Great post Orkan.

The tool was not fully sharp, anyway I think that it was not made with mill because it would take too long.

Did you take any measures of pins and channels?

Bill, I NEVER said that the testing methodology was flawed.

I said that the testbed (TTV) was.

Can have the world's best sculpter. You can give him a blunt tool, but don't expect miracles, that's all I'm saying.

Approx Pin array dimensions -

34mm x 34mm
1mm channel between the pins.

About the HS mechanical issue, could it be practical and efficient to use a IHS like this one below and then to secure it in place using the same kind of lever system that Intel has for the socket 775?

http://slike.slo-tech.com/12037sm.jpg
http://slike.slo-tech.com/12027sm.jpg

Lee, I like this arrangement for many reasons but there is one problem with the vertical hole for the temperature sensor that I think is insurmountable.

Basically it means that there is a large uncertainty in the vertical location of the point that is measured which makes it impossible to use the data for any kind of modelling. The vertical shaft will have a different temperature gradient than the copper around it and the position of the sensor is uncertain.
Compare this to a horizontal hole. The gradient (in the vertical direction ) is the same as the surrounding copper, the measured temperature is some kind of average of the temperature of the hole surfaces and does not vary by more than a tiny amount along the length of the hole. In the case of the vertical shaft the same applies, namely that the measured temperature is some kind of average of the shaft surfaces, except that the temperature is varying hugely along these surfaces and a small "position error" would have a large impact on the measured temperature.

Edit: I have drawn something to try and explain explain this, I believe it is very important that you don't do this. The holes should be horizontal. Always. Their effects (heat shadowing) can be modelled, the effects in a vertical shaft can't.

even with the horizontal locating is a bit of a question mechanically
- I guess the 'effective' location could be back calculated with some accuracy

and, as has been discussed here before, need to quantify the clamping force (with load cells ?)
load cells will reveal mounting hardware issues clearly and provide repeatability

more and more sophisticated, how many still in ?
(not to be negative, but these inevitably become so complex than none continue)

if those are solder tabs that may be a TTV, all TTVs are mounted in their respective sockets
since the CPU/socket/mobo will have to be replicated (as part of the source) to mount a sink or wb that locates off the mobo, there is a lot of hardware involved;
other systems let the mobo float between the chassis and the sink,
and it will be a brand new exercise for BTX

It is CNC milled.

You'd think it would be sawed... but it looked milled to me.

Thanks Incoherent,

Yes, I agree with what you are saying. I did a quick calc and for a 14mm^2 cross-area with 100W load thru copper, I get ~1.3 deg C per mm... not so good.

"Today" I am leaning towards a one-piece heat die and heatspreader with two or three temp sensors located in the sides of the die to generate a heat flux curve and that use an extrapolated temp value for calculating the WB dT and C/W. I also like the idea of a rectangular die (maybe 14 x 16mm?) since that seems to be the direction more of the newer CPU's are going in. (I know, much controversey there... :) )

Bill: Yes, I have considered integrating a load cell underneath the die and creating the upwards force by regulating the air pressure to a small cylinder instead of a spring. Considered "clamping" the WB under test and then pushing the die/heat spreader up with a known force. This is great for analytical testing but removes the variability of specific mounting hardware that the end user will have to deal with. As you said though, complexity often leads to no action...

knew there'd be a proper name for what was going thru my mind... *makes a note of Load Cell*

hmm - just a (lateral) thought, but could you overcome this by using a spring to push upwards on the sensor in the vertical hole?

There's a spring variant known as a constant force spring. It's a tightly wrapped band of spring steel that has a known load force, which varies very little over its active range because it has a very small spring constant and a very large range (most of which is not active).

I haven't seen a constant force spring that's designed for compression, only tension, but that doesn't mean they don't exist. You might be able to rig some kind of crazy lever system to get compression. I'd go that route before looking at using pneumatics.

I don't think so. The problem isn't so much the sensor position itself, it's the position of the focus of measurement. Hard concept to get across, but in a vertical shaft you are measuring the temperature of the air (or whatever it is filled with) in the hole as well as the surrounding copper. The point within the sensor that the temperature reading represents is in an unknown location but in a horizontal hole it will be more sure that it is halfway between the top and the bottom of the horizontal hole (in terms of gradient) with the conditions in the Axis of interest (vertical) being the same, as opposed to the vertical one where the conditions up and down are totally different (copper one side, air (or whatever) on the other).


It is one of my concerns with the TTV which experiences a similar effect, but they are mitigated somewhat by a very concise description on how to affix the TC to ensure a good contact and thus ensure that the temperature measured is in fact the temperature at the point expected. I am reasonably happy with the TTV concept, although it is not the way I would have done it. In working with Intel I have found that they generally know what they are doing.

Is it just me.. or are many online retail stores not selling the Storm any more? And at the same time, the price seems to have increased.

This leads me to the conjecture that Swiftech is end of lifing the Storm.

Kinda goes without saying doesn't it? They replaced the storm in all their apex kits. Their intentions seem pretty well stated.

Here's a good point: the Apogee will work well as a TEC block too.

the pin array is only 34mm^2 ... that enough cooling area to work on a 50mm^2 TEC like a 200+ watt?

Maybe, maybe not.
Notwithstanding, predicted performance on 32x32mm Heat -Die is no better than MCW6000 - see up-dated Post133.

I don't know if you guys are aware of this thread on [H], slightly worrying :/

Uhh, just look one page back in this thread. http://forums.procooling.com/vbb/sho...&postcount=300

That's the complication I was worried about.

From my perspective, the IHS is mounted in a way to spread the applied mounting pressure between the core area and the rim of the IHS. In what proportion, we do not know.

To estimate it, we'd need all of the measurement of *everything* (core, substrate, IHS), then we'd need to know how thick the seal is on the rim of the IHS, then we'd need to know how hard it(seal) is; then we could quantify how the mounting pressure is spread out.

It's a nightmare.

The load cell idea is great, but I'd rather use the heat dies straight out (10 by 10 and 14 by 14mm), no IHS, and apply an offset for temperature calculations (if we can ever figure them out), which should make up for the missing IHS in the test bench.

Anyone disagree?

Alternatively, we could put on a free-floating IHS, and use mounting pressure similar to bare core mount specs (i.e. 24 to 30 lbs), but that assumes that that's the actual pressure.

Hummm,

If the IHS temp is all the TTV uses then how is that different than the die temp of a die sim? The only thing that bothers me about the IHS is the heat spread/flux. Without putting dozens of probes all over the IHS we really don't know how much of the IHS is being cooled/warmed. Therefor guessing a dimension is useless without the needed data from dozens of calibrated probes of insane resolution.

I say **** it and will be using a die sim. :) :nod:

Remember this? Even though a heat spreader is used, the heat die (core) size still plays a major role...

http://pages.sbcglobal.net/water.gro...at-Flux1mm.gif

Post #57 by Groth http://forums.procooling.com/vbb/sho...3&page=3&pp=25

Yup. This is why I don't understand/like the taking of the IHS temp only. It would be nice to see how well our dies are being cooled as well as the IHS. Also that image would very from block to block being the cooling area inside the blocks are not all the same.

More reasoning IMO to stick with die sim.

The IHS temp measurement (Intel style) is easy to do, and can be done by anyone else, in the same way; that's why it's so tempting.

Otherwise the simulated core temperature can be calculated by an offset (which granted isn't easy to quantify, but not impossible).

So JD, what clamping pressure will you be using?:sneaky:

I'm thinking about an Intel-style temp measurement, using a die simulator without an IHS.

Its my understanding that is why the IHS temp probe must always be mounted directly over the center of the "core" - heat die. This is where the temp should be the highest and can be estimated/calculated the most accurately.

Re: Apogee from Swiftech...
 

I don't use Intel and probably never will unless AMD goes under. I have no interest in testing under Intel conditions.

Clamping pressure? Was using 25lbs. Not sure what the new AMD CPUs are recommended at. Probably 50? The stock HS I just put on my sempron 2500+ Socket 754 was amazingly hard to lock down. Has to be a lot of pressure. Already having nightmares thinking about getting that thing off not to mention the thermal paste suction effect. The HS fits perfectly in the little square and cannot be moved side to side to break the suction....

At this point I am just about to the "**** testing" stage again. Getting to ridiculous.

Re: Apogee from Swiftech...
 

Oopsy - want minimum 60lbs, max 90lbs - http://www.amd.com/us-en/assets/cont...26633_5649.pdf - Page 18

http://marci.over-clock.com/amd64tsd.jpg

Appendix A makes good reading - page 37 onwards. Page 41 goes on about Load Cells for testing heatsink downforce, and specifically which make and model to use....

Re: Apogee from Swiftech...
 

Snide's Model of Storm G4
Taken as: 35 Cups(3.1242mm Diam x 4mm Deep) in 4.7625mm Base with 0.7366mm(ID) Jets
Modelled as: 37Cups(3.1242mm Diam x 4 mm Deep) in 4.7625mm Base on a 20.7x20.7mm Area, with 0.7366mm(ID).
Cup walls modelled as hexagonal Arrays of (0.4or0.8) x3.1x4mm Pins on 0.7625mm bp Effective Base Area=21.5452x21.5452 mm = 464.195643 mm^2
Effective Cup Base Area=330.8391314 mm^2
Effective Cup Wall Area=1694.330101 mm^2


Notes on Modelling
All very crude
Applies to uniform flux sources and are average Temps and Watts
All need fudge factors for h(conv)
All need better correlation between h(conv) and h(eff) : particularly Storm G4
Dimensions are sometimes assumed and/or hearsay:applies particularly to MCW6000 and Storm G4

Edit: Corrected boob in StormG4 h(eff) profile > Stupidly, was showing the "Cupped Area" to be double its actual size