Thermal Conductive Tape testing

[VW_Factor]

Tonight, in anticipation of doing some heatsink modding, I realized that I was nearly out of thermal tape. Digging around in my stuff I found a roll of some double sided tape, however this roll of tape wasnt really thermal tape, it was just supposed to be permanent tape.

Got me thinking.

I went to the garage to test this stuff. Keep in mind this regular old double stick tape is pretty damn thin stuff. Thinner than most paper I think. Otherwise, I stuck a heatsink to a small slab of copper I have around with the tape, and head to the garage. (attached a smallish corner of the heatsink to a corner on the slab) I used a laser temp probe for reading the temperature of the heatsink that was attached to the copper slab. I heated the slab with a regular old cigarette lighter for 20 seconds, from the furthest point from the heatsink and watched the heatsink temp.

The temperature rose steady and quickly.

I repeated the test with the small bit of thermal tape I have left. It is Sekisui brand tape. Its about the thinnest stuff in the price range. I ordered another roll of it tonight anyway, however.. The test commences.

I was very surprised to see that the temperature difference was only 5 degrees, and there was no obvious glaring difference in how quickly the temperature was transferred either..

Should I look into this further? Is there another method I could use to test this stuff? I was only curious, so I thought I'mma check this out. Now Im left with more questions than answers.

For linking pleasure.. This is the double sided tape I have.. http://www.duckproducts.com/products...id=126&PID=333

Sekisui 5760
http://www.bestbyteinc.com/Cooling_for_CPU_,_VGA_,_HDD_,_Chip/Thermal_Compounds_&_Adhesives/Thermal_Tape,_Sekisui_5760,_25mm_(1in)_X_50_Meter_ Roll/Page_2/COTISK02.html

[OverclockN]

What was the temperature that it actually reached?

The real test is how hot it gets under extended periods of load. I'm interested in what you're doing here, keep us updated.

[VW_Factor]

What was the temperature that it actually reached?

The real test is how hot it gets under extended periods of load. I'm interested in what you're doing here, keep us updated.

The flame from the lighter in the 20 seconds reached a temp of 140f and 145f.

Not enough to "heatsoak" the metal here, but that wasnt my intention either. I was merely hoping to find out how well or not so well the regular tape conducted or allowed heat to pass. This initial test, showed that it didnt do too bad.

[VW_Factor]

More testing this morning.

I whipped out the old Dell. The machine is a basic Intel chipset with 2.8ghz HT P4 power. The northbridge gets quite hot, however because of the nature of this testing, I did my testing on the southbridge as it didnt require a heatsink, giving me the opportunity to get readings off the top of the chip itself, hoping to see if the tape was actually insulating heat instead of conducting it.

Onwards to some pics..

Here is my temp probe.



Here is a quick reading off the Geforce4MX Dell special at idle..



How I went about this testing was pretty simple. Firstly, I left the chip bare, and ran through 3DMark2001 (yeah yeah, its old) to put some heat into the chipset. The bare chip alone reached a temp of 103f in the center. I will add, that after taking a bunch of pictures, I realized it was pretty boring, as they all looked the same. Needless to say. Results were had.

First idle with the test heatsink and the "duck"



Ran through a couple runs of 3DMark and at the end of the second test (immediately after) I got a reading of 92f on the heatsink itself. Now, this could show two things. Either its doing its job, and exchanging the heat properly, or its insulted under the heatsink and in the chip.



So, to find out the exact story, right after the reading on the heatsink, it was quickly removed, and then tested on the chip directly under where the heatsink was placed. I got a reading of 94f



After doing this testing with "the duck" double sided tape, I repeated the same method with the Sekusui tape. There would be more pictures here, except really, they show the same thing, (with the exception of 93f on the heatsink after two runs, whoopie). Not very much excitment there.

What does this all mean? Is this regular old double sided tape something special? Or is my actual thermal tape something that isnt so special? Is the testing not showing the media enough heat? Which other testing methods can I try to "break" this stuff. Shouldnt there really be a bigger difference in the two?

[OverclockN]

I would definitely think there is a difference between the two. I'm kind of confused by the results I see here. To me, it would say that the "thermal tape" you have is just garbage. Without even further testing, it's hard to see if that's actually the case.

Do you have any thermal paste on hand? Any Arctic Silver or something similar?

[VW_Factor]

I would definitely think there is a difference between the two. I'm kind of confused by the results I see here. To me, it would say that the "thermal tape" you have is just garbage. Without even further testing, it's hard to see if that's actually the case.

Do you have any thermal paste on hand? Any Arctic Silver or something similar?

Got lots of pastes and such around..

Think I should attach the sink with AC and retest?

Even still. The standard old double stick tape, shows that its not insulating much heat into the chip after removal of the sink, and getting a heat reading. Should I be testing hotter chips/equipment?

[OverclockN]

Think I should attach the sink with AC and retest?

Yeah, I'm curious to see what the difference would be.

Should I be testing hotter chips/equipment?

With newer hardware that runs at higher temps, you may see a bigger difference between the two. Not sure I would bother with trying it on newer stuff, but I would definitely see what kind of difference there is with the thermal paste.

[VW_Factor]

Going to attempt to get at this more today.

Is OC'in the only person interested in this sort of stuff?

[OverclockN]

I've messed around a bit trying to get the best cooling with what I had in the day. Wet sanded a heatsink I had to lap it, and it worked FANTASTIC. Also did the old graphite pen trick to unlock an old Athlon CPU to crank up the FSB. Slapped on a bunch of different VGA coolers for different video cards I've owned (Ti4600's, Ti4200, X800Pro, etc..)

Now, I'm not so ambitious with the projects. I'd like to try watercooling, but don't want to pay the price of it. I just ended up the buying what I thought was the best case I could get and loaded it up with 120mm fans. It's not quiet...but my AthlonX2 with a 700MHz overclock is at 28C most of the time in idle and about 46C under load.

Anyway, yeah...I rather enjoy this stuff.

[VW_Factor]

Well.. Final tally thus far.

Given that the double stick tape works somewhat.. I whipped out my old 9800Pro 128M. It long ago developed an artifacting problem. After playing with it some, I had found that the RAM was overheating, even at desktop. Not sure why, however perfect for more extensive testing right?

Using the same Dell computer, I used double stick tape and mounted these heatsinks to the RAM chips.



It worked.. Somewhat. About 10 minutes go by and it began with the artifacts. (usually, it was nearly immediate after getting to desktop) I reopened the case of the Dell (clamshell) and grabbed the trusty boxfan. Setting the boxfan to low, was enough and the artifacts went away.

Next was to test, with heatsinks that were half as tall. (gold in color, I think Uber has seen these before) Using the doublestick tape, I ran into the same thing, however the boxfan required to be on high.

Now, onto the real shat.



My new roll of 50 entire meters of Sekisui thermal tape. I should have thermal tape for a lifetime. Its 1inch wide and of course, a large roll.

I installed the tall heatsinks (seen above) with this tape. Guess what? No extra airflow needed, even with the case open, and sitting in the stagnant (non moving) air, didnt need the box fan. I proceeded with installing the smaller heatsinks, and with the case closed the case airflow provided enough movement to keep artifacts away, even while running 3DMark.

Conclusion. It works. As advertised, and regular old doublestick tape while enough for smaller, cooler things and it does help.. It cannot match the performance of the real deal.

Some temps were taken during this testing, however turned out to be irrelevant to this style of testing.

Now, with a small modification (addition of the slow LED 80mm fan, running on 5v) this revived card has found its way into my shuttle media box.







The sidepanel fan, will eventually find itself cleaned up and smoothed into the side. Likely with some fiberglass built enclosure and some bondo. Make an art project out of it. The 120mm fan on the rear with adapter was of course for silence (the 80mm in the case would spin up all the time and make too much noise for a media box).

I hope someone enjoys this.

That, and now looking at the pictures of the back fan, I should get around to cutting the "grill" away from the heatsink radiator in the rear of the machine. Gain more airflow area.

[Uber_Tiny]

Impressed!

Oh and yeah I know what gold heat sinks you speak of.

[OverclockN]

Very nice work!

I figured at some point we would start seeing the difference with the Thermal Tape, just wasn't sure when that was. I was pretty sure that using stuff that conducted much more heat...we would notice more of a difference right away.

Thanks for sticking with it man!