Innovation Cooling 7 Carat Diamond TIM Test results thread *NOW ACCEPTING RESULTS!!*

[3 of 7]

Originally Posted by rjbarker

Thanks 3 of 6 ...gotcha!

Your welcome and it's 3 of 7, rj
I think people here confuse me with 3 oh 6 who is someone else on the site.

[sswilson]

Originally Posted by 3 of 7

Your welcome and it's 3 of 7, rj
I think people here confuse me with 3 oh 6 who is someone else on the site.

Heh... I'm not sure he'd look as good in spandex as the real 3 of 7 does...... :)

[3 of 7]

Originally Posted by sswilson

Heh... I'm not sure he'd look as good in spandex as the real 3 of 7 does...... :)

Your thinking of my aunty 7 of 9

[sswilson]

Heh... yet more name confusion.... can you tell I was never a bit TNG fan? :)

[ICD7]

Latest Results added -Check it over and let me know if it needs corrections

[rjbarker]

Your welcome and it's 3 of 7, rj
I think people here confuse me with 3 oh 6 who is someone else on the site.

Yep your right..."3 of 7"....."3 oh 6"....dod you guys do this on purpose

I actually knew that but was wondering how long it would be before I screwed up..

Go canucks Go

[zlojack]

I gotta get mine done. I've been slacking, Nademon! I'll get to it this weekend.

[Nademon]

No worries zlojack...I know you're good for some results. Chart is looking very good Andrew!

[ICD7]

Originally Posted by 3 of 7

Lots of dual cores and quads on here, so I figured I'd try this on my old P4 workhorse.
It's a 3.0C Northwood with a Zalman 7000ALCU in an MSI 945GCM478 mobo with 2G of DDR2 and a 7600GT ( PCI-E)
This rig runs open case all the time so that skews what one would expect the temps to be.
Load temps were taken 15 minutes into a 20 minute run at 100% using OCCT 3.01 and are all non-corrected as reported by everest Ultimate.



Ambient temps were 23C for all the tests

AS5

Idle 25
Load 44
RPM 1560 constant


IC7 First run ( second install after botching the first one)

Idle 25
Load 46
RPM 1560 constant

IC7 Second run
Idle 25
Load 45
RPM 1560 constant

The first install I used about twice the amount of IC7 than I would have used of AS5 and I knew immediately that something was wrong when it idled at 31 and load was 49 so I tore back into it.
Here's the size of IC7 I used



Here's the pattern on the HS. Too low and too small.



The line down the middle is from the copper strip on the HS



Here's my second application of IC7 ( about 3 times the size of AS5 )

Observational note:

That's a very unique spread pattern with the paste. The stripe down the middle is obviously much thicker than what would be considered a normal spread.

Looking at the construction of the heat sink along with the axis of the mounting system it seems to me while a probably a high pressure mount, the sink is bowing in the middle with higher contact pressure on either side.

Application is great no problems there.

Andrew

[ICD7]

Originally Posted by sswilson

Here's what I've come up with WRT my impression of the IC7 peaking at max temp vrs MX2 staying fairly consistent.....

The graph at the top is the IC7 and the one on the bottom is the MX2. Don't get caught up with the actual posted temps, as these are coretemp temps, and I'm not even sure if these MX2 readings are the actual ones from the run that ended up being recorded... I only know from the time frame that they are MX2.

If you look at the temps on the top graph (IC7) you'll see that the lower "load" temp is fairly consistant, while the max "load" temps are temporary spikes rather than the norm.

Looking at the temps in the lower graph (MX2) you'll see that the trend is just the opposite. On this graph the higher "load" temp (max) is fairly consistant, while the temporary spikes tend towards the lower temps.

Based on this, I'm pretty sure that the IC7 "average" temps over the whole 1 hour time frame would be at least 1c better (probably more) than the max/min peak readings would indicate.

Does that make sense folks or am I looking at this the wrong way?

One possibility

Just a thought from more of a mechanical point of view. I believe the electronics are pretty solid but the problem may be one of resolution.

Where I see fluctuations in readings on a synthetic die with metered power supplies the fluctuation mostly occurs in the ambient air reading which affects the power source temp.

I usually place multiple thermocouples for ambient temps and average them and when I take a reading I stand a few feet away. Thermocouples are so sensitive that just standing near them a couple of feet away they will pick up your body heat and the warmed up air you breathe and cause temp fluctuations of as much as a couple of degrees C, even 5 feet away normal air currents will move the reading up and down 1/2 a degree.

This in turn affects the source reading - if the sink is warmer or colder the source heat will be warmer or colder. Whatever happens upstream affects the down stream heat flow and when the downstream changes it affects the upstream heat flow.

To me it looks like the electronics are picking up the ambient fluctuations indirectly and/or directly from downstream heat flow changes

Andrew