Xigmatek HDT-S1283 & SD964 "heatpipe direct-touch&q

Want to talk about one of the articles in SPCR? Here's the forum for you.
rpc180
*Lifetime Patron*
Posts: 309
Joined: Sun Feb 09, 2003 8:01 pm
Location: Washington, DC

Post by rpc180 » Sat Apr 05, 2008 8:17 pm

The kit's specific to the Intel mount unfortunately.

Bluefront
*Lifetime Patron*
Posts: 5316
Joined: Sat Jan 18, 2003 2:19 pm
Location: St Louis (county) Missouri USA

Post by Bluefront » Sun Apr 06, 2008 2:18 am

My take on this setup....the aluminum ribs between the heat-pipes at the base are there for a reason. They prevent the pipes from being crushed during the mounting procedures. Those heat-pipes are relatively fragile. I had to bend an 8mm pipe once before.....it can almost be crushed by finger pressure.

Looking at the base....the round heat-pipes are placed in the aluminum base. Then the setup is pressed relatively flat. After that, the base is machined slightly.....enough to attain the final base you see. The walls of heat-pipes are very thin.....not much metal can be removed by the machine process.

Without the aluminum base, you'd have thin-walled copper heat-pipes that would loose their flatness with the slightest pressure.

Remember the Scythe "Heatlane" technology......this used a wide, flat heat-pipe, rather than the usual round heat-pipes. This was a fragile device, that needed support metal on both sides to prevent crushing. And this is the weak point of the technology. The wide "pipe" never could make good contact with it's own heat-spreaders. (there was no good way to machine the "Heatlane" completely flat).

Anyway...

Wibla
Friend of SPCR
Posts: 779
Joined: Sun Jun 03, 2007 12:03 am
Location: Norway

Post by Wibla » Sun Apr 06, 2008 6:23 pm

Performance speaks for itself, I'm getting one to replace my sh*tty Nexus SkiveTek PHT-7750

Mats
Posts: 3044
Joined: Fri Dec 26, 2003 6:54 am
Location: Sweden

Post by Mats » Sun Apr 06, 2008 8:20 pm

Bluefront wrote:My take on this setup....the aluminum ribs between the heat-pipes at the base are there for a reason. They prevent the pipes from being crushed during the mounting procedures. Those heat-pipes are relatively fragile. I had to bend an 8mm pipe once before.....it can almost be crushed by finger pressure.
I disagree. First of all, there are heatpipes that have thicker walls than the ones you described.
You don't need the ribs to keep the pipes from damage, as long as you have the aluminium base touching the IHS closer to the edges where there's less heat dissipation.
Imagine just keeping the two ribs that are closest to the mounting holes in the pic I posted earlier.
This would make the HS less useable for CPU's without an IHS, of course.

The (inner) ribs are there to keep the pipes in place, with this construction they get support from three sides.
Without them you need to fixate the pipes to the base higher up.

nitram_tpr
Posts: 96
Joined: Wed Feb 06, 2008 5:40 am
Location: Sarf of Engerlund

Post by nitram_tpr » Mon Apr 07, 2008 12:31 am

If the pipes are empty then yes they would be pretty fragile.
I have always been under the assumption that heatpipes have some sort of liquid in them and that they are sealed at both ends. That gives the heatpipes some pretty good structural integrity due to the internal pressure and the fact that pressure isn't allowed to escape.
You try crushing a garden hose pipe when no water is in it, it is fairly easy, and then try doing it when the pipe has water in it and the water isn't allowed to flow, it will be rock hard.

floffe
Posts: 497
Joined: Mon May 08, 2006 4:36 am
Location: Linköping, Sweden

Post by floffe » Mon Apr 07, 2008 1:03 am

nitram_tpr wrote:If the pipes are empty then yes they would be pretty fragile.
I have always been under the assumption that heatpipes have some sort of liquid in them and that they are sealed at both ends. That gives the heatpipes some pretty good structural integrity due to the internal pressure and the fact that pressure isn't allowed to escape.
You try crushing a garden hose pipe when no water is in it, it is fairly easy, and then try doing it when the pipe has water in it and the water isn't allowed to flow, it will be rock hard.
But heatpipes are pretty much empty, they only hold a tiny amount of liquid (often water), and at pretty low pressure to allow the water to boil at ~0C instead of 100C @ normal pressure. If they'd be filled with a liquid pressure would get much higher as soon as some of it evaporated, which could break the pipe (water, for example, expands over 1500x when evaporating).

NeilBlanchard
Moderator
Posts: 7681
Joined: Mon Dec 09, 2002 7:11 pm
Location: Maynard, MA, Eaarth
Contact:

Post by NeilBlanchard » Mon Apr 07, 2008 2:35 am

Greetings,
I wrote:Please remember a couple of things: The heatspreader is (relatively) flat, and it is wider than ~2 of the heatpipes. Also, there are aluminum "ribs" between the hollow copper heatpipes -- so, the aluminum would have to also be crushed in order to bend the copper heatpipes.

It isn't called a "heatspreader" for nuthin' -- all the heatpipes are in contact with the heatspreader. Also, the grooved aluminum base is in contact with the heatspreader and with the heatpipes; so I think that the heatpipes that are a little off center from the actual CPU will still get heated.

In fact, they will get heat more "directly" than they do on the big Thermalright. The Ninja has a second level of heatpipes, and I think it is the contact (or lack thereof) with these that hurt the Rev. B's performance.
I'm agreeing with Carl on this. Without the aluminum ribs, when there is enough pressure to get good contact with the CPU, there is no way that a flattened, hollow, copper tube would NOT distort.

Aris
Posts: 2299
Joined: Mon Dec 15, 2003 10:29 am
Location: Bellevue, Nebraska
Contact:

Post by Aris » Mon Apr 07, 2008 7:15 am

i dont see what all the fuss is about. the numbers speak for themselves.

Smaller than a thermalright
Lighter than a thermalight
Cheeper than a thermalright

All with the same perforamance, and slightly better at lower airflow?

The pushpins arnt going to go away until intel dictates it no matter how much we all hate them. Cant really blame the heatsink makers, they are afterall just following intel's specifications.

Its a slam dunk people. Go out and buy one and stop complaining.

FartingBob
Patron of SPCR
Posts: 744
Joined: Tue Mar 04, 2008 4:05 am
Location: London
Contact:

Post by FartingBob » Mon Apr 07, 2008 7:42 am

Aris wrote:
The pushpins arnt going to go away until intel dictates it no matter how much we all hate them. Cant really blame the heatsink makers, they are afterall just following intel's specifications.
They could at least offer the option of a backplate. Its not hard to take off the pushpins and put them back together, bundle in a backplate as well and everyone will be happy, and it'll only add a few dollars at most for them to do it.

NeilBlanchard
Moderator
Posts: 7681
Joined: Mon Dec 09, 2002 7:11 pm
Location: Maynard, MA, Eaarth
Contact:

Post by NeilBlanchard » Mon Apr 07, 2008 8:43 am

Aris wrote:i dont see what all the fuss is about. the numbers speak for themselves.

Smaller than a thermalright
Lighter than a thermalight
[Cheaper] than a thermalright

All with the same [performance], and slightly better at lower airflow?

The pushpins arnt going to go away until intel dictates it no matter how much we all hate them. Cant really blame the heatsink makers, they are afterall just following intel's specifications.

Its a slam dunk people. Go out and buy one and stop complaining.
Ditto -- the KingWin version is the best deal at NewEgg ($4.01 less shipped than the Xigmatek) -- it is $40 shipped:

http://www.newegg.com/Product/Product.a ... 6835124019

figment
Posts: 232
Joined: Fri Feb 01, 2008 10:15 am
Location: "Fake" Virginia, US

Post by figment » Mon Apr 07, 2008 11:17 am

While I really enjoy SPCR and the review made for a good read, I really wish MikeC wouldn't have published this review.

This is the heatsink I had targetted for my upcoming build. I'm still about 3 months out and now that SPCR and AnandTech and everyone else is starting to notice this HSF, I'm worried that I'm going to have troubles finding it in the near future. Being smaller, lighter, and cheaper than the TRUE, but from a (potentially?) smaller company, I hope they don't end up selling out at retailers like the TRUE is.

I'll be getting the TR bolt-through kit, though. I really hate pushpins. And I'm looking to replace the stock fan with a Scythe Kama PWM fan. I suppose in the end, its only going to be about $10-15 cheaper than the TRUE, but I really like not having to worry about stressing my board with the TRUE.

Mats
Posts: 3044
Joined: Fri Dec 26, 2003 6:54 am
Location: Sweden

Post by Mats » Mon Apr 07, 2008 1:35 pm

Just to clarify, I didn't mean that there should be no ribs at all, one on each side must be kept. I think that is what andyb meant as well.

It just wouldn't work with no ribs, like everybody else says.

butters
Posts: 64
Joined: Sun Feb 18, 2007 9:48 pm

Post by butters » Mon Apr 07, 2008 3:20 pm

I'm convinced that mounting pressure is just as significant as thermal resistance in evaluating heatsinks. It's not immediately obvious to me how this could be measured quantitatively in a method consistent with SPCR's ideology, but perhaps somebody smarter than me will figure it out.

This multi-brand "family" of heatsinks tops off with the iCEAGE Prima 120, which has four 8mm heatpipes, a noticeably improved surface quality, more fin area, and an impressively low thermal resistance of 0.09 C/W (half the listed resistance of the 120mm heatsink tested here). But if the mounting pressure isn't as good, then the overall performance could be worse.

What I'm saying is that it's pointless to debate the finer points of heatsink design unless we can eliminate the variable of mounting pressure. It's already the conventional wisdom of the SPCR community that the current versions of the Ninja are limited by insufficient mounting pressure. The sample tested in this review apparently had its push-pins configured for substantially more pressure, and the test results bear that out.

IMHO, SPCR should test all heatsinks with a spring-loaded bolt-through mount as well as the stock mounting system. This way we can ensure that the products are tested at a consistent mounting pressure and send a clear message to the manufacturers as to just how much performance they are leaving on the design bench by not shipping their heatsinks with the obviously superior bolt-through mount system.

Mike, if you really dislike the push-pin system, you have a pretty nice soapbox from which to spread the bolt-through gospel.

Bluefront
*Lifetime Patron*
Posts: 5316
Joined: Sat Jan 18, 2003 2:19 pm
Location: St Louis (county) Missouri USA

Post by Bluefront » Mon Apr 07, 2008 5:05 pm

Could someone please explain to me just how more pressure between two 100% flat surfaces could improve thermal transfer? The idea behind thermal paste is to fill-in slight surface imperfections, removing the small air pockets. More pressure would flatten warped surfaces (supposedly), but that's about it. If the surfaces were completely flat against each other to begin with, neither TIM nor great pressure would be necessary.....IMHO.

MikeC
Site Admin
Posts: 12283
Joined: Sun Aug 11, 2002 3:26 pm
Location: Vancouver, BC, Canada
Contact:

Post by MikeC » Mon Apr 07, 2008 5:21 pm

Bluefront wrote:Could someone please explain to me just how more pressure between two 100% flat surfacescould improve thermal transfer? The idea behind thermal paste is to fill-in slight surface imperfections, removing the small air pockets.
You just answered your own question. Both surfaces are never "100% flat" whatever that means. Actually, that phrase is meaningless if we're talking scientifically. It's more accurate and realistic to say "flat enough" -- whatever that means. :lol:

In any case, it is my experience with many dozens of heatsinks tested that higher pressure usually leads to lower temperature. If it's not very tight, then simply pushing down firmly on the HS with a finger or two is enough to lower the CPU temp a couple degrees almost immediately. If it is very tight already, then added pressure makes no or little difference.

Bluefront
*Lifetime Patron*
Posts: 5316
Joined: Sat Jan 18, 2003 2:19 pm
Location: St Louis (county) Missouri USA

Post by Bluefront » Mon Apr 07, 2008 6:10 pm

So it's a matter of lapping both surfaces, or machining the surfaces better, or designing a retention mechanism that holds the surfaces flatter against each other.....not particularly pressure.

Pressing down on a heatsink harder to get better thermal contact, simply proves the surfaces were not touching each other correctly, something that could be improved by other methods besides more pressure....which warps boards and breaks retention devices. IMHO

cmthomson
Posts: 1266
Joined: Sun Oct 09, 2005 8:35 am
Location: Pleasanton, CA

Post by cmthomson » Mon Apr 07, 2008 6:40 pm

In pedestrian cooling systems, any TIM with any pressure will perform adequately.

However, at the high end, a top-quality TIM (eg, AS5 or any other silver-based TIM) must be extremely thin to achieve maximum performance.

Given this, there are two variables: flatness and pressure. Lapping improves flatness, but again for maximum performance, thinness is important.

Thus mounting pressure is important, even when both surfaces are very flat. In the case of LGA775 CPUs, there have been many reports that the IHS is not really flat unless subjected to at least 60lb of pressure (I have no idea whether this is true).

But what I do know is true is that with AS5, higher mounting pressure is always better, and when coupled with a high-performance heat sink such as the original Ninja, the resulting performance is outstanding.

matcote
Posts: 39
Joined: Tue Jun 05, 2007 6:16 pm
Location: Québec

Post by matcote » Mon Apr 07, 2008 7:50 pm


MikeC
Site Admin
Posts: 12283
Joined: Sun Aug 11, 2002 3:26 pm
Location: Vancouver, BC, Canada
Contact:

Post by MikeC » Mon Apr 07, 2008 8:13 pm

Wow, that's quite an imprint.

I had a look at our 1263 sample. It has some slight marks but nothing like the above. Perhaps the example above is of sample that was handled roughly during installation? Or maybe just left on the CPU/mobo for a long time. That HS does seem to apply a fair bit of pressure. Mind you, just about every base develops some kind of blemish at the perimeter of the CPU heatspreader.

Image

Bluefront
*Lifetime Patron*
Posts: 5316
Joined: Sat Jan 18, 2003 2:19 pm
Location: St Louis (county) Missouri USA

Post by Bluefront » Tue Apr 08, 2008 1:03 am

Yipes....that picture proves how easily those heat-pipes can be crushed. But I suspect that particular heat-sink sample was somehow manufactured out of square.....either that or the board retention device was faulty. Something to consider/check before installing one of these devices.... :?

zoob
*Lifetime Patron*
Posts: 380
Joined: Tue Apr 13, 2004 7:12 pm
Location: Toronto, ON
Contact:

Post by zoob » Tue Apr 08, 2008 3:57 am

It could just be a horribly concave heat spreader.. I've had to lap so many C2D CPUs because of high temperatures at stock speeds.

psyopper
Friend of SPCR
Posts: 336
Joined: Fri Mar 28, 2008 1:15 pm
Location: Portland, Oregon

Post by psyopper » Tue Apr 08, 2008 7:29 am

It's an interesting conundrum, this heat pipe crushing issue. One could argue that "crushable" heatpipes, assuming that the overall structural integrety remains solid, maximizes contact area with the heat spreader on the CPU without the need for lapping either the heat sink or the processor.

Crush washers are commonly used in automotive bolt-seal applications, ironically usually copper, where the pressure on the washer between the bolt face and the block curshes the washer into an airtight seal at lower torque values. This is an alternative to a bolt face to block direct seal or forcing a seal through the threads themselves which nets a positive seal while reducing thread strain on the bolt and preventing over-torquing, bolt stretch, etc.

This "crush-factor" may actually be a significant contributor to the better than expected results of these heatsinks. I don't think it was an intended feature by the manufacturer.

[EDIT: Horrible sentence structure fixed]

jhhoffma
Posts: 2131
Joined: Mon Apr 25, 2005 10:00 am
Location: Grand Rapids, MI

Post by jhhoffma » Tue Apr 08, 2008 7:37 am

psyopper wrote:Crush washers are commonly used in automotive bolt-seal applications, ironically usually copper, where the pressure on the washer between the bolt face and the block curshes the washer into an airtight seal at lower torque values. This is an alternative to a bolt face to block direct seal or forcing a seal through the threads themselves which nets a positive seal while reducing thread strain on the bolt and preventing over-torquing, bolt stretch, etc.

This "crush-factor" may actually be a significant contributor to the better than expected results of these heatsinks. I don't think it was an intended feature by the manufacturer.
[EDIT: Horrible sentence structure fixed]
I doubt it. While copper gaskets/washers have their uses (I used them with an ultra-high vacuum chamber in college) I would say that crushing a heatpipe from its designed specs would ever be "good" for it.

I'd much rather have the lap the IHS for good contact than have to distort the HS device to make it usable. Lapping an IHS makes it better regardless of the HSF used. Crushing a heatpipe essentially makes it "one and done".

psyopper
Friend of SPCR
Posts: 336
Joined: Fri Mar 28, 2008 1:15 pm
Location: Portland, Oregon

Post by psyopper » Tue Apr 08, 2008 8:53 am

jhhoffma wrote: I'd much rather have the lap the IHS for good contact than have to distort the HS device to make it usable. Lapping an IHS makes it better regardless of the HSF used. Crushing a heatpipe essentially makes it "one and done".
Right, but with the compound Aluminum and Copper base in this HS lapping would be more difficult. Not impossible, but greater care needs to be taken because with the differing hardnesses of the two materials. Besides, it looks as though the heat pipes have already been machined, and with the "crushability" I don't think I'd want to take much more material off of it anyhow.

As for "one and done," with the exception of super-power users who have the time, money and inclination to repeatedly swap heatsinks to save yet one more degree C, I would say that most common-average users don't swap and sell heatsinks like Magic cards.

NeilBlanchard
Moderator
Posts: 7681
Joined: Mon Dec 09, 2002 7:11 pm
Location: Maynard, MA, Eaarth
Contact:

Post by NeilBlanchard » Tue Apr 08, 2008 9:00 am

Hello,
psyopper wrote:It's an interesting conundrum, this heat pipe crushing issue. One could argue that "crushable" heatpipes, assuming that the overall structural integrety remains solid, maximizes contact area with the heat spreader on the CPU without the need for lapping either the heat sink or the processor.
I think you may be onto something with this. Obviously, whatever is happening at the interface between the CPU and the HS is working. :o

Another interesting thought occurs to me: if a typical CPU touches a little less than the width of 3 heatpipes, then with the iCEAGE Prima 120 HS, it will touch the center 2, and just touch about half of the outer 2 heatpipes (since the aluminum ribs are narrower):
Image
(click on image for link to OEM)
Last edited by NeilBlanchard on Tue Apr 08, 2008 10:53 am, edited 2 times in total.

psyopper
Friend of SPCR
Posts: 336
Joined: Fri Mar 28, 2008 1:15 pm
Location: Portland, Oregon

Post by psyopper » Tue Apr 08, 2008 9:58 am

NeilBlanchard wrote:Another interesting thought occurs to me: if a typical CPU touches a little less than the width of 3 heatpipes, then with the iCEAGE Prima 120 HS, it will touch the center 2, and just touch about half of the outer 2 heatpipes (since the aluminum ribs are narrower):
What's really interesting about the 3R iCEAGE is it's 775 retention mechanism. They include a plastic bracket that attaches to the motherboard via the Intel holes, the heatsink is then attached to the bracket like an AMD-AM2.

BTW- I can't seem to find any US based vendors for the 3R Systems Iceage 120 and the US distributor link at Frostytech works but has nothing about this heatsink.

NeilBlanchard
Moderator
Posts: 7681
Joined: Mon Dec 09, 2002 7:11 pm
Location: Maynard, MA, Eaarth
Contact:

Post by NeilBlanchard » Tue Apr 08, 2008 10:42 am

Hello,

Then that means it can only be mounted in two positions (180 degrees apart); whereas the KingWin/Xigmatek can be mounted in four positions (90 degrees apart).

psyopper
Friend of SPCR
Posts: 336
Joined: Fri Mar 28, 2008 1:15 pm
Location: Portland, Oregon

Post by psyopper » Tue Apr 08, 2008 5:12 pm

NeilBlanchard wrote:Hello,

Then that means it can only be mounted in two positions (180 degrees apart); whereas the KingWin/Xigmatek can be mounted in four positions (90 degrees apart).
Fortunately it can be rotated in 90 degree increments. Remember that the 775 holes are square to each other so it's simply a matter of rotating the base 90 degrees when mounting to the board.

Check out the review at FrostyTech. The last picture on the first page shows the mounting for the 775.

matcote
Posts: 39
Joined: Tue Jun 05, 2007 6:16 pm
Location: Québec

Post by matcote » Tue Apr 15, 2008 10:31 am

are they available in canada?

cmthomson
Posts: 1266
Joined: Sun Oct 09, 2005 8:35 am
Location: Pleasanton, CA

Post by cmthomson » Tue Apr 15, 2008 5:10 pm

It occurred to me the other day that there is another variable we haven't been discussing in this pressure vs performance debate: the internal TIM.

LGA775 CPUs have a TIM between the CPU die and the IHS (inside the CPU case). Maybe pressing harder on the top of the IHS improves the performance of this TIM.

Unfortunately for us in forum-land, only Intel engineers could possibly answer this question, and they aren't likely to...

Post Reply