Apogee from Swiftech...

[Albigger]

Originally Posted by nikhsub1

The interesting part of pH's idea is that if say 2-3 blocks were tested with and without the IHS, we can get an idea of what role the IHS actually has in terms of core temp. My thought is that the IHS will make the die temp go UP in a linear fashion regardless of the block when measuring CORE temps. I could be wrong but I don't think so. So, Block A reads 15C with IHS and reads 12C without IHS. Block B reads 14C with IHS and 11C without IHS. Block C reads 13C with IHS and 10C without. At least, I hope that is the case, in which case, couldn't the IHS be taken out altogether to remove that variable?

I'm not sure I agree, in the strictest of senses. How many waterblocks would you have to test before you felt confident saying there is a linear offset, so then no need to test with an IHS - 3? 4? 5? It would only take ONE block not in agreement to disrupt this 'conclusion.'

[nikhsub1]

Absolutely right Al, but if 3 were in agreement, I think we would be good.

[pHaestus]

Eh or a nonlinear relationship (then you'd need LOTS of points right?)

[Cathar]

Originally Posted by unregistered

pH
think simple, you only have to characterize the TIM joint once
then assume all good joints will measure the same

See, this is where I have the most trouble with. That assumption that a flexible variable TIM joint doesn't change with pressure profiles from flexing/mounting variations.

[pHaestus]

I would think that as long as you're measuring it with a tc then you gain a bit of confidence in the whole system. The "do it once and forget it" troubles me, and the "skip IHS and assume it's just an offset for all blocks" also troubles me.

[pHaestus]

OK since think simple is one piece of advice...

What about just trying to work a piece of fishing line in between the block base and the IHS top when remounting and break the seal by "flossing" the CPU? Simple enough?

[Albigger]

Originally Posted by Cathar

See, this is where I have the most trouble with. That assumption that a flexible variable TIM joint doesn't change with pressure profiles from flexing/mounting variations.

I think this would be very important especially with the trend to thinner baseplates. Badly designed waterblocks (ones with excessive flex) may only put pressure on the IHS edges when clamped down, going concave over the core and resulting in less pressure on the IHS->core joint and less efficient heat transfer, and a hotter core.

A more rigid waterblock base would put more pressure direclty on the core, and result in a lower dT (core to IHS temp). This is one reason I highly doubt a constant linear offset would work.

Originally Posted by pHaestus

I would think that as long as you're measuring it with a tc then you gain a bit of confidence in the whole system. The "do it once and forget it" troubles me, and the "skip IHS and assume it's just an offset for all blocks" also troubles me.

I think this is good news, as you're the one to do the testing, and if you don't like it you're more likely to do more tests (both ways) so that we can all (hopefully) gain a better understanding

[Orkan]

What about just trying to work a piece of fishing line in between the block base and the IHS top when remounting and break the seal by "flossing" the CPU? Simple enough?

Genious.

[bigben2k]

Originally Posted by unregistered

...Ben
compare (speculate on) the repeatability of the assembled TIM joint vs. a factory one ?
(Shin Etsu is silk-screened, that grease will not be used; solders are out, no procedure; we know AS is not used; next ?)

Ok, speculation:

I'm not counting on the ability to replicate the inside TIM joint, exactly the same way that it is manufactured. I'm counting on the ability to apply an offset, which means that we have to have a repeatable TIM joint during testing, both inner and outer.

So... this means that we have to have a TIM joint that can consistently be repeated. (great, now that I'm caught up...)

How do we make sure it's repeated? Back to Cathar's objection; it's not quantified. PH remounts 5 times, and measures variations, but it doesn't actually quantify the TIM joint. Quantifying it would require a temp measurement on both sides of the TIM joint.

So... we add a thermal probe down low in the heat die (horizontally, of course), and do a dT between it and the IHS?!? (how much more coffee am I going to need today to sort this out? )


Orkan; I don't think that anyone is going to dispute that an IHS does spread out the heat. What's really at issue is how significant it is. Then it comes down to wether or not it can be compensated by an offset, when using a straight die. (Darn, ya'll have caught up to all this while I was typing!)

Someone (Les? Groth?) projected a 1mm thick IHS, but it looks like it's actually 1.4 mm.


Flossing ?!?

[Albigger]

Originally Posted by bigben2k

Orkan; I don't think that anyone is going to dispute that an IHS does spread out the heat. What's really at issue is how significant it is. Then it comes down to wether or not it can be compensated by an offset, when using a straight die. (Darn, ya'll have caught up to all this while I was typing!)

Flossing ?!?

How significant the spreading is will depend on the specific waterblock cooling properties, I suspect. However the variance may be negligible, again, won't know till we try to quantify it.


Also confused on the flossing, PH - are you talking about between the actual core and the IHS or just around the edges/seal of the IHS?

[Joe]

hes talking about between the block and the IHS - this goes back to his concern about busting the CPU having the CPU and the block stuck together.

[stev]

Originally Posted by Cathar

... I am questioning the effects of block geometry, and the applied pressure and deviations in effect as a result of this. 1.5mm thick copper flexes fairly easily, we know this. The edges of the IHS are mounted against the packaging substrate with a compressible bond. The IHS will warp and flex in accordance with the pressures, and locations of the pressure applied as a function of the block geometry.

I agree with the statement above since the Corsair HydroCool pre-production and review cold-blocks, being 1.70mm in base thickness, warpped making a concave base after the braze process. We had to quickly scramble to change the braze process and the silver-solder material to maintain flatness for the production run of parts.

However, those blocks were made for the early P4 478's and AthlonXP's loadings.

Today's 70-90lbs of force for the latest processors would require a circular step(s) to add strength to the thin base thickness.

I can understand, all to well, how a think base can flex on a cold-block.

Stev

[nikhsub1]

Originally Posted by Albigger

waterblocks (ones with excessive flex) may only put pressure on the IHS edges when clamped down, going concave over the core and resulting in less pressure on the IHS->core joint and less efficient heat transfer, and a hotter core.

This is PRECISELY what I think is happening with the TTV... now if a core temp was ALSO attainable with the TTV...

[Albigger]

Originally Posted by nikhsub1

This is PRECISELY what I think is happening with the TTV... now if a core temp was ALSO attainable with the TTV...

Right. But its not attainable, and neither are TTV's themselves according to Bill. But basically the talk earlier in the thread was toward a heat die with and IHS with BOTH temps known, so we could quantify this (although it would only be representative of a CPU or TTV if the edges of the IHS were supported as well).


The conclusion of that thought process? Method too complicated?



Have to think more but may be advantageous of two different setups:
Setup A: CPU with stock IHS, modified for a temperature measurement at the IHS surface. In theory both IHS and core (ondie) temp are known.

Setup B: Heat die with separate IHS, temp sensor in both die and IHS surface.

If results correlated between the two setups with the IHS on, the blocks could also be tested on the heat die without the IHS to indicate both IHS and bare-core perfomance?

[bigben2k]

Originally Posted by Albigger

How significant the spreading is will depend on the specific waterblock cooling properties, I suspect. However the variance may be negligible, again, won't know till we try to quantify it. ...

Wrong (mostly); the heat flux flows through the IHS regardless of the cooling solution. The heat flux however may take a bit of a variable path, but in no way significant to the effect of the IHS itself; the IHS is pretty good at smoothing things out, as is the baseplate of the water block itself.

Specifically, if the heat flux is concentrated within a smaller area, then there's a "hot spot" and conversely, if the heat flux is spread out, then the overall temperature is lower, but the flux remains the same.

The size of the area through which the heat flux flows can have an impact on the performance of a water block. As we've demonstrated though, it's not significant.

Al, agreed, comparing the two would be interesting, just to put out this issue. FYI, Storm's design would be most susceptible to the difference. The big Q though; can it be compensated with an offset?

[jaydee]

Originally Posted by Marci

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....

Note I was refering to Socket A cpus on the 25lbs not anything above that being that is what I was setup for.

[bigben2k]

Originally Posted by Albigger

...
Have to think more but may be advantageous of two different setups:
Setup A: CPU with stock IHS, modified for a temperature measurement at the IHS surface. In theory both IHS and core (ondie) temp are known.

Setup B: Heat die with separate IHS, temp sensor in both die and IHS surface.

If results correlated between the two setups with the IHS on, the blocks could also be tested on the heat die without the IHS to indicate both IHS and bare-core perfomance?

Many problems:

1) the power level would have to be matched (*very* hard to do)

2) you're not likely to be able to mount a temp probe within the heat die, to replicate what a CPU would return, due to physical limitations.

[plywood99]

Originally Posted by pHaestus

OK I am talking in Pro/Chat with Scott.

I think I might have what is the best solution for me:

1) Test a waterblock with IHS
2) Pop off IHS and retest
3) Fiddle with the CPU until I hit upon a method to reliably reaffix the IHS in a way that yields similar temps to (1) but the IHS separates from the CPU when I change coolers so I don't destroy my CPU.

There are a lot of advantages to this:
I could put a TC into the IHS and then get tim joint estimates
I get better CPU diode temps than you can from any motherboard
I don't destroy my CPU within a week or so
I don't have to argue with the die sim folks
I get the flexibility of choosing whether to test bare core or IHS

Is there a disadvantage?
It might take some fiddling to get the IHS back on totally reproducibly
It is an alteration from a pristine CPU (but so was this right?)

Concerning issue three. You could set up ihs to be removable per mount. I am considering doing this myself.
1) Remove ihs.
2) Coat the top and outside edge of ihs with wax.
3) Now remount ihs.
4) Use JB Weld or some other epoxy to form a ring around ihs. This insures ihs will be placed in same spot through frequent removals.

This would negate testing bare die. But guess you could also have a small jig machined to keep ihs in same place also.

[pHaestus]

Think I like the idea of flossing the CPU much better than putzing with the IHS. Gotta test it out though when I can get a spare hour.

[plywood99]

As was mentioned, A64 tim joint degrades over several mountings. I think they use Shin etsu compound for the tim. Most certainly is not a glue or epoxie. As much as I hate Intel, they have by far the better ihs mounting...

[Marci]

Hmmm, just straying back to the main topic... knew Swiftech had replaced Storm in the kits, but didn't know they'd completely discontinued it...

I was going to return my Storm for exchange, but I was told the Storm is now discontinued so I'm SOL.

http://www.ocforums.com/showthread.php?t=426899

[mad mikee]

Originally Posted by pHaestus

OK since think simple is one piece of advice...

What about just trying to work a piece of fishing line in between the block base and the IHS top when remounting and break the seal by "flossing" the CPU? Simple enough?

Why use fishing line, use Glide floss tape ?

[pHaestus]

Originally Posted by Marci

Hmmm, just straying back to the main topic... knew Swiftech had replaced Storm in the kits, but didn't know they'd completely discontinued it...

Why would they keep it in their lineup when it performs worse and costs much more?

Only it doesn't on all CPUs; only for sure on the TTV.

//edit for 100% less speculation

[Orkan]

Originally Posted by pHaestus

Why would they keep it in their lineup when it performs worse and costs much more?

Only it doesn't

Did you get around to testing them? Don't tease us man... post some data.

[Marci]

Demand is still there for it no matter what at the moment, therefore there is still money to be made from it.

Marketing BS - force users over to Apogee by removing the alternative. Sucks.