some of that Nanotherm stuff but like someone said earlier it doesn't seem to be available in the UK. :(

The Xp diode isnt accurate enough either.

plus as has been allready stated application is very important. It cant take a whyle for you to learn how to apply each paste, especially awquard ones like shin etsu.

CAN YOU OVERCLOCK ANY MORE ?

If you can then its doing some good, if you can then its not, simple and easy test :D

Quote:

---

Originally posted by MrLavender
It depends on the testing procedure, which is usually not explained in enough detail. E.g. ...Get a mobo that reads the XP diode:(

---

Would that be any motherboard having BIOS software (found here) capable oF reading the AMD XP diode using a "IC LM90" or other such device? And/or ABS II equipped mainboards?

Got some coming over for review in a coupel of weeks, so i'll post up results when i've tested the stuff!

Quote:

---

Originally posted by gobbo
Got some coming over for review in a coupel of weeks, so i'll post up results when i've tested the stuff!

---

where'd you get it from?

Quote:

---

Originally posted by Johnny Knoxvill
where'd you get it from?

---

Ditto. :D Will it be on sale in the UK? I'd like to test it for myself.

from 8balls hardware's graph, a 4C drop in temp over AS3 is pretty impressive just by changing the TIM. it's kinda like changing a new HSF, just cheaper :D

Quote:

---

Originally posted by aLpHaEv7
from 8balls hardware's graph, a 4C drop in temp over AS3 is pretty impressive just by changing the TIM. it's kinda like changing a new HSF, just cheaper :D

---

Exactly, that is why I am so suspicous of these results. If there was really that much difference then, as I said earlier, no-one would use AS3 or anything else!

I dont mean to be rude, but both those graphs are pointless.

8ballshardware are testing using a duron and insocket thermistor (they dont explain where the temps come from but im assuming that if they had propper test equip they would say) so there results are pointless.

Techangel use a P4 system, but dont say how the temps were measured. They also fail to state ambient temps. There results may be more credable but they dont provide the required info to validate them.

PCreviewpot also use P4 but they use case temp as there ambient "Ambient case temperature was around 32'C."
As they cant get an accurate measure of ambient there results are also highly dubious

Im not a big fan of roundups, people spend an awfull lot of time and energy working on them, but often fail to do a good test. if your measuring temperature to within 0.5*c you need to be pretty damed accurate with everything, and provide evience that your methodology is sound.

This is why I dont and wont write a roundup, it so complex to do it right that its not worth doing at all IMHO.

Quote:

---

Originally posted by Dexter
Hey,

I spose with testing all TIM there are a heap of variables which can alter the results.

Here are some results from other sites, no one else scored 6 degree difference, but it was clearly the winner in most cases.



http://www.8ballshardware.com/articl...ts_celcius.gif
Thats from 8 Ball hardware

http://www.techangel.co.uk/reviews/nanotherm/graph.jpg

From tech angel

and another graph here

http://www.pcreviewspot.com/reviews/...rm/results.gif

---

EDIT -- Made several edits as I got confused, I think its all correct now.

I got in touch with Nanotherm directly and requested a sample for review on a website i write for and they are sending one over. I'll e-mail the guy about UK stocksits, but i don't think there are any!

In my review i'll try to be as honest as possable with the reults, although it may be vergiing on a roundup, as i plan on covering about 4-5 different TIM's.

Holst,

Valid points. But while all the tests were off in one way or another they all do agree on one point, this new TIM is worth a look.

gobbo,

What other TIMs do you intend to include, and what site will you be posting your results at?

Yep, the fact they all agree is the only positive thing there.

if you want any help working out how to do an accurate test gobbo id be hapy to pick faults before you test anything.
To do it right will take some planning.

Id like to see you try and test max overclock, although thats a tough challenge as there are allot of variables.

now all we need is some Xtreme members feedback ;)

Quote:

---

Originally posted by gobbo
I got in touch with Nanotherm directly and requested a sample for review on a website i write for and they are sending one over. I'll e-mail the guy about UK stocksits, but i don't think there are any!

---

if there are any leftovers, you know who to pm :cool:

Holst,

Did you get my PM about this? :D

Anyone have any updated observations to post?

This site did a good comparison, both P4's and AMD's.

http://thecrucible.ca/images/tptemps_intel.jpg

http://thecrucible.ca/images/tptemps_amd.jpg

What got me was the TCB 0.891, the author said it was a fictitious name to hide a certin companies Alhpha product. He said it was dry, powdery, and felt like moth wings ??

Any updated results to post?

Quote:

---

Originally posted by gobbo
I got in touch with Nanotherm directly and requested a sample for review on a website i write for and they are sending one over. I'll e-mail the guy about UK stocksits, but i don't think there are any!

In my review i'll try to be as honest as possable with the reults, although it may be vergiing on a roundup, as i plan on covering about 4-5 different TIM's.

---

Nanotherm sent me a sample few weeks ago. Here's the results;

All tests were completed using an Intel P4 2.4 C and SLK900U, Vantec Tornado 92mm. Abit IC7-G motherboard, OCZ GolD 3700 4x256MB, PCPower&Cooling Silencer 400W

Results at 2.4GHz @ 1.525Vcore;
Artic Silver 3 = 39C
Naotherm Blue II = 46C
Nanotherm Silver XTC = 38C

Results at 3.6GHz @ 1.550Vcore;
Arctic Silver 3 = 44C
Nanotherm Blue II = 50C
Nanotherm Silver XTC = 42C

I didn't do a review because there were so few brands used. I also running way low on Artic Silver III. I'm waiting on Some OCZ, Shin Etsu, AS3, and Ceramique. Once I get those I'm going to do a pre-lapped Vs lapped review with my 2nd P4 2.4C.

I have the Nanotherm PCM, which is described as "Phase Change Material for air and liquid cooled electronic devices..." it's a liquid compound. I was aprehensive about this, because the bottle's instructions read, "Shake Well". I'm skeptical of a thin viscocity liquid for such an application, but am going to try it later tonite because i think it may be ideal for a Lapped heatsink, and IHS which I did a few days ago.

yea its very good stuff. Works very well for me. Some dont seem to like it and there have been these type of debates sence i can remember, but i always say use what works best for you.

Ive seen much better temps over when i used as3.

I'll probably get in trouble for posting these, but here are some pics of TCB 0.891

http://thecrucible.ca/images/tcb1.jpg
http://thecrucible.ca/images/tcb2.jpg
http://thecrucible.ca/images/tcb3.jpg
http://thecrucible.ca/images/tcb4.jpg

Silly Silly overclockers :eek:
All you need is cheap white paste nothing more:)
Its all in the application and the 2 surfaces being perfectly matched.
Focus on the 2 surfaces not the TIM
Just my 2 cents Ive tried almost everything and when you have 2 surfaces that are mated properly you could use white silicon or gold palladium comet dust it matters little to none
Hell for awhile i ran out of white and was using mercedes benz disc brake quiet stuff performed just as good as as3.

i notice the Nanotherm PCM+ outscored the Silver XTC, I now see was in error not using it.

Kenneth I see in your sig you've modded your VDIMM on your IC7-G. If I were able to di this I feel it would be an ever better platform then the 4PCA3+. Wopuld you please PM me the process, or link me to where you found the mod? Was it difficult?

IC7 vdimm mod is here
http://www.xtremesystems.org/forums/...threadid=12914

But if you want an even easier way, pm me with an e-mail addy, I have a pic but no way to host. But you can only jump to one voltage 3.15v, and the VTT stays locked at 1.37v which is off of the recommended 1/2.

Heh, i just went to get the link.

Yea, thats what i did. Didnt have any major problems.

Thanx guys. Hey where can I get that Gold Palladium Comet dust TIM? They sell it at the Mercedes dealership along with the Disc brake stuff. Or do I get the Comet Dust at the Palladium night club in NYC? I'll ad them to the round-up.

I must diagree with what your saying Spicey, of course it makes a differnce. Sure Lap the heatsinkl, and CPU, BUT why not then add the best thermal paste you can find as well and gain another 3C-4C?

Agreed.

Btw, im about to lap my p4, thanks for the guied liquid, looks like it should give me some good results.

Liquid3D,

If you have the PCM+ and you've read the reviews that it performs better than the top competitive TIMs, and you can figure that a "liquid" will most likely fill in the surface imperfections and more effectively gap the air out of the thermal junction (moreso than a powder, thick grease or "plumbers-putty" equivalent), then why wouldn't you test it and see how it performs?

I want to get some of this stuff, but every resaler want's over $7 for shipping, when the stuff is only $9 to begin with!. That is just stupid. I don't need anything else from any of the stores either to justify the shipping cost, and I am not letting them make profit off of my shipping. Funny cause I stuck in a $15 120mm just to see, and the shipping stayed the same. Guess I will stick to AS3 for now.

Quote:

---

Originally posted by Dexter
Hey,

I recently got myself a sample of the ESGN Nanotherm EXP now know as Nanotherm PCM+. I was very hessitant to try this TIM because it was extremely watery. But when i finnaly tried it I was amazed.

here are my results, results recored after one week of use on a epox 8k9a2+, with a 1800 t'bred at 1.7v @ 1900Mhz. Volcano 7+ on low was the heat sink. All temps are load.

AS3 - 43
Nanotherm blue II - 44
Nanotherm Silver XTC - 40
Nanotherm PCM+ - 37

Thats a whopping 6 degree drop, from AS3. i re tested the AS3 and got the same temps.

I was very surprised at these results. Just thought you might be interested in this stuff. Of it is xtremely easy to apply as well.

---

i get 3-5C drop from AS3 with PCM+ and my P4s but really you have to measure the rooms at the time to be accurate i.e. my day and night room temps vary from 16C night to 23-28C day

Quote:

---

Originally posted by LikwidKool
I want to get some of this stuff, but every resaler want's over $7 for shipping, when the stuff is only $9 to begin with!. That is just stupid. I don't need anything else from any of the stores either to justify the shipping cost, and I am not letting them make profit off of my shipping. Funny cause I stuck in a $15 120mm just to see, and the shipping stayed the same. Guess I will stick to AS3 for now.

---

Check your PM for more info ;)

Heh... that .80 or whatever stuff looks surprisingly familiar.......................

Liquiid + concave/convex P4 heatspreader = bad.

Liquid + flat AMD core/lapped P4 heatspreader = super good contact.

Damn you mean macci and me can get even better temps with all these Glorified metal tims.
The best contact surface is the one with the least amount of interference between the two opposing objects.
fill the pores in the metal surface with copper dust if you want the best heat transfer. If thats to out there for ya.
Just use the white stuff:)
Its al or nothing man!
Of course this is just my opinion;)

Spicy does have a valid point...that's part of the reason that the Nanotherm PCM works so well, it gives you the best contact between the die and HS. That seems to be it's biggest difference from the others. :D

alright I ordered some from ithinkpc.com they had it for $9.99 and free delivery. They also sell internationally for you guy's across the pond. I guess we will see what it does when it comes in.

Quote:

---

Originally posted by Dexter
Hey,

I spose with testing all TIM there are a heap of variables which can alter the results.

Here are some results from other sites, no one else scored 6 degree difference, but it was clearly the winner in most cases.



http://www.8ballshardware.com/articl...ts_celcius.gif
Thats from 8 Ball hardware

http://www.techangel.co.uk/reviews/nanotherm/graph.jpg

From tech angel

and another graph here

http://www.pcreviewspot.com/reviews/...rm/results.gif

---

Hmm... it beat AS3 by .5 degrees idle and a few degrees load, where arctic silver ceramique (Arctic Silvers high end product) was not tested... Ceramique usually beats AS3 by 3 or so degrees.

Quote:

---

Originally posted by LikwidKool
alright I ordered some from ithinkpc.com they had it for $9.99 and free delivery. They also sell internationally for you guy's across the pond. I guess we will see what it does when it comes in.

---

Thank you - we appreciate your support.

no prob Nano, just look so damn interesting I wanted to try it! Plus I like your involvment here;)

I finally had time to try the nanotherm PCM+ on a problem temp pc. With ambient room temps staying exactly the same, temps went down 3c right away as measured in MMB5. I will be watching it to see if it drops any more. I am very happy with that.:D

UPDATE: Several hours later, temp has not changed, still 3c lower, just by changing paste...I don't remember what used to be on it...it was gray and thick. That's a big deal...this pc runs hot as hell...going to change HS next.

Quote:

---

Originally posted by Nanotherm
Liquid3D,
If you have the PCM+ and you've read the reviews that it performs better than the top competitive TIMs, and you can figure that a "liquid" will most likely fill in the surface imperfections and more effectively gap the air out of the thermal junction (moreso than a powder, thick grease or "plumbers-putty" equivalent), then why wouldn't you test it and see how it performs?

---

Because I'm an Idiot Savant.
By the by I'd enterd the wrong values from my "round-UP" on Pg.2 of this thread. I found where I'd written them down. I was reading screen shot temps from the Epox 4PCA3+ (Doh). And during those measurments I was using Q320A287 2.4C. It doesn't run as cool as my L310A735 2.4C, and most likely this is the reason it's not as stable at 300FSB 3.6GHz. I'tmay also be the Epox NB-silicon.

Unfortunately I hadn't seen the reviews on the PCM+. However as soon as I stop typing I'm switching out the paste for the liquid. [B]I belive I mistakingly surmized paste would more effectively displace any air, and fill imperfections. I was concerned under the very heavy SLK900U, it's clamping mechanism, and with high temperaturs, a liquid might simply be forced out from under the heatsink, leaving a sporadic coverage. This is because as temp rises, viscosity increases, however; as pressure is applied viscosity is lowered. Given the cancellation effect, and the following quote;
"In liquids the intermolecular forces are stronger. No simple theory is available to predict the behaviour of liquid properties. In most cases, we find that dk/dT < 0. Water is an exception, with dk/dT > 0. As the liquid is nearly incompressible, pressure has very little effect on conductivity."
I now realize, my ignorance in this matter was the sole reason for my procrastination.

On a side-note. It's so dam difficult to accurately measure the true on-die temp's, given the large discrepencies among temperatue diodes. For example the temps on Pg.2 were acheived on the Abit IC7-G. My Epox 4PCA3+ happens to be reporting slightly higher (2C-6C+). The temps below using Winbond's DrHardware software, at 3.6GHz @ 1.575Vcore (this was prior to my realizing my 2.4C could attain the same speed at 1.525Vcore) are further proof of these discrepencies;

Gathering reasonably accurate temp measurements from a PC based test rig can be a daunting task. I've made a few posts about measuring Die Temp and Ambient Temp (at the intake of your HSF or right around your Waterblock) to derive Delta T (Temp Rise above Ambient) in order to eliminate the variability and direct effect that Ambient Temp has on Die Temp. I don't have the exact posts bookmarked, but I'm sure you're getting the gist of what I'm about to say.

It really doesn't matter how inaccurate your Die Temp sensor is because it will be equally as inaccurate on every test you do (hopefully). Just make sure you get a good, standalone temp sensor to measure Ambient Temp (not the Mobo mounted System sensor either). Place the probe at the intake of HSF, note your Ambient Temp and Die Temp at the same time and then subtract Ambient from Die Temp at each measurement point to derive Delta T. Then plot the Delta T data every "x" minutes for "y" period of time - and do the same for each compound you test.

This will provide you with a fairly accurate "relative" profile of temperature data (Delta T, once again, not Die Temp) and will also provide a solid means for comparing the performance of TIM vs. another. The Temp data will all be "relative" to your specific system and will include the inherent inaccuracies of the sensors employed. BUT, the results will be consistently inaccurate from one test to the next, therefore, the actual comparison data will be quite relevant - "relatively" speaking. ;)

http://www.zzz.com.ru/art174.html

Thanx xgman that's the type of article I just love to read. I've read breifly about Carbon, and Diamond type materials, and their potential for cooling. Yet due to the prohibitive costs, few manufacturers would attmept to make such materials. They'd be fighting an up-hill battle against habit. Poeple have been conditioned to pay low prices for thermal materials, and change is difficult, especially when it's an increase in price. Personally I'd be happy to pay the price if the performance was there. Would i pay $20 for a few grams of thermal paste? I've paid $80 for a 1.5oz bottle of Colonge. Hell I'd buy my Processor all the Colonge it wants, and give it my Fortex watch if it perform better for me.

I have to say Nanotherm, I for one am happy your here at Xtremesys. I hope you remain. Your wisdom is appreciated, and I don't see you bashing other brands, just mentioning yours. I tried the PCM+ and I'm very impressed. One critical aspect of the article link above, (and of course the most important), was contact between surfaces. I knew I was on the right track in investigating attributes such as viscosity, but didn't know enough to understand what comprises a better thermal transfer material. . Your PCM+ is unique, in that it seems to have ameliorated the contact issues which plague other materials. I have a new found respect for your R&D. Looks like your goal was to produce a low cost material, the properties of which eliminated the larger particles found in most paste's, which create, and even increase gaps between surfaces. I was slightly perplexed at whether the blue "solid" in the liquid base, should be made to cover th entire surface, or just the clear base itself? I used a few extra drops, and took more time ensuring the "blue" was at 100% coverage, tobe safe. I wonder if next time, I should simple spread till moistened, not concerning myself in spreading the "blue" as well? Irregardless, my temps have dropped significantly. And even if my temp monitor is inaccurate in reporting the actual on die, there's still accuracy in the consistency from one measurement to the next. See what I mean? Because I used the same software to verify temps prior to the change-out, it's the one constant among the variables.

Agreed, nanotherm makes good points and is a generous contributor to the forums. Thanks man!

But, what I want to know is, where can I get carbon black, ethyl cellulose and polyethylene glycol? :D

Shame it can't be bougth nowhere here in Portugal :( , still using ASII i've bougth some years ago, wondering what improvements I could get!

btw great thread :D

Why do want want these components? Do you plan on trying to simulate the work that Dr. Chung has done?

From my experience, the only thing that may be unique to the good Dr.'s formula that might make this TIM patentable (and maybe not) is the addition of some type of specially engineered carbon black into the formulation - and perhaps a special method for processing the material. Carbon black is nothing special - burn a piece of paper and you've got carbon black.

Carbon black is readily available on the market (and has been for years) and factors little or nothing in terms of the performance of the TIM. Sure, carbon black has some thermal conductivity to it - and it also has some electrical conductivity to it. If you want thermal conductivity and you're not concerned about electrical conductivity - then use silver powder, not carbon black - it will perform better. It will also be less "dirty."

Look guys, I've read the research paper - I've read the "Carbon Black" posts and threads on this and other forums - I've even met with Dr. Chung personally on a few occasions. I've held off until now about saying anything. Mainly because I find it amusing about what a big deal everyone is making about this "Carbon Black TIM."

What I'm going say now may startle you a bit, but what we're talking about here is not an earth-shattering, new TIM technology, despite what everyone seems to think. To her credit, Dr. Chung is a very smart lady and is an awesome Materials Engineering person, but she's got several more unique and innovative research projects going on than this.

Now, I've never seen this material, nor did I know anything specifically about it until just recently. What I can tell you is that you may be better off looking at some of the other new TIMs on the market than trying to recreate what she is doing with this TIM. First of all, you don't have the processing equipment that she has available to her and, secondly, why would you want to put a highly engineered, electrically conductive "soot" on your CPU? You need to ask yourself this - seriously.

Of course, this is just my own humble opinion based on the overall concept of the work - and I may be completely and entirely wrong in my assumptions. This is not intended in any way to slight this new invention or the wonderful work that Dr. Chung does for the venerable University of Buffalo, for industry and for the community at large. Her achievements are many and large - and I hope this new product challenges the Thermal Management Industry as we know it today and drives the technologies and performance benchmarks of tomorrow.

Carbon black is easy - soot from a kerosene lamp is easy enough to collect. Unless I read the paper incorrectly the carbon black used was simply that, carbon black - nothing was specially engineered. The others I'll probably search for, and see if they're available to the general public.

Why? For fun, that's why :) And, because I'm curious. The same curiosity pushed me into buying some PCM+ yesterday. Can't wait for it to get here :D

(Edit) Not worried about electrical conductivity, I used to use (and still re-try it, from time to time since I still have it) CCW conductive silver paste. At one time it was the bestest around and if applied carefully doesn't cause any problem.

You may find this thread interesting.

Good read, thanks!

You're right, Mr. Thompson, is was an interesting thread.

sjohnson,

I'm certainly not trying to deter you from trying it and satisfying your curiousity. By all means... have fun with it. :p:

Believe me, I've experimented with many different forms of carbon - and I've experimented with PEG, also. Let me just say that the carbon black idea - try it for a filler if you like - but it's not magic pixie dust by any means. As a matter of fact, carbon black would be on the bottom of my list of the forms of carbon that I would choose to use a thermally conductive filler. Take that for what it is - my own opinion.

If you're not worried about electrical conductivity, which most people are, then you might come up with some interesting combinations if you play around with other types of carbon fillers. Carbon is a funny animal, though, expecially when used as a filler material. You should look into different "sizing" agents that essentially serve as a surfactant to help the carbon be more efficiently and effectively introduced into fluid matrix you plan on using. Perhaps that is what the ethyl cellulose is used for, but I couldn't say for sure. The proper "sizing" of the carbon is a very important factor to consider and it will vary from from to form (fiber, particle, crystal, nanotube, sheet or what have you).

Hope this helps with your experiment ;)

If you are curious, i'm going to say give graphite a try... 8-)

You're getting a little warmer now...

Nanotherm, always appreciate your point of view and knowledge.

Reading the paper, it would seem that carbon black is not so much a filler as it is a synergistic agent that complements the PEG and ethyl cellulose. Much like I *assume* PCM+ uses different materials to reach its design goal(s).

If I understand the paper correctly, the carbon black isn't so much a filler or conductive agent as it is the carrier of the agents PEG and ethyl cellulose - plus the deformability of carbon black is presumed to fill in the microscopic "voids" while allowing metal-to-metal contact at the "peaks".

I work with an altered form of carbon black all the time at work - copier/printer toner. A shame that the plastic component of toner precludes using it for this "fun" I'll try to have.

The reason this intrigues me so is that I've been trying to come up with a low-tech way to use solder for the TIM for several years, without success.

ASII and III were great commercial successes but I've always found them to be little to no better than well-applied rat-shack generic paste. Plus, I "collect" new paste tech - hence the purchase of the PCM+. I even tried the automotive copper anti-sieze paste - why not? (I had terrible temps with that though, the copper wasn't fine enough is my guess).

I have access to the necessary measures to "formulate" the CB TIM, and to play around with various percentages of the materiels.

Best of luck and success to your company!

Very cool :cool: You obviously understand the nuances of carbon better than most. I've got a butt-load of all types that I mess around with here and there. I still believe it has some possiblities, but haven't had as good a luck with carbon as I've had with other types of fillers. One of these days I'll come across the right combination - but it probably won't be with carbon black, though. God bless Dr. Chung if she's made it work for her in some way shape or form.

You work with toner, huh? Nasty stuff that is - goes and gets everywhere. I live right down the road from Xerox and Kodak and know a thing or two about toner myself. The particles might be about the right ize, but, yeah, to bad about that insulative plastic component.

You might try Conoco or some specialty divisions of the major oil companies. Good luck and keep us posted on your experiments.

Woulnd't Diamond Dust be an even better phase-change/filler if mixed with the correct catalyst?

I clean up toner lol - Computer tech/printer tech/copier tech. Pays the bills, but the job is only (just barely) a step above Mickey D's in pay :( The price you pay to live in a small town, but I gladly traded the 6 figure salary and the corporate shivs for the life ;)

Theorectically speaking, yes - practically speaking, no. The problem with diamond particles is that they are crystalline and multifaceted in nature with a lot of surface area. The particle-to-particle contact is, therefore, severly limited and whatever small amount of heat that is transferred between particles (or even between surfaces) is not enough to make it even a mediocre choice as thermally conductive filler material. If you could make me a nice smooth, plate-like diamond particle, I'd be interested in buying some, though ;)

Quote:

---

Originally posted by sjohnson
I clean up toner lol - Computer tech/printer tech/copier tech. Pays the bills, but the job is only (just barely) a step above Mickey D's in pay :( The price you pay to live in a small town, but I gladly traded the 6 figure salary and the corporate shivs for the life ;)

---

Well, I have to commend you on that - and you'll live longer and be happier. :)

Just some warnings, from my initial research:

"Ethyl cellulose is an off white granular powder that is hazardous to humans so contact and inhalation should be avoided" - http://mason.gmu.edu/~jdiesel/ethylcellulose.html

-AND-

"Polyethylene glycol-electrolyte solution (PEG-ES) is used to cleanse the bowel before a gastrointestinal examination or surgery. It works by causing diarrhea." - http://www.nlm.nih.gov/medlineplus/d...r/a601097.html

Carbon black is, by itself, safer than mother's milk.

Let's be careful, K?

Nice stuff, eh? I'll eat the carbon...

Actually, the remark about carbon black is my personal opinion. California considers Carbon Black to be a carcinogen. In some scientific circles it's a *possible* carcinogen when respirated (breathed in).

Heck, I've seen studies that show that water, applied to the skin of rats caused cancer.

Ethyl cellulose, on some web pages is indicated as only causing irritation.

I just wanted to post some links that erred on the side of caution, hence my post about my opinion regarding carbon black.

Live and be well, don't worry, be happy :D :D :D

Nanotherm - what are the chances of beeing able to provide me with some samples (that will then provide to 2 local hardware sites and of course try it out myself ;) )?

Quote:

---

Originally posted by Liquid3D
Thanx xgman that's the type of article I just love to read. I've read breifly about Carbon, and Diamond type materials, and their potential for cooling. Yet due to the prohibitive costs, few manufacturers would attmept to make such materials. They'd be fighting an up-hill battle against habit. Poeple have been conditioned to pay low prices for thermal materials, and change is difficult, especially when it's an increase in price. Personally I'd be happy to pay the price if the performance was there. Would i pay $20 for a few grams of thermal paste? I've paid $80 for a 1.5oz bottle of Colonge. Hell I'd buy my Processor all the Colonge it wants, and give it my Fortex watch if it perform better for me.

---

I am going to be testing this substance very soon and from my understanding, the costs to manufacture are very low in fact. But someone has make a deal to supply it first.

Quote:

---

Originally posted by Rudzer
Nanotherm - what are the chances of beeing able to provide me with some samples (that will then provide to 2 local hardware sites and of course try it out myself ;) )?

---

That's a very distinct possibility. Please drop me a pm and we'll discuss it further ;)

Quote:

---

Originally posted by xgman
I am going to be testing this substance very soon and from my understanding, the costs to manufacture are very low in fact. But someone has make a deal to supply it first.

---

You're right, the manufacturing costs for a diamond-based product aren't as horribly prohibitive as one might think - but they are not "low" either. What is prohibitive is the performance. Try it for yourself and see (and please read my post about 7 posts ago).

The other problem is that diamond is ultimately Abrasive. No matter how carefully you try to apply it to your nice polished and/or lapped HSF or Waterblock, the surfaces are going to wind up being "roughed up" substantially no matter what you do, as will be the surface of your CPU die. If you would like to see picks of what small micron-sized diamond particles can do to a surface, please email me and I'll send some to you. It's not pretty.

Of course, I'm not saying that you should take my suggestions to heart or learn from my own experiences. Try it out if you are determined to do so and and satisfy your own curiousity. You might want to pick up a lapping kit beforehand, though, so you're ready to refinish your surfaces after you've discovered the truth for yourself and you want to go back to your old, favorite TIM ;)

Quote:

---

Originally posted by sjohnson
ASII and III were great commercial successes but I've always found them to be little to no better than well-applied rat-shack generic paste. Plus, I "collect" new paste tech - hence the purchase of the PCM+. I even tried the automotive copper anti-sieze paste - why not? (I had terrible temps with that though, the copper wasn't fine enough is my guess).

---

This supports the already obvious conclusion of the research that states "the performance of TIM's is mainly governed by the conformability and speradability and conformabitlity of the material rather than the thermal concuctivity."

Quote:

---

Originally posted by Nanotherm
You're right, the manufacture costs aren't terribly prohibitive - I won't call them low, though. What is prohibitive is the performance. Try it for yourself and see (and please read my post about 7 posts ago). The other problem is that diamond is Abrasive. No matter how carefully you try to apply it to your mice polished and/or lapped HSF or Waterblock is going to "rough it up" substantially, as well as the surface of your CPU. If you would like to see picks of what even small micron-sized diamond particles can do to a surface, please email me and I'll send some to you. It's not pretty.

I'm saying you take my suggestions and learn from my experiences, though. Try it out your and satisfy your own curiousity. ;)

---

Diamond? Of the substances tested, the one containing diamond was about 5th below carbon black.

That statement hits the nail right on the head - and is essentially what I have been saying to people for 2 years when they ask me why I don't spec thermal conductivity measurements for our thermal compounds. Why bother with numbers? The best way to determine if a thermal compound works for a particular application is to try it and see if you like it. Forget about the numbers - the numbers mean little or nothing when it comes to judging the true performance of a TIM. Try to tell that to an engineer, though... :eek:

Quote:

---

Originally posted by xgman
Diamond? Of the substances tested, the one containing diamond was about 5th below carbon black.

---

Pardon the misinterpretation then.

Nanotherm you got a new pm ;)

Here's a post from Nevin in a thread from December of 2001 on diamonds. The particle size he was working with ranged from 0.125 to 1 micron. ;)

Quote:

---

While the thermal conductivity of diamond is very high, there are other physical characteristics that make it a poor choice for a thermal compound. Microscopic diamond particles have spikes and sharp edges so the physical contact from one particle to the next is extremely small. The heat travels very well through a particle, but does not transfer well to the next particle. When you are talking about 50 to 100 particles to make up 1/1000 of an inch, this lack of particle-to-particle conduction is a major problem.

Also, diamond has a great affinity for diamond. Once the compound sits for a few hours, it forms clumps as the diamond particles bunch together. A very bad thing in a thermal compound.

And finally, the wands on high-speed homogenizers are about $1800 each. The bearings on the wand are submersed in the compound being mixed. The highly abrasive diamond particles ruin the bearings each time a batch is mixed. That's a loss of $1800 per 400cc of compound.

Given all this, I have mixed up several test batches of diamond compound.

They all suck.

Nevin

---

Well, that further substantiates and validates my previous posts about the viability of diamond as a thermally conductive filler material for TIMs. You've got the word from two good sources now. It's a safe bet that you won't be seeing a diamond TIM from either of us in the foreseeable future.

Quote:

---

Originally posted by sjohnson
...Heck, I've seen studies that show that water, applied to the skin of rats caused cancer....

---

This is another example of scientific methodolgy. The importance of controlled environments, and variables. Lets say the study used coolant water from a leaky Reactor. Well of course the the Rats got cancer. More likely, there could have been any number of contaminants (variables) unbeknownst to the scientists.

I can't tell you the amount of posts, replies, and irratic conclusions I've read in hardware forums, which are simply too open ended for retort, other then on the priciple of their structure. For exmaple, someone earlier, made the comment "All thermal pastes suck" (or simliar). In some respects this may be true. For example, they MAY all suck, compared to IDEAL theoretical solutions. Such as a process that would metamorphosize the multifacted "edges" of the particles, but leave their thermal conductivity in tact. It's also akin to saying Samsung TCB3 IC's suck, because their rated to DDR333. Well OCZ engineered TCB3 Rev-E IC's to perform at DDR510. Doesn't this empirical premise negate the universal claim "all samsung TCB3 IC's suck"?

BTW sjohnson we have much in common in respect to your lifestyle choice. There was time all I wanted was to enter the Advertising feild, and makes lot's o money. Then I fell in love with Philosophy, specifically Neruophilosophy. My father said,"chck the paper, you don't see any Want-ads for Philosophers do ya?" So I changed my major to Law, where i could get away with a BA in Philsophy, a great Pre-Law major. Growing up in Connecticut, ALL I wanted to do when i was young was live it up in Manhattan nightclubs. One day, I went skiing. A year later I eneded extreme skiing in the Wasatch Mountains, and majoring in Philosophy at teh University of Utah. Since then I lived in a number of small towns from Costa Rica, to Provincetown on Cape Cod. I wouldn't trade the natures's sprirituality, for the largest home on Fith Avenue. I love small towns, and those of us left, who still beleive in a community. I'm materially poor, and sprritually wealthy. I'm sure you can empathize with that centeredness?

Two things - the experiment was done with distilled water, onto the shaved necks of immobilized rats. One drop/second for (IIRC) 180 days. I believe that the irritation factor would be the most likely cause of the cancer. My point is that *anything, misused* can be hazardous and/or dangerous. Conversely, anything used properly and carefully will have little likelyhood of causing harm.

And, I certainly do empathize. Why is where I live so "good"? It's a well kept secret :D :D

Does anyone in Australia sell this Nanotherm PCM?


I'd like to try some out too.. :)



FT.