Requested proof of Thermalright quality control issues

[Spawne32]

Personally i say give the wedge of cheese a go, nothing cuts better then vermont sharp chedder thats for sure

 

[nd4spdbh2]

Yes, it will be better because it puts out more heat, but it also requires I re-run all of my previous tests because the absolute temps and deltas will not be comparable between the two.

As for the thermal resistance stuff, I think the electricity analogy explained it fairly well. Fact is, if the die / IHS interface is bad you'll have high temps, but if one cooler works better than another, it should still work better and the temps will reflect that. This should be fairly clear. If the U120X is supposed to beat the Maxorb, it should beat the Maxorb regardless of how much power is dissipated through it, provided that the amount of power dissipated doesn't change between tests. The die / IHS interface isn't going to be bad for the U120X, then magically get better for the Maxorb. So it doesn't explain the Maxorb / U120X comparison w/ the 1st and 2nd heatsink. The 3rd one, because it performs somewhat better, is more up for debate.

how hard is it to put the ultra 120x, put it on the pentium d... run a stress test for like 5 min to see if the ultra 120 works??? then take the ultra 120 off and pop on the max orb... its like your trying all you can to NOT test this right, so you can just say to not buy thermalright products.... thats just what im seeing... dunno if its just me tho.

 

[maxfly]

what will testing another cpu prove?that his 4400 runs hot?am i missing something here?

 

[nd4spdbh2]

what will testing another cpu prove?that his 4400 runs hot?am i missing something here?

it will prove that the ultra120's rnt to blame (even though he presists on not listening to any logic)...

 

[Drinkyoghurt]

hmm i will test my 120 ultra extreme tomorrow, i installed it, i can twist it a bit...and it isn't rock solid(the screws fastened too early IMHO)

will post up tomorrow

 

[Metallica]

what will testing another cpu prove?that his 4400 runs hot?am i missing something here?

If he test is on the PD with the two sinks, and gets the same results (they perform the same) we will have something to go by.

I don't want to get into this arguing, soo .. just explaining why it would be a good idea

 

[maxfly]

i understand what is being asked.what i dont understand is how there is supposed to be a difference between testing a pd or his c2d?how will testing a pd prove anything about the hsf hes tested?
the testbed hasnt changed according to the op.so what is left to debate?
me tinks you guys should just agree to disagree on this one.
im not trying to start a flame war here but it seems like there has been enough proof presented for everyone to just live and let live lol.

 

[Spawne32]

i understand what is being asked.what i dont understand is how there is supposed to be a difference between testing a pd or his c2d?how will testing a pd prove anything about the hsf hes tested?
the testbed hasnt changed according to the op.so what is left to debate?
me tinks you guys should just agree to disagree on this one.
im not trying to start a flame war here but it seems like there has been enough proof presented for everyone to just live and let live lol.

The differences on the two processors really wont mean much, however what your really looking for when comparing two heatsinks is the deltaT which is found through your Load Average - ambient temp = dT. That would measure the exact difference between the two heatsinks and how well they perform. I tested the Ultra 120 extreme against a variety of heatsinks, and heres how the numbers stacked up.

the lower the deltaT the better the heatsink performed, as shown by this chart.

it basically removes the variations created in ambient temperature fluctuation's. As you can see by the charts a few of the heatsinks changed ranking positions on the deltaT chart because of the equation. So in order to take the two chips into consideration you would have to have a set of test results for the E4300 and then a set of test results on the Pentium D, however the pentium D's ability to record temps is far less accurate then the E4300, as you can get exact core temperature readings on core 2 duos.

 

[nd4spdbh2]

The differences on the two processors really wont mean much, however what your really looking for when comparing two heatsinks is the deltaT which is found through your Load Average - ambient temp = dT. That would measure the exact difference between the two heatsinks and how well they perform. I tested the Ultra 120 extreme against a variety of heatsinks, and heres how the numbers stacked up.

the lower the deltaT the better the heatsink performed, as shown by this chart.

it basically removes the variations created in ambient temperature fluctuation's. As you can see by the charts a few of the heatsinks changed ranking positions on the deltaT chart because of the equation. So in order to take the two chips into consideration you would have to have a set of test results for the E4300 and then a set of test results on the Pentium D, however the pentium D's ability to record temps is far less accurate then the E4300, as you can get exact core temperature readings on core 2 duos.

its not really even that... its just to get the heatsink on a different chip as its looking like the c2d chip notlag currently has, has bad contact between the core and IHS, causing bad temps overall... or GOD FORBID, hes doing something wrong... but notlag is perfect so that couldnt be... *cough*thermaltake rep*cough*

 

[Spawne32]

its not really even that... its just to get the heatsink on a different chip as its looking like the c2d chip notlag currently has, has bad contact between the core and IHS, causing bad temps overall... or GOD FORBID, hes doing something wrong... but notlag is perfect so that couldnt be... *cough*thermaltake rep*cough*

Well actually i just received a maxorb a few weeks ago from Thermaltake again for testing as well, so along with the other 5 heatsinks ive been testing this week for review, im gona test the maxorb and ill get an exact comparison in deltaT on my own equipment to see how well the performance of the maxorb is compared to a ultra 120 extreme. Whether or not notlags 120's have been damaged or not, i still cant decide.

 

[Drinkyoghurt]

Yoghurt reporting in

i have found a gfx card to conduct my tests with the TRUE 120
this is with a E6420 @ 3.2ghz
load temp generated by TAT so its higher then when using orthos
Results:
Idle: core1:24c , core2:24c
Load: core1:36 , core2:41c
i'd say ambient is around 20c..but i have no way of measuring this

its holding up well enough for me
btw, this is an without pressure mounting(so no coin or other mod), just used everything supplied with it(except thermal paste, i used AS5)

so I am guessing it's just Notlag's chip that is running hot..

EDIT: I am running a Enermax Warp fan on the HSF at full speed, which produces around 83cfm according to the site,

link http://www.enermax.com.tw/english/product_peripherals_detail.asp?PrID=43

and the proof

 

[Notlag]

how hard is it to put the ultra 120x, put it on the pentium d... run a stress test for like 5 min to see if the ultra 120 works??? then take the ultra 120 off and pop on the max orb... its like your trying all you can to NOT test this right, so you can just say to not buy thermalright products.... thats just what im seeing... dunno if its just me tho.

Trying all I can to NOT test it right, huh? I'm sorry, but there's no fault in the OP:

1. Try one heat sink
2. Swap it w/ another w/o changing anything else
3. Test the other heat sink
4. Then compare the deltas

I've also tried some of the suggestions here such as rotating the heatsink (nevermind that the whole reason I got the Maxorb was because of a suggestion here from the original thread where I was asked to try a comparable heatsink at OC'd speeds).

What I'm "seeing" is that certain people don't like accepting the idea that Thermalright's QC is crap lately, that they have been turning out some mediocre parts, and those same people are now grasping at straws trying to explain it away.

Consider that the 3rd heatsink was 4C better right out of the box compared to the first two. Consider the 1st two were barely capable of beating a Maxorb. Consider that other people have posted w/ problems w/ their bases and some bad performing heatsinks. Nothing with my rig or CPU (that I haven't already eliminated) can explain those three findings. Nothing.

 

[Notlag]

... or GOD FORBID, hes doing something wrong... but notlag is perfect so that couldnt be...

Consider my other posts and the detail I've put into this thread. I'm not doing something "wrong" mounting the U120X, applying TIM, providing air flow, etc...

*cough*thermaltake rep*cough*

LOL. Ok, ok... For the record, I don't like ThermalTake products, with the possible exception of their higher end cases (and for cases I prefer Antec). Most of their cooling products (like fans) perform nothing like how they're spec'd (lower output, higher sound pressure level, etc...). I'm not recommending people buy Maxorb's Or Big Typhoons for that matter. I wanted to compare a TR heatsink to something from somebody else and my local computer shop (I didn't feel like waiting on shipping or paying someplace for next day) only carries Cooler Master, Thermaltake, and Zero Therm. I needed something I could modify to use the TR bracket, so that cut the options down some.

 

[Notlag]

so I am guessing it's just Notlag's chip that is running hot..

Folks, this doesn't explain the Maxorb and first two U120X's tying, or the third one performing better by a 4C margin.

 

[WonderingSoul]

Folks, this doesn't explain the Maxorb and first two U120X's tying, or the third one performing better by a 4C margin.

Actually it does. As someone else stated; the temps of your IHS are limiting the cooling effectiveness of your heatsinks.

 

[Notlag]

Actually it does. As someone else stated; the temps of your IHS are limiting the cooling effectiveness of your heatsinks.

Read my post somewhere in the last page or two explaining heat transfer. Thermal resistances add in series just like electrical resistors add in series. If at 85W power the TR U120X is 9C cooler than the Maxorb on a working CPU, then even if the CPU die / IHS transfer is bad and results in temps that are 10C higher overall, the TR U120X will still be 9C cooler than the Maxorb.

The only exception comes in the fact that there are other paths heat can take out of the CPU, like going through the package. If the transfer through the IHS and heatsink is inefficient, then, like electricity, it will use other paths as well. However, the thermal resistance of the package and the transfer of heat from it into the surrounding air is extremely high, so even if the die / IHS interface is bad, the vast majority of the heat will still go through the IHS.

Consider an electrical circuit with two resistors in parallel connected to one in series:

<PRE>
     |
  -------
  |     |
  R1   R2
  |     |
  |     |
  R3    |
  |     |
  -------
     |
R1 = heat transfer from die to IHS (low resistance)
R2 = other paths for heat to leave the CPU (high resistance)
R3 = heat transfer from IHS to air through heatsink (low resistance)

R(Total) = 1 / ( (1/(R1 + R3)) + (1/R2) )
</PRE>

If R2 is high (like 1 KOhm) and R1 and R3 are low (like 10 Ohm), then even if R1 is double or tripled, the overall effect of R2 is still very small - virtually everything still goes through R1 and R3. Say the resistance of R3 for the U120X is 10 Ohms less than the Maxorb, then the overall resistance of the circuit is APPROXIMATELY 10 Ohms less w/ the U120X than the Maxorb (because R2 is many times higher it has little impact on the circuit and you can approximate it as a simple series circuit of R1 and R3). If we raise the resistance of R1 (even doubling or tripling it), the overall resistance of the circuit for the U120X is still approximately 10 Ohms lower than the Maxorb.

Heat transfers similar to electricity. This analogy holds. If the CPU runs hot then the CPU runs hot, but it doesn't change whether or not the TR U120X should be cooler than the Maxorb.

EDIT: Original version mistakenly had R3 in series w/ the combination of R1 and R2 in parallel (versus R1 and R3 in series, parallel to R2). Moved R3 to correct the diagram and changed the formula to reflect that. At the time of the edit (which followed immediately after the original posting when I proofed the post), nobody had yet replied to the original version, but please refer to the correct diagram shown above if there are any later discrepancies.

 

[CGR]

Wow, ten pages of absolute nonsense.

Why is this thread still going?

 

[Jon]

Wow, ten pages of absolute nonsense.

Why is this thread still going?

Why are you still posting in it?

I've been through every page of this thread and have seen that since your first post, you have contributed nothing but absolute nonsense to this thread.

If you don't agree, don't want to participate, or just don't like this thread, simply stay out of it.

 

[Drinkyoghurt]

btw at stock it cools amazingly:

IDLE temp: 21c on both cores
LOAD temp: around 31-33c

nice

 

[inkfx]

btw at stock it cools amazingly:

IDLE temp: 21c on both cores
LOAD temp: around 31-33c

nice

Could you get temp readings with CoreTemp? I have trust issues with TAT.

stock temps look pretty.