The biggest "Heatpipe" won't make the best Cooler
ฝัง
- เผยแพร่เมื่อ 2 ส.ค. 2024
- Support me on Patreon:
/ der8auer
---------------------------------------------------------
Save 10% on your iFixit purchase: DER8AUER10
eustore.ifixit.com/der8auer
Pro Tech Toolkit: bit.ly/2JOFD8f
---------------------------------------------------------
Find my products at Caseking:
Delid Die Mate 2: bit.ly/2Rhv4y7
Delid Die Mate X: bit.ly/2EYLwwG
Skylake-X Direct Die Frame: bit.ly/2GW6yyC
9th Gen OC Frame: bit.ly/2UVSubi
Debug-LED: bit.ly/2QVUEt0
Pretested CPUs: bit.ly/2Aqhf6y
---------------------------------------------------------
My Equipment:
USB- Microscope*: amzn.to/2Vi4dky
My Camera*: amzn.to/2BN4h2O
(*Affiliate Links: If you buy something, I will get a part of the profit)
---------------------------------------------------------
Music / Credits:
Outro:
Dylan Sitts feat. HDBeenDope - For The Record (Dylan Sitts Remix)
---------------------------------------------------------
Paid content in this video:
- /
Samples used in this video:
- /
---------------------------------------------------------
Timestamps
0:00 Intro
0:43 Cooler Master Masterair G100M
1:06 The cooler in detail
2:46 The cooler from the inside
4:12 Advantages & disadvantages of the cooler
6:42 Outro - วิทยาศาสตร์และเทคโนโลยี
![[ไฮไลต์] แบดมินตันหญิง ไท่ ซือ หยิง vs เมย์ รัชนก อินทนนท์ รอบแบ่งกลุ่ม | โอลิมปิก 2024](http://i.ytimg.com/vi/LUCG0H5qkpQ/mqdefault.jpg)
Now I am kinda wishing you had properly tested its performance to actually show us this saturation effect you're referring to _before_ you had cut into it 😂🤔
130 watt TDP cools the Ryzen 5600X easily with its 65 watt TDP
@@malcolmogilvy4885 Ryzen seems to understate the Box TDP... mine was labeled to be 65W as well... but according to power monitors (HWinfo) mine pulls 140W (OC'd) at 4ghz.
@@BeezyKing99 Well duh, if you OC of course it won't be the same as the box rating :p
That being said, the "commercial TDP' rarely has anything to do with the actual power consumption.
@@Bourinos02 I'm aware, but capabilities and cooling matter the most over box ratings.
@@BeezyKing99 lmao that is overclocked but fair enough ima have an eSports Duo sitting there
Coolers like this will always remind me of the old "Zalman Flower" coolers
In the late 1980's Sony made stereo receivers whose heat sinks used heat pipes. It was the first time I had ever encountered the technology (I am an audio repair tech). Those heat pipes were about 18-20mm in diameter and perhaps 200mm long. I was amazed at how they worked.
Were there cooling fins or fans?
Sony SA-20F
Too cool!
Thanks Scott
@@bloodybucket213 of course. Sony was serious about their best stuff! You can't really use evaporative heatpipes without a mass to sink that heat into. Oh hey, a heatsink!
Also tech person here :) I'm not employed but consider myself and am widely regarded as that audio nerd fixing local soundsystems.
I occasionally assemble some of the Nelson Pass amps and have always wondered what would happen if I could cool the mosfets with heatsinks.
I'm older now. I'm wiser. More tempered. And also richer, which is exactly why I now have a class A amplifier with PC heatsinks on them.
@@bloodybucket213 I looked that number up. I had not seen the powered speakers before. They're using a heatpipe and a heatsink (within a vent, no less!) to cool the woofer. The receivers I remember were conventional except for the use of the heat pipes to carry heat from the output transistors to the heatsink fins, and the the fact that there were many very thin fins (like today's CPU coolers) instead of the thick heat sink fins popular in the 80's and prior. I don't remember seeing any fans, but in some models there might have been.
@@AnnaVannieuwenhuyse I actually have a PASS AM that's awaiting me to fit AMD Wraith heatsinks onto, and a F5 Turbo V2 that im going to use repurposed AIO blocks without pumps, plumbed onto a EK pump res. And also another PASS AM that's going in a traditional chassis with heatsinks running the side of it
The old Zalman Flower coolers were great. I still have them. The fan bracket is still great for adding fans over top of expansion cards and the like.
Look at the speed he cuts through it with the hacksaw. German efficiency!
Any person with experience with a saw can do it, there's nothing "German" about it.
@@itisabird it might have been a joke about the footage being sped up
Forget the cooling efficiency, it looks really cool !
I seem to recall in the very early days of heatsinks for coolers the best was one that looked like a hedgehog, a block of copper with what looked like nails sticking out of it. That was cutting edge in those days, but I have forgotten the name of it.
Your title is the obvious result of the inverse square relationship between volume and surface area (aka the inverse square law) and the fact heatpipes entirely rely on surface area to operate
Another weakness with this kind of design is the surface area of the heatpipe (both internally and externally) is smaller for the same volume of heatpipe than a lower diameter one. This means that the internal area for phase change is smaller, and also the external area for conduction to the fins is smaller.
It would have been better if they made it hexagonal or any shape whatsoever. A cone would have also been cheap and better than just a cylindre
@@laharl2k The problem still stands that there's a higher air volume to surface ratio inside than in a standard heatpipe. You'd have to... I dunno, turn it into some kind of highly exaggerated star shape with a dozen very thin points so that there's very little air volume left in the chamber. Even then, what'd be the point of that? That's a lot of extra manufacturing steps just to get decent cooling out of something that's only ever going to be a marketing gimmick.
@@pirojfmifhghek566 i think this higher air volume actually helps for quicker heat transfer, the vapor can go freely anywhere to condense out, and the condensate just flow to the hot surface.
On a normal thinner heatpipe, the vapor when travels to the cold side, it interferes with the condensate on the walls, and i think it might slow down the actual "evaporate-go to cold side- condense - go to hot side" work inside the heatpipe, becouse they drag at each other.
@@_decky4ever_ Unfortunately, the benchmark tests for this product beg to differ. You can find some pretty damning results on Toms Hardware. They even tested it a second time using a much lower-wattage CPU and it was still at the bottom of the charts. It's an infamously bad cooler.
@@pirojfmifhghek566 then the phisics is not working as i tought, then the smaller pipe is better :)
Heat-column technology was originally used primarily for LED lighting or large-scale heat exchangers. As der8auer explained well, it is an inexpensive process and it has been widely used in thermal solutions of CPU manufacturers such as AMD Wraith Spire and Intel's TS13A, TS15A, FCLGA4-S. The most recent one is the Intel RH1 cooler, which is very hip.
Maybe they should had added more surface into the chamber to improve the surface area, if they wanted to keep with the single huge heatpipe design. Adding pins to the inside (either vertically or horizontally) would make for a good transfer increase.
Also, grab a water jet cutter for your shop, if you plan more "investigative" work in the future, as it causes less damage to the surroundings than a bench saw.
theres no point increasing the interior surface area that would just make it saturate itself faster because youre not actually increasing the EXTERIOR's ability to remove that heat.
plus anything you do to increase the surface area would porbably be more expensive and less effective than just using multiple heat pipes instead
Heat pipes saturate when the rate of evaporation of the fluid exceeds the rate at which capillary action can bring the fluid back down to the location from which it evaporates. Increasing inner surface area increases the rate of capillary action, so that should make it more difficult to saturate the heat pipe. I was really expecting this heat pipe to be stuffed with a wick or something to use all that vacant space to help with this.
Your inner surface will just dry out very quickly because it cannot eject heat to the outside and there is no condensation happening. Without the fluids condensing there is no capillary action.
I just wanna say your workshop looks awesome! The CNC machine behind you, wow! In any case, I kinda like the look of this cooler master ufo looking heatsink, even though it doesn't live to its cooling expectation.
My kitchen isn't that clean
What a weird design! I wonder if they’re actually selling many of those, I’ve never even heard of something like that. I wish you could’ve tested it compared to a similar-sized cooler. (maybe an intel stock cooler?) Would be interesting to see.
Like most things in computing, intel did it first - theres a Core2 Extreme stock cooler with a similar design
I doubt they're very high volume, considering the best low profile cooler on the market bundled with the best slim 92mm fan on the market is only a few bucks more. This looks more like a product that breaks even and let's CM promote a wider cooler range.
@@joemarais7683 probably more 'the tooling is ancient but its making money so why shut it down'
with intel it's actually a slug. copper slug. Even for the latest alder lake i7 12700 cooler. Not sure about the i9 one
with the AMD Ryzen 5 1600 & 1700(and 2700) coolers its an big ass heatpipe like that
Intel had few stock coolers that include a VC slug. QX9650 stock heatsink came with one and the newer TS13A and TS15A are also using VC slug.
Very interesting! Happy you cut one open
Thanks for the lesson. I definitely learned a few things.
You forgot to mention it "looks siiiiick"!
That's a cool cooler
Back in the P4 days I built my own custom loop using a heater core, an aquarium pump, and some fittings from the hardware store. My first "water block" was a heatsink that had a copper core. I drilled holes in V shape and jb-welded some copper tubes to it just so water would flow through the copper core. It performed quite a bit better than the actual purpose-made waterblock that I later ordered. I believe it just had far more surface area since this was before the "micro fins" of modern water blocks.
It is actually large enough to not get into a liquid lock situation which makes it possible to increase the liquid charge in it and avoid the sudden overload you would normally find.
Instead it would just increase in temperature (both hot and cold side) until the cold side can get rid of the heat again. Would be fun though as you would be able to hear the boiling.
If you use such a large bottom surface as evaporator you'd have a hard time feeding the middle of it condensate from the sides. Overfilling a bit would probably help this but I'd expect actual liquid to drip out when cut.
The first time I seen a cooler like this was the stock heatsink, FCLGA4-S , Included with some Core2 Extreme CPUs.
Thank you for confirming my memory wasnt fooling me about intel having a heatpipe rather than the slug on a couple of models - i couldnt find anything googling XD
The Zalman coolers where dustcollectors, the thight slots between the fins closest to the P4 CPU clogged up really fast, my sons computer had one of those and we cut a hole in the sidepanel and attached an aftermarket K&N type conical airfilter with a large tube ending just above the fan! The mod looked hilarious but worked pretty okey since it also drew air from the outside of the old very much closed off Fuijitsu Siemens case! :)
Flower type coolers used to be the standard until tower coolers took over. Zalman used to cool very well the CPUs at the time. I have one that is made with only copper and it is almost 1kg.
One of my old favorites was the cnps 7000A-cu, used to have a couple of them and had modded a 775 era bracket to fit 1156 which worked really well but sadly lost all that to an arson fire back in 2012.
back in those days where those heatsinks really that wicked ...thermaltake like thos disc type heatsink
Cooler Master cannot beat their own cooler - Hyper 212 Evo which was best bang for the buck people could get 5-8 years ago
More like a cheap cooling solution that simply doesn’t keep up with modern computers heat outputs
ahah true, im using a Hyper 212 led turbo (2x Fans) with an i7 10700k and it works alright
The hyper 212 evo is already 11 years old, and is not that different from the best seller that was the Hyper 212 plus (around 15 years old lol)
@@RomainCavallini yep my 212 plus cooled quad phenom 2 overclocked very well
3.5ghz phenom 2 x6 here with the 212+. Cant do more because of the 63*C limit but it could really shine with a newer cpu that can do 100*C.
The poor cooler did not even get used. I t failed it's purpose in life. Straight into the bin always to be known as a bad one. A no good money grab.
It’s gorgeous, like the old zalman coolers!
Back in the day I had a Titan Vanessa strapped to an XP1700, on an Abit NF7-S with Samsung TCCD-F ram. Absolute air cooled overclocking monster.
I remember when I first encountered heat pipes; I thought some marketing guy was pulling my leg. This is one of the few instances where there actually is a basis in physics for heat pipes. Simple, and nifty.
Good sheet
Would love a sort of analysis on it
When does the vapor chamber quit working?
Compared to a heatpipe
I think that as its a direct single heat pipe its prop quite good for low TDP parts
I have one of these that i put inside of xbox performs really good
I like the copper bottom, that's going to help to move heat quickly to the fins.
I have two of those coolers. Bought them on sale $40 Canadian a few years ago. That's about $32 US. Seen them for that twice. Works fine on a 3200G.
Actually just bought them for the looks. But would not have paid any more than that for them. Would have bought the hyper 212 if they had not gone on sale.
I had one of those old Zalman's when it first came out on my athlon 64. It worked fine and that chip made some heat. I don't suspect it would work on today's CPUs with such concentrated heat, though.
Those copper slugs are used in alot of stock cookers, famously the ryzen wraith stock coolers that came w/ the 1000 series and alot of the intel stock coolers that have a copper slug, I'd you take the Fan off either of those coolers you can see the top of the heatpipe, wich has a nipple on it from manufacturing as its not just a mashined piece of copper like many think it is
I use one of those in my i7 3770k, works ok in a small form factor dell case...
i kind of wonder what a like 3 mini ones would behave like in a tri form or 4 instead (more skinny tubes) with some extra material around them from the cold plate/bottom in place where the big one sits and how much difference that would do for this or if anything. its surely weird though but cool looking.
Like most things in microcomputing (good and bad) intel did it first - there was a Core2 Extreme cooler of a similar design
bro you deserve to have at least 1 mil subscribers
I didnt know that either, but thats why I come here:
Roman doesnt know something, so he gets out his hack saw and we both learn something.
Thats how the world should work.
I'm curious on how a custom heatpipe would perform now, similar to this having the fat heatpipe being the coldplate, but instead of it being one pipe going through the whole thing, if there's multiple connections to the fat one, all going to different areas of aluminium fins, would that help
I'd like to see a performance comparison between this cooler and other conventional circular top blower coolers and other coolers at that price point.
I always thought heat pipes were more like a pipe inside of a larger pipe with the inner pipe being wider at the bottom to catch the hot rising vapours and narrowed at the top to eject them onto the outer copper wall so they can condense as they trickle down the outer walls.
And this is how Der8auer keeps in shape....
Now it's der8auer rig everything.
Cool stuff tho.
I think that main cost source of that cooler would be this exact heat-pipe.
cool video!!
hi I have a g100m that I
polished (lapped) and it gives very good temperatures, I am currently using it with ryzen 5 2600 at 3.8ghz in the 6 cores and it is usually around 60°c rendering or doing stress tests, I have seen it reach 72°c once but room temperature was 40°c
Couldn't help but think of ways to improve such a "monolithic" heatpipe design.
Best solution I could figure is to just add some protruding walls sticking out towards the middle, then add the sintered powder. This would allow water to more easily get to the center of the "cold plate". Nor would such protruding walls need much additional work since a single weld point would be sufficient, the rest of the bonding can be done by the sintered powder, since the walls is to help water flow back, not really to conduct away heat in their own regard.
The Zalman actually came in pure copper at around 2002. :)
I'm curious about how it perform when it is mounted sideways like on the tower pc. Because my very old toshiba laptop when hot enough, it is easy to make the solvent stuck on the colder fins area when the cpu area tilted higher than the fins. Thus, instantly overheating the CPU
Luckily, when the laptop is placed on the tilted cooling pad, the CPU always lower than then heatsink area, cooling the CPU more efficiently.
I wonder if this case still happened today?
Does anyone know if vapor chambers sometime use grooves as well as sintered metal interiors?
Easy to see why this cooler would have under performed mainly due to limited surface area before the heat makes it to the fins just like the old solid blocks of the P4/Athlon era.
Would have loved to see some real world temps before the cutting...
one thing about this cooler though ,depending on the ram slot spacing and the height of the ram sticks , using this cooler may prevent some tall ram sticks from being used..also it looks like the Coruscant Senate building . XD
What is the performance like?
I hoped to see visual how this chamber works. You could have made a small hole iside and put camera in it
Old s478/s775 Intel coolers used a solid cylinder of cooper or one of these?
At least one of the newer stock AMD Wraith Spire coolers has that same thing in it and the back side of it's fan also says Cooler Master.
Can you ask GN to make a review of the G100M? I'd be interested to hear there thoughts.
Would have been interesting to test this cooler... but this a job for GN I'd say.
I like those 3D printed jaws on your vice! Great idea!
How about heat pipes connected to heat pipes cooled by a fan made of heat pipes?
That single heat pipe has a much larger volume than multiple thin heat pipes, but the overall surface appears to be lower. That's where the saturation is likely occurring, and it may be fixed by simply using a longer heat pipe with a more typical "tower" array cooler. It may then just come to a diminishing profit returns problem, as those miniature copper heat pipes use significantly less copper.
good ol inverse square law
1:02 Common cut it, cut it, cut it now!!!!
(love you Roman)
2:55 Yes, yes, yes, you never disappoint!
2:58 That bench is moving!
OMG I used that Vanessa thing. There's till pieces of its weird puffy logo decal floating around in my drawers-full-of-comptuer-junk.
a huge heat pipe is such a weird compromise, You lose out on the advantage of having surface area in exchange for a void that is not doing work.
So, how does this work if you turn it on its side? The liquid wouldn’t be touching the surface connected to the CPU.
i remember a "orb" or "duo orb" cpu cooler that used a similar single huge heat pipe I think.
we need Fact Revealing like this more in the content.
Favorite cooler of every ebay prebuild pc seller.
I wonder what the original Wraith Spire with vapor chamber looks like inside. I know they did away with that with the newer version (just a big blob of aluminum).
Probably similar. I have that one and it's just OK, it doesn't get that much lower than the cheaper cooler. The copper plate probably contributes more to better temperatures.
Considering how that cooler would be mounted on a vertical motherboard, how would that condensation even touch the coldplate? That heatpipe would be horizonal.....
Amd also had heatpipe like this in first revision of wraith spire cooler. Also made by coolermaster
Are you sure it was a heatpipe/vaper chamber design and not just a copper slug?
@@nathanpeterson5609 yes 100%
And was pretty much junk
@@christophervanzetta not really, in terms of small stock coolers it was fine
It was dry?
Cooler Master should see this as review and take notes. So, it can reform and remake that cooler. I would suggest cooler master to make it fully based copper and instead of rgb just goes for Argb. Also, to make it true low profile.
some powermac g5 coolers were like this
I think some Intel Extreme-or-something coolers were also like this.
Surprised no liquid came out. Small pipes generally do when cut in half. Also, aren't solid copper slugs probably more effective but more expensive too... At least for certain loads? I guess in this case that'd be a big ass slug lol.
Heat pipes have a thermal conductivity that is dozens of times that of solid copper (greater length = higher thermal conductivity for heat pipes). That said, a solid piece of copper isn't going to suffer from dryout, but neither will a well-engineered cooling solution using heat pipes.
I'm interested how the 25mm heatpipe fares against the roughly 17 and a third 6mm heatpipes that add up to the same cross sectional area. A=πr², ergo 6mm pipes have 113.09724 square mm while the 25mm pipe has 1963.49735 square mm, which comes to 17.36111 times as much. HOWEVER! This is a surface area based game, not a volume based one. We want perimeter instead of cross sectional area, thus 6mm pipes have 18.84965mm circumference while the 25mm can has 78.53975mm circumference. Multiplying 18.84965 by 17.36111 gets 327.24894. Dividing it by 78.53975 gets us 4.16666. My hypothetical bundle of 17 heat pipes should be roughly a fraction over four times as effective per unit length of heatpipe. Makes me wonder if a vapor chamber with a porcupine array of heatpipes would combine the best of both worlds here.
I am disappointed that you didn't do any test.
I hate to make you buy a new cooler, but if it would be possible to remove the fins, would the cooler still work at all? I would really like to see you test this.
Roman, Have a question for you, I use your thermal compound and always had this question in my mind, from where did it come from, like did you invented or some one else made the stuff with your specs ?
He grows it in his back yard, hand picked thermal pate
@@falsemcnuggethope hahahahahahahaha
Nice
Why didn't you heat it up to de-solder it mate? For a guy that does deliding everyday you should have known this..
wouldn't it be effective to have on the end of the heatpipe or vapor chamber a vacuumpump instead of cut and closed down?
with a controller board added to compensate temperature versus internal pressure.
The vacuum chamber is only a partial vacuum to lower the vaporation point of the cooling fluid. You don't want a high vacuum because it reduces condensation. It is a balance between low vaporation point and high condensation point.
@@kazedcat thanks for the background explanation! but i feel my question stands except it is now between a smaller band of operation
@@michel5148 The benefit is insignificant over the cost of active pressure control. A much cheaper way to improve cooling is increase surface area by adding more heat pipes.
I sad you gave up on dielectric fluids as a self circulating fluid that reacts to sunk heat. I getting very close but not there yet
Why would multiple smaller heatpipes be more resilient to saturation than one big pipe?
I didn't hear the vacuum chamber being filled with air or see the liquid bubbling out.
Isn't this exactly the same as the "copper core" in various Intel and AMD stock coolers?
no, tho intel did one similar model on the C2E
It would be interesting to compare the performance of that heat pipe center cooler to one with a solid copper center
2:28 nah mate that has nothing to do with vapour chambers, thats the power button.
Me seeing that power supply : Oh, remember that days
I'm pretty sure its the same design as the AMD wraith spire cooler, one big vapour chamber in the middle and a bunch of fins, in AMD's case extruded ones.
Each time I hear this accent I remember Tinger duel in "Fury" XD XD XD
I saw no liquid coming off from it. Maybe it was acetone or something like that, it may have vaporized as soon as you "unsealed" it?
yes it vaporized the moment it gets in touch with air.
@@Jeroensgambling thats the opposite of what would happen, the vacuum keeps it a vapor, with that removed the tiny amount of liquid would just condense into the sintering.
What has been soldered can always be unsoldered.
Yeah good for a cool cpu and low profile design needs but that's about it
Yo i had a huge copper 360* heat sink (Zalman? Somthing like that/) oh my old Phenom X3./ It was the shit
you can get generic chinese coolers in that style that, I think, have the pipe/chamber, for like 15-27 bucks
Мож надо было сначала температуры потестить и сравнить?