Buy a GamersNexus medium modmat here: store.gamersnexus.net/products/medium-modmat-gpu Watch our review of the Arctic Liquid Freezer II: ruclips.net/video/KPaSEGe6ML0/видео.html Or read it on our website: www.gamersnexus.net/hwreviews/3571-arctic-liquid-freezer-ii-cooler-review-benchmark Watch how liquid coolers are made: ruclips.net/video/y16azp7Wh14/видео.html
Steve, what calipers do you use? I know Mitutoyo is pretty much the standard, but I'm not finding any that look like the one you're using. I'm just looking for something tried and true and you use yours all the time. I'm asuming they work well and you haven't had any problems with them falling out of specs.
@@mikeoleksa my name is not Steve but I am a machinist for 25 years, and one of my favorite daily toys is the mitutoyo SOLAR POWERED caliper. It's at least 12 years old and still flawless. I would have gone through a garbage bag full of batteries by now. Think about that.
"That screw driver has a Corsair logo on it - that's the best thing they made with the A500" Corsair still getting burned on a completely different video
@@zaaajac Corsair aren't the first to make a product worse than Noctua, what intrigue me is on the level of wasted material and weight, it is way more spectacular of a failure than normal air cooler.
You can run the pump into a cup and measure how much it pumps over time to measure the flow. For liters per minute or whatever increment you want . We measure that way in the automotive industry.
@@GamersNexus Well, but I guess you really have to take into account the flow resistance of the cold plate and also of the radiator. (The bigger radiator will help with this, too, btw.) But I'm not sure how much this pump type is affected by this. Measuring this in a good way seems like a nightmare. Especially when you don't want to destroy the unit by placeing a flow-meter between/inside (non proper English?) the tubes. :-/ GL & TY!
@@Reaper035 It's actually pretty easy. Cut the intake tube to the pump and allow it to pull water out of a container. Measure the water the exits the radiator side of the cut tube. Yeah, it 'destroys' it, but you've got to crack a few eggs to make an omelette.
@@thegribbs Ah yea, that's right, haven't thought about cutting it at this point. (You could also add a flowmeter then there, wonder if it makes a notable difference.) Mentioned the "destroying" because GN doesn't like that, but parts would still be usable for a even more custom Custom Loop. ;)
The impeller is probably plastic with magnets molded in. That's how the stator controls it. IG the impeller were steel it would interfere with the magnetic fields from the coils.
@@zombieswitapple You can clearly see the poles in the rotor of a reluctance motor which this doesn't have. I would not use a reluctance motor in this application because of the difficulty (expense) of controllers, especially if you want different pump speeds. I recommend looking up reluctance motor rotor design if you're interested. The synchronous type rotors can get pretty fancy to lower torque ripple. Another reason to have the rotor magnets encased in plastic is that you don't have to worry about maintaining a gap between the stator core. If the plastic brushes up against it in operation, since it's nonconductive there's no issue.
@@samfedorka5629 From the looks of it it's a PM 3 phase synchronous motor. 9 stator coils. 3 per phase. Probably 12 pole rotor (magnets). @sas281gt Loads of fans have steel rotor cans...
I bought this AIO RIGHT when it came out, had it upside down, saw GNs 3950x build video with it recommending the proper config, ripped mine out and remounted immediately. This is a product I would personally tell everybody to get.
No kidding, my 980 ti bugged the piss outa me and the solution was obvious. I only mountwd it that way because the tubing lined up better but jeeze it made a lot of noise after like 4 years of running. It was always gurgling. Lol
mark tackman Same situation here. I am thinking about changing it to the "correct" configuration. But my concern is that putting the ports to the bottom increases the height different between the pump and the ports which will lead to slower water flow. Also, air bubbles are easier to be trapped. Did you notice any difference in the thermals after the correction?
Great, revealing tear-down of this CLC - I have resisted AIO/CLC in my powerful workstation builds for clients but I think it is time to test this product and to consider using these in my builds where appropriate - much thanks for all you do!
The Arctic looks like a super low restriction setup, super rare in AIOs. Particularly in the return line feeding directly into the center of the impeller - you never see that on other CLC designs, and with how dinky the pumps in these things generally are the low restriction should help a ton. Most other AIOs have the impeller facing the cold plate which means they need to draw fluid from a convoluted clusterf* of flow channels. That and the added fluid volume alone IMO should contribute to the performance.
I'm not an expert but I'm fairly certain the RGB has a lot to do with how the other brands design their pumps the way they do...Arctic avoids all that by just ditching the light show to make things simpler. I approve.
@@LastSecBloomer Not likely. The inverted pump over the cold plate has been SOP since Corsair's original CoolIT units (h100/H80/H60), well pre-RGB. 1st gen Asetek units I believe actually had a less restrictive CPU block/pump combo, but it resulted in a tall unit and Asetek at the time were using the pathetic little 3-rotor impeller that moved water like a senior citizen drooling on their cardigan. The improvement here appears to come from moving the pump off center. I assume this mandates a smaller cold plate which would mean it's less adaptable to HEDT sockets, but if Arctic's aim was to maximize performance on the mainstream platforms then that's a non-issue. Also, the pump housing has 2 tangential outlets that look pretty open. So the pressure drop before the flow hits the cold plate is minimal. And the center outlet on the cold plate is, again, directly in line with the outlet tube. Also great for low restriction. All in all I'm very impressed with the engineering that went into this unit, much more emphasis on optimizing the flow path and allowing the pump to work while fighting the other components as little as possible.
Although the cold-plate on the arctic is less dense the fins are taller. I would guess that the actual surface area is very close on both however Arctic having more room between the fins for better flow might just be the leg up on its design.
I own the same Artic Freezer ii cooler and I applied Thermal Grizzy Kryonaut thermal paste for my AMD Ryzen 2700x cpu. thanks to you Steve Burke for the recommendations.
A patent in China? 🤣 A patent is meant to protect innovation. Simply putting two items (that aren't patentable) together is not innovation and *shouldn't* be patentable. I guess Acetek has some good lawyers! This is the equivalent to patenting the location of a fan in a case.🙄
this is how every company operates you think any of them give a shit about the consumer? the ceo's aren't even majoring in their business' field half the time, they just are there for the money. it's why there won't be too many more cell phone companies too /it is difficult to get new tech onto them because the second you add anything all the other companies come with their patents suing them. microsoft also used to do that iirc, they patented other people's designs then sued them because the other people were too slow to patent or didn't want to patent it themselves
The worst part is that CoolerMaster took them to court and in the US, they upheld this bullshit patent. While a minor reason, this a reason why I avoid AIOs. I see Asetek as a patent troll that doesn't deserve my money. So, I don't give it to them.
@@bananya6020 Other companies patent things that actually deserve to be patented, though. And when they don't, like Mayhems and the word "pastel", I have as much scorn for them as I do for Asetek.
Arctic is high quality and reasonable prices for all of their products ..i have 3 Arctic fans & Arctic freezer 34 duo (cpu cooler) better than other overpriced brands
I agree with your statement 100% here. I recently completed a build with 2x 140mm for intake and 1x 120mm arctic freezer bionix f series case fans and the arctic freezer 34 esports duo for my cpu cooler. Arctic makes great cooling products and temps in my build have been great. I think a lot of people are sleeping on Arctic tbh. They are a great value product for the price.
While using this cooler in my newest build i lost a screw and they responded to my email in 24 hours and sent me a new one within the week. During covid that was pretty amazing customer service imo
Yeah, I thought the fin stack looked taller too, although whether the fins are cut deeper or just sit a bit higher is anyone's guess. Not sure which would actually result in more surface area, although I'd assume the lower density on the Arctic would be a little more clog resistant. Fingers crossed that these hold up better than enermax's coolers in the long run.
Even if it's a couple of mm higher I'm not sure in terms of surface area that will overcome the fin density. I'm not about to calculate the difference, but when your're adding height you're adding more surface area on 2 surfaces(the sides of the fins), when you add fins you're addding basically a whole new 3d space. For the increase in height you'd basically have: Surface area increase=increase in height * length of fins* number of fins For the increase in surface area by making fins denser you'd have surface area increase= [surface area of fin( area of each side including top)-difference in space between fins- difference in top surface of fin]* number of fins added. Given you're losing only a little bit of space on the top and between fins by reducing thickness and spacing, but gaining 4 additional faces you're gonna make up that number pretty quick by only adding a couple of fins. That being said, I'm wondering if the extra height is a way to keep some surface area while keeping the lower flow impedance of the looser fin density.
@@asthmaticrhino Long version of my thoughts. Would love to see flow rate testing. And impeller / motor combinations data versus loop size (total volume). We may be splitting hairs but I'm curios if it would make a difference. Insert a flow meter in line with one of the hoses, test in multiple installation positions.
i got 1 the first day they come in on amazon for my brother mounted on a Ryzen 5 2600x great resoult very quiet and very nice price ! Thx for this tear down just confirmed its a good product
Artic is such an underrated company. I once bought their F series fan as it was the only thing in stock locally and everything from the price to packaging to quality was excellent. On par with noctua quality and silence wise and way cheaper.
What's your opinion on plastic impellers build into internal water-pumps on vehicles then? My old 2003 sebring with the notorious 2.7L in it had this and was one of it's major fault points. Newer versions of it's water pump replaced it with a metal one. Such a stupid idea to put plastic in contact with such a hot fluid and it being one that's constantly moving too. Man chrysler can be dumb with design...
@@GamersNexus They call it the "Rev 2" - IMO you should do a teardown of your Rev1, as well as some basic tests on the Rev 2, to determine if the cooling performance is still the same. (I'm not accusing Arctic of pulling a bait-and-switch, but it has happened in the industry, and as GN is the channel I rely on for authoritative, thorough and honest testing, I would hope you'd be interested in doing this comparison - as I would hope Arctic is happy for you to do.)
@@iankemp2627 I'd assume if the rev2 cooler they received scored differently to their rev1 in any significant way when they benched it then it would've been mentioned.
Yea, but thats air. In an automotive application air is being compressed with an insane flow rate, whereas water is simply being moved with low pressure. And plus, the method of rotating the impellers are completely different. Honestly, an aquarium pump is much closer, though Acetec has a very specialized design.
@@jacobrzeszewski6527 Remember, air is a fluid, so, the same basic rules apply. The only thing you have to worry about in liquid vs air is cavitation, which is a similar issue to air when considering flow separation.
Great tear-down...I got the Arctic Liquid Freezer II 360mm version at launch and I am now using it on a 10700K overclocked to 5GHz all core and the temps are just exceptional. I think the radiator being thicker makes a difference and after watching the tear-down, the fin density which seems slightly lower may allow for better water flow through the cold plate...For those who are looking to buy this, definitely take the radiator thickness into account as you may struggle to fit this in some cases.
Hey Steve! So what you're seeing with the screws being so tight in the block might be another example of a mixed metal reaction. The screws and copper block are different metals and a reaction can take place between them at the thread interface or where the screw head meets the block. This happens a lot with aftermarket aluminum car wheels and their steel studs/bolts. They make an anti-seize compound that gets applied to the threads and keeps this from happening.
Shady97342 well I see the electric connection, but where is the ion transport between the copper and the aluminium? In a AIO its the Liquid and the electrical connection is the Common ground. Also i think the screws are anodized, so the ion transport is much harder/impossible.
@@herpderpson4712 I didn't mention anything about galvanic corrosion, just that it might be another example of a mixed metal reaction. I think the galling issue is seen more commonly between metals of different types due to their differing hardnesses.
I got this exact cooler & I swear by Arctic for all my cooling fans as well. I can tell you, that's quality at very reasonable pricing. And I'm not paid to say that.
Agree. In my country the Arctic P12 costs a mere 4€ and is high up there in terms of quality and performance. And you get 6 years of guaranty on top of that.
Thanks for the detailed info. I'd just picked up the Liquid Freezer II 240 version last week. First AIO for me. Makes me even happier with my choice for a first AIO.
Same here, hope I won't run into leaking, as at the moment I couldn't afford to replace my System if things go wrong. That's my only concern pretty much. I definitely have to say, that the Ryzen 9 3900x stock cooler is just not really fit for the task, even at idle it ramps up at higher room temps. Positive side effect I noticed: My GPU runs significantly cooler with the heat being expelled out of the case by the radiator.
For removing very tight screws, use an electric screwdriver. The high initial torque is much less likely to strip a screw than the wiggling and inconsistent pressure you get with a hand screwdriver. It's the same reason a breaker bar is much less likely to strip a tight bolt than a small wrench.
Hi im from Poland good work Steve !! Arctic LF2 number 1 in 2020 I think you are the first big youtuber who shows what this cooling can do, let the truth come to light, maybe it will mobilize other companies to try a bit: /
I mounted this in my NZXT 700 case, in the top based on the performance numbers. I SHOULD have re-watched this video first. I should have mounted the rad "inside" the top, and put the fans under the top. I don't have any more thermal paste, otherwise I'd take it all apart and re-do it. I am probably going to move this whole build to a new chassis, and will re-do it then. I'll be ordering ANOTHER one of these for my new Ryzen 9 5900x build - if I can find one in stock ever..... Thanks for this coverage. Really appreciate the detail, when I remember to watch out for it :D
Ordered the 360mm version, should be getting it this week. Excited to replace my evga clc 280 with the shitty rgb implementation, terrible proprietary software, and loud fans.
I just installed the 360 in my new PC. It's a fantastic AIO. You won't be disappointed. You may, however run into issues mounting if up top in your case if that's where you want it. The rad is considerably thicker than other AIOs and you could run into clearance issues...I put mine in the front of my case which is where I wanted it anyway because cooling performance is better in the front rather than the top. I also replaced the fans with Corsair ML 120 fans because I already had them. Together, the performance on this is better than any other AIO on the market.
I have had the 360mm version since launch and it is just stellar especially for the price. I have owned AIO's from the beginning going through various Corsair ones and this is without a doubt the best I have ever used though take into account of the radiator size as it is quiet a bit thicker so make sure you have the space....
Any chance I could get an update on this? I'm looking to upgrade and wanted to have a look into AIO's. This one seems like a steal but I've heard that CLC's in general are less reliable. No failures or anything to report? I'm kind of concerned since they only offer a limited 2 year warranty unlike most other AIO producers who offer 5-ish. Are you supposed to replace them when the warranty expires just to be safe or can you just leave them around without them leaking/breaking?
Awesome video, I appreciate the detailed content on this CLC. I am an air guy but the quailty build and your A+ reviews have given me a reason to switch. If my new 3700X needs more cooling, this is what will cool it instead of the Noctua I was planning to upgrade to. You guys do phenomenal work.
Thank you for this. I made a mistake and use 2x AMD screws (there are 4 in total) to mount steel bracket to pump body (should have use 2 slightly smaller screws). They are just slightly larger diameter. Screws made some "thread" in plastic (hole with small fins inside) and pushed steel nuts inside. I can hear nuts making noise when I shake it. I figured out my mistake, tried to use proper screws but it was to late, nuts are not in place any more. :) I'm using it with Intel LGA 1200 brackets, so at this point I don't need 4 AMD screws, and "wrong" screws are holding Intel bracket tight, there is about total length of thread they made of about 3/4" in plastic fins. :) Friction and difficulty to make few last turns with screwdriver was serious, I have a blister now :) So, long story short it is nice to see watching your tear-down that I need to remove just pump cable and 2 small screws to remove plastic cover not risking any leak and reposition steel nuts where they should be. Thank you!
To comment on impeller design you also need to know rpm of the motor. Smaller high rpm impellers can equal the flow rate of large ones but are usually designed for higher back pressure loops. Being a mechanical engineer and seeing the impeller comparison in this video, I immediately decided that this was the clc for me. I honestly wanted to design and open loop but couldn’t justify the $300-$400 usd of the systems after seeing this review
I had a recent thought watching penguinz0 "we did it" video, you guys are surprisingly similar in ways I would have never expected, I think I've discovered the two best youtubers
Really enjoyed this tear down. This looks like a well made product. I don't design impellers and the two look similar to me except for size. The bigger impeller almost certainly flows more for a given pressure than the smaller one. At least this is true for compressor impellers on turbochargers. When you are talking about the flow across the block at 16:16 I think you have it backwards after having a look at the impeller and housing at 23:25. The liquid will be forced to the outside of the impeller fins due to centrifugal force and will flow through the two passages on the outside of the housing and then across the block and through the hole in the middle of the housing.
Ahhh, I was just reading up on those. Been wondering if it's possible/makes sense to add another row of fans on the other side to have push-pull configuration.
If you have the space a shroud can add 5-10%, but it depends on the size of the fan hub. You get more with large hub fans like noctua nf-a12x25 and vardar, and less with small hub fans like the provided ones. On thin radiators push-pull isn't any better an twice the noise, unless your case is very restrictive and you need the extra static pressure. Lowest effort improvement is sealing up all the gaps between the fans and the radiator (and fan-fan gaps) with the kind of foam adhesive tape you use for doors & shit. Which is why dumb open frame fans like the corsair ones suck at radiator. Looks like Arctic has a small gap between the tanks and the fans - basically all radiators have this - you might gain a degree just by sealing that up with 50c of material.
I would just buy some silicone rad/fan gaskets. They seal your fan right up to the rad and provide a marked increase in static pressure through the fins, especially since many fans do not seal to the face of the rad AT ALL
I've got the 360mm on my 3950x, works great. I wonder if they will go up in price now that GN so highly of them. I assume they will sell a lot better now. One thing that made me hesitate was the lack of in depth reviews, which is now no longer the case.
@@MsChupar Clocks are a bit... hard to answer, with AMD boosting clocks and whatnot. But I will say that with PBO on, PPT 395w TDC of 225A EDC 225A, A cinebench R20 run pulls about 166.5W according to ryzen master resulting in a single run(not reaching equilibrium) of 79.5C and clock speeds of 4032. With Prime 95 V29.8 build 6 I am pullin 195w, equating to 87.8C after 15 minutes with a clock speed of 3950Mhz(ironic, yes?) Keep in mind I normally keep PBO off as I don't need the added performance and i pulls nearly 90W more power\heat, But what better way to test a cooler than push it as hard as I possibly could with my setup? Case is a Thermaltake Core V71, and the rest of the setup consits of ASUS tuf x570 WIFI, seasonic X series 1250W powersupply, Sapphire Nitro+ 5700xt, samsung 950PRO nvme, Crucial P1 NVME ssd, and a cheapo muchkin sata SSD.
@@cazandmal First of all, thanks for the brief reply! My focus will be on video editing and I found several things, very useful in the answer. One of them is the case, which supports several hdd's (I have been making a spreadsheet with hardware to buy, and the one that was there was NR600 (but it sinks in storage and limits the radiator on the top to 240mm) Another is the radiator in the case(your case), operating this CPU at very nice clocks (I don't intend to overclock, just use the maximum performance at a pleasant temperature, and yours seemed to me well under stress ...) Now my sheet will change, since your case seems to be the right choice (for me), each case is different... Future pc now: Thermaltake Core V71 GigaByte X570 Aorus Master AMD Ryzen 9 3950X + Arctic Liquid Freezer II 360mm 4 x Kingston 16GB HyperX Predator DDR4 3200MHz CL16 DIMM -HX432C16PB3 / 16 RX 5700 XT Nitro +
Corsair AX1000 1000W 80 Plus Titanium Full Modular MasterBox Corsair 1TB SSD M.2 2280 Force MP600 3D TLC NAND NVMe - CSSD-F1000GBMP600 and some of my HDD´s ... Once again thanks!
Depending on what software you intend to use, a nVidia card may benefityou. It is an area I know little about with the majority of transcoding being done with handbrake. I used to try to use Sony Vegas pro but it is not something I have a talent in.
I have a Arctic LF2 and am in automotive industry. Looking at the impeller, you can see how much longer and taller the blades are, it will move more water at same rotational speed. It being metal and looks to be coated , it will last a long time and withstand the extreme heat of the cpu's. I see too many plastic impeller waterpumps fail and fall apart in the modern car designs. I believe they are trying to save money but also maybe mass and energy required to spin it but , bits of plastic thru out your cooling system is a pain in the ass to try to flush out. Some things are better left alone. I believe the LF2 pumps are supposed to vary in rpm as well, based on heat load, so in order to test flow rates, that needs to be put into consideration. PS- I think they have good potential to improve even further if they upgraded the copper plate fin density/height.
I'm missing literally 1.5mm of clearance, preventing me to mount this at the top of my Silent Base 801 case, Imma go and cry, just 1.5mm of bullshit stopping me from getting this(not fan on front mounting)
Depending on what's in the way you could file down the shroud of the fans, or the case, or if the case has enough flex you cam force it in but that might be detrimental.
Put it in the front of the case...AIOs belong there anyway. Putting the rad up top will always yield less cooling performance because of the heat released from GPUs and their backplates. Bitwit did a in depth video on this and the numbers were astonishing. He saw temps 5-10c higher when the AIO was up top vs the front.
@@Saltssaumure well it depends on how you have your fans situated. I'm running my fans in a different way now vs the traditional way. The three fans on the rad are pushing through the rad out the front of my case. The three top fans are blowing down into the case and the rear fan is blowing in. My temps have been great for the GPU and CPU this way in a Thermaltake A500.
When I had no mixer for vodka I used the liquid from this cooler. Out of all my AIO fluids I have drank, this made me equally sick for a week. I should try the arctic silver III fluid to compare
Just a few remarks this time: 7:13 Propylene glycols (including polypropylene glycol) are odorless (so is Ethylene glycol used in cars). the smell is likely from the corrosion inhibitors, which are volatile aromatics, or a bitterant added to prevent you from drinking it. Note that since Propylene Glycol is non-toxic, they have to add biocide to it, but you don't need to do that with Ethylene glycol since it's toxic. I looked it up just now, it's pretty cool. 21:54 "checking if it's metal." Not sure why you are checking if it's metal or not with a magnet.... The rotor is definitely plastic. It contains magnets so the stator can move it. That's why it's magnetic. You will notice that when he drops the magnet on the side twice it sticks at exactly the same spot, it even moves forwards a bit the second time. Your magnet test did show that the spindle is stainless steel (not very magnetic) You can clearly see the mold marks from the ejector pins on the impeller face thanks to the zoom at 22:32. You can see the 2 marks from the gate on the back at 21:58 (play at 0.25x and freeze it as he flips it over). 24:56 "the electromagnet" is called a stator. it's the part of a motor that stays in place. The part that spins is the rotor. Otherwise excellent content as always! Keep it up!
I've had an Arctic 360 for a good while, it works great, is cheap, and came with a gross amount of fans, more than I could put in the PC. I'm very happy with it.
What I would like to see from you when you compare such coolers is the flow rate! The impeller choice is significantly less telling than actual flow rate. What pressure is used, how much water is pumped, what is the power consumption of the pump. After that you'd look at the inside and you'll probably see that the ptfe/teflon type tubes will have a better flow rate (less drag) which is another benefit to it. The higher density of copper fins is also not that telling, higher density means more surface area and more drag, so you need a better pump to actually make use of it or it will be a disadvantage.
They changed their internals completely Steve. I've seen the inside of mine and it is different half willing to take it apart again just to show you. It is cooling my gfs PC so I'll have to do it tomorrow. It is only a 120 but looks to be the same as your older 280. The VRM fans are fragile as well.
it isn't just the spacing of the fins, it is the thickness, you still need to conduct heat from the plate into the fin and then into water, so if the fin is paper thin, then it can only receive the heat from the plate through that cross section of the fin that is making contact with the plate (I know they are skived, so technically not making contact, since it is all one solid, but you get the point), which is why thicker fins actually take more heat from the plate and into the water if the flow rates are even (which they aren't), then I think the thicker fins is the reason why Arctic is performing better
Would be great if Arctic either did a complete liquid cooling system for a whole PC or an AIO for GPU's for a low price just like their current coolers.
I"m a machinist and tool designer. I can say with 100% certainty that the impeller + coil + pump housing on the Arctic Liquid Freezer II Cooler is about 5 -6 times stronger and more efficient than NZXT cooler. which means it can recirculate the liquid approximately 5 times quicker. and in the long run, can outlast the NZXT cooler. just by seeing the two designs. however, the coper plate on the Arctic cooler needs 8 screws instead of 4 to prevent leakage. but overall made really well. this pump design could destroy any competitor .
I bought the 360 version of the Arctic Liquid Freezer II 3 weeks ago. While the performance was very good I'll return my unit (probably had a bad unit): The first issue I noticed: After unboxing and moving the unit, I noticed that there was a considerable amount of air trapped within the assembly (you can literally hear the air move etc.), which is quite noisy after cold starting my PC as well (air had been settled on places, where it doesn't settle while actively running the cooler). The unit doesn't leak though. The second issue: Having screwed the mounting plates the wrong way my first time, the disassembly caused one of the internal mounting nuts to become loose. I had to disassemble the case (the 2 small ones near the pump case and the 2 larger screws holding the top cover) and glued the nut in place to be able to mount the unit on my CPU. Very rarely there is a very loud and high pitched grinding noise (plastic on plastic) coming from the pump at >= 1400 rpm. Reducing the rpm to < 1000 and going back up strangely fixes that temporary... Apart from that, there is nothing wrong with this CLC. I don't know if I'm just a unlucky guy concerning water cooling (4th time trying and failing) or there is something generally wrong with these units, but I'm going to reliably air cool my 3950X again.
@@GamersNexus I installed it the recommended way: VRM fan pointing up to the VRM heatsink, radiator on top in push config with the tubes coming out at the front part of my case. So you expected the air noise to happen in CLCs? It is my first CLC. Before it was only custom loop without any trapped air noise...
19:35 Steve " What is this? ( points finger ) that's a gummed up hole! " 19:43 Steve " That's a fill port that they jammed up with rubber!!! " Steve is doing the hard but much needed work of informing us and protecting us here on the internet.. Letting us all know that if you happen to come across a hole that is " Gummed up! " Before you jam it all up you need to use your rubber......
Yep...picked mine up from Amazon at launch on sale for £69.99 for the 360mm version!!! and it is rock solid, great cooling especially for overclocks. Though as Steve has mentioned the radiator is thicker so do take that into account with your case!
Pause @22:57 The impeller design on the right looks better designed with my understanding on fluid dynamics. The left one has holes on the base of the impeller (same base holding the fins), whereas the right one, the hole/gap is around the centre pin, but raised, so as for liquid to flow unimpeded, in one direction. The left one has to flow at an angle. Centrifugal force, blade design and coupled with the spacer/gap/hole on the right fin woul appear to be a better design choice but probably quite a significant level of difficulty to design. Would lose to see a super closeup of the centre pin and how it is held in place because it has the gap around it. Thanks!
This is why I am about to buy a bigger mod mat. I specifically am trying to avoid Acetek patents and thanks to you I can do that and get a good idea of how the device works! It is between this and the EK AIO for me.
For flow rate testing, it would be rather easy. 1. Cut the radiator off from the tubes 2. Hook to 2 different containers 3. Decide how much fluid rise you are going to have. 4. Fill one container with water 5. Run pump for a set duration. 6. Measure fluid moved. So long as fluid rise is a fixed function, then you would have a standardized test. Preferably, the lower water tank would be wide to reduce the effect of water pressure from fluid depth.
thanks. I'm curious to know if there is a performance difference by changing the fan with some even more quieter. like a standard silent wings 3 (not high speed, too many noise) even if the fan is not optimized for radiator.
Not really. The P14 is incredibly good for the price. The only fan where you could actually notice a big enough difference would be the NF-A12x25, but those are 30$/€ a pup. You could get 3 Arctic P14 PWM for that...
4 года назад
@@herpderpson4712 I just bought 5-pack of P14 PWM PST (basically the best they have) for 32€, and you can have 5-pack of cheapest P12 (120mm, no PWM and no PST and shorter warranty) for as low as 15€ here. Insane value.
I like those fans, they're a derivative of the Cooler Master Silencio fans, they're a static pressure fan that also throws a ton of airflow, while remaining reasonably quiet when against something dense like a fan filter or radiator. Great fin design.
You are doing things as an adult that I did as a child. I lost my curiosity somewhere. To tear down things and make money at the same time, what a dream.
3:08 - We're unable to read the pump speed?! So we have no monitoring in case the pump fails... deal breaker for me. (I mean I know you'll see sky high temps but still.. I like to monitor with Aida64 and get an alert if I get a failure so I have a chance to save it!)
Jason Gaunt Your pc is not gonna die from 1 time overheating man, if the pump is dead, it will just shutdown, and after changing the cooler, everything gonna be fine again
any idea if there's a way to measure the actual flow rate of water inside an aio cooler, say by opening the loop and adding in a relatively low flow resistance flow sensor into it? flow rate not only affects the final temperature of water coming off the coldplate but also the actual behaviour of heat transfer between water and fins so it would be quite relevant to know in order to compare how good diferent coldplate designs actually are
I notice you took the fans off and placed them in a pull configuration - I would need to do this to fit in my case (mount the fans externally with the rad inside the case) - did you manage to do this without needing the cable extensions?
I have the older version which came with 4 fans with a push and pull configuration for the same price. Still works today after about 4-5 years. Got 2 of them, one in a 3930k and another one i moved to a ryzen 3600
Awesome to see you call out the true thickness of the actual radiator at about 28mm... think they are getting away with a bit of "false" advertising claiming a 38mm tricking some thinking it will be a much better rad because its thicker... that said, still think its an amazing radiator, especially for the cost.
This is super informative, thank you! I'd love to see a comparison between the NZXT vs the Arctic with the fans normalized to see how much they contribute to the difference. That would be very useful for those of us who want a CLC but are willing to swap out fans (I'm planning a small build at the moment with a 240 but with the fans replaced with Noctua A12x25s).
Arcric p12's perform the same as the noctua a12x25's in both cooling and noise for 1/5 of the price, £4.99 per fan compared to £25, watch optimum techs video on it
@@danieleasthope5517 Thank you. Yeah I saw that Optimum Tech vid. He did say that the quality of the Noctua noise was a little more pleasant, I think, but the Arctics are amazing value for the money 👍🏻
Thank you for explaining the loop being on the bottom. I changed mine to the bottom months ago when you mentioned it but didn't know why. I do think my temps went up a little bit when I did that, but it's having to run by my GPU now so that could explain it. BTW this radiator make the Kraken look like a child.
Steve, have you ever thought about measuring the amount of liquid in the cooler in grams? since 1g of water = 1ml, it might be an easier/more accurate way to look at that! or use both!
If you were to take apart a water pump when working on cars like guys did when I was a teenager, you would see a very similiar design to the "new" impellers that Asetek are using now. Its a time proven design for almost 100 years now. A gas engine powered water pump used to pump water during a flood uses the same type of pump impeller design as well.
I recommend parts to quite a few people, and build PC's for people often.. I am SO glad to finally not have an Asetek option, I will definitely be pushing these into more builds I do.
![Blox Fruits Dragon Rework Update [Full Stream]](http://i.ytimg.com/vi/EqDAp8udhm0/mqdefault.jpg)
Buy a GamersNexus medium modmat here: store.gamersnexus.net/products/medium-modmat-gpu
Watch our review of the Arctic Liquid Freezer II: ruclips.net/video/KPaSEGe6ML0/видео.html
Or read it on our website: www.gamersnexus.net/hwreviews/3571-arctic-liquid-freezer-ii-cooler-review-benchmark
Watch how liquid coolers are made: ruclips.net/video/y16azp7Wh14/видео.html
What's the radiator size? You measured the nzxt but not the arctic
I have a question, why do they always put the pump at the head nearest to the cpu? cant they put the pump at the radiator or integrate it?
Test 360 version of this aio!
Steve, what calipers do you use? I know Mitutoyo is pretty much the standard, but I'm not finding any that look like the one you're using. I'm just looking for something tried and true and you use yours all the time. I'm asuming they work well and you haven't had any problems with them falling out of specs.
@@mikeoleksa my name is not Steve but I am a machinist for 25 years, and one of my favorite daily toys is the mitutoyo SOLAR POWERED caliper. It's at least 12 years old and still flawless. I would have gone through a garbage bag full of batteries by now. Think about that.
"That screw driver has a Corsair logo on it - that's the best thing they made with the A500"
Corsair still getting burned on a completely different video
21:33 well they deserved this :D
simoSLJ89 when you make a product that is worse and more expensive than Noctua you deserve to be roasted in maaany other videos xD
To be fair, noctua's l handle screw drivers are really nice.
Getting burned is what happens when your cooler sucks 😉
@@zaaajac Corsair aren't the first to make a product worse than Noctua, what intrigue me is on the level of wasted material and weight, it is way more spectacular of a failure than normal air cooler.
You can run the pump into a cup and measure how much it pumps over time to measure the flow. For liters per minute or whatever increment you want . We measure that way in the automotive industry.
Would it work if you use 2 big reservoirs and measure how much time it would take to increase the level of one in the output by X?
I like that idea. Will try it in the future. Thanks!
@@GamersNexus Well, but I guess you really have to take into account the flow resistance of the cold plate and also of the radiator. (The bigger radiator will help with this, too, btw.) But I'm not sure how much this pump type is affected by this. Measuring this in a good way seems like a nightmare. Especially when you don't want to destroy the unit by placeing a flow-meter between/inside (non proper English?) the tubes. :-/ GL & TY!
@@Reaper035 It's actually pretty easy. Cut the intake tube to the pump and allow it to pull water out of a container. Measure the water the exits the radiator side of the cut tube. Yeah, it 'destroys' it, but you've got to crack a few eggs to make an omelette.
@@thegribbs Ah yea, that's right, haven't thought about cutting it at this point. (You could also add a flowmeter then there, wonder if it makes a notable difference.) Mentioned the "destroying" because GN doesn't like that, but parts would still be usable for a even more custom Custom Loop. ;)
More teardowns, please! We'd love to see the inner workings of the cooling components and what goes into each liquid cooler!
The impeller is probably plastic with magnets molded in. That's how the stator controls it. IG the impeller were steel it would interfere with the magnetic fields from the coils.
You'd also have to worry about galvanic corrosion
It could also be steel moulded in for a reluctance motor
@@zombieswitapple You can clearly see the poles in the rotor of a reluctance motor which this doesn't have. I would not use a reluctance motor in this application because of the difficulty (expense) of controllers, especially if you want different pump speeds. I recommend looking up reluctance motor rotor design if you're interested. The synchronous type rotors can get pretty fancy to lower torque ripple.
Another reason to have the rotor magnets encased in plastic is that you don't have to worry about maintaining a gap between the stator core. If the plastic brushes up against it in operation, since it's nonconductive there's no issue.
@@samfedorka5629 From the looks of it it's a PM 3 phase synchronous motor. 9 stator coils. 3 per phase. Probably 12 pole rotor (magnets).
@sas281gt Loads of fans have steel rotor cans...
Arctic: A company so promising that it improves its products even before Steve gets a chance to complain about them!
Arctic has a lot of experience with Cooling, after all. they're not a new player.
This will be our new slogan.
hahaha
After first revision the question always asked is "what will Tech Jesus complain about?"
Only if you're like 12.
Arctic has been in the industry for quite a while...
13:47 One of the most useful things I've ever learned about mounting CLCs, thanks Tech Jesus!
Yeap
That's been said in several GN videos, yet I always see other tech channels install a front mounted AIO upside down. I internally facepalm every time.
I bought this AIO RIGHT when it came out, had it upside down, saw GNs 3950x build video with it recommending the proper config, ripped mine out and remounted immediately. This is a product I would personally tell everybody to get.
No kidding, my 980 ti bugged the piss outa me and the solution was obvious. I only mountwd it that way because the tubing lined up better but jeeze it made a lot of noise after like 4 years of running. It was always gurgling. Lol
mark tackman Same situation here. I am thinking about changing it to the "correct" configuration. But my concern is that putting the ports to the bottom increases the height different between the pump and the ports which will lead to slower water flow. Also, air bubbles are easier to be trapped. Did you notice any difference in the thermals after the correction?
Fun fact -40°C and -40°F are the same temperature so you can say just -40°. Only temp you can say that at!
Sashimi grade ;)
@@deejeh9494 lol
-32
Great, revealing tear-down of this CLC - I have resisted AIO/CLC in my powerful workstation builds for clients but I think it is time to test this product and to consider using these in my builds where appropriate - much thanks for all you do!
The Arctic looks like a super low restriction setup, super rare in AIOs. Particularly in the return line feeding directly into the center of the impeller - you never see that on other CLC designs, and with how dinky the pumps in these things generally are the low restriction should help a ton. Most other AIOs have the impeller facing the cold plate which means they need to draw fluid from a convoluted clusterf* of flow channels. That and the added fluid volume alone IMO should contribute to the performance.
I'm not an expert but I'm fairly certain the RGB has a lot to do with how the other brands design their pumps the way they do...Arctic avoids all that by just ditching the light show to make things simpler. I approve.
@@LastSecBloomer Not likely. The inverted pump over the cold plate has been SOP since Corsair's original CoolIT units (h100/H80/H60), well pre-RGB. 1st gen Asetek units I believe actually had a less restrictive CPU block/pump combo, but it resulted in a tall unit and Asetek at the time were using the pathetic little 3-rotor impeller that moved water like a senior citizen drooling on their cardigan.
The improvement here appears to come from moving the pump off center. I assume this mandates a smaller cold plate which would mean it's less adaptable to HEDT sockets, but if Arctic's aim was to maximize performance on the mainstream platforms then that's a non-issue.
Also, the pump housing has 2 tangential outlets that look pretty open. So the pressure drop before the flow hits the cold plate is minimal. And the center outlet on the cold plate is, again, directly in line with the outlet tube. Also great for low restriction. All in all I'm very impressed with the engineering that went into this unit, much more emphasis on optimizing the flow path and allowing the pump to work while fighting the other components as little as possible.
I didnt understand a single thing in that comment, damn im dumb.
Yup, I noticed EXACTLY the same thing (@22:57)..that centre of the impeller definitely could make the difference.
@@cygnusx1858 Not even the bit about senior citizens drooling?
The most expensive item in this desk is that alcohol bottle, stop flexing dude
@Florian Oprescu woah there don't get yourself killed
We got a bad ass here.
walter
Although the cold-plate on the arctic is less dense the fins are taller. I would guess that the actual surface area is very close on both however Arctic having more room between the fins for better flow might just be the leg up on its design.
It's probably cheaper too since I'm guessing the tolerance is easier to maintain with the larger/thicker fins
@@nskpsycho Yeah and most of all less machining time which is the costly part!
I own the same Artic Freezer ii cooler and I applied Thermal Grizzy Kryonaut thermal paste for my AMD Ryzen 2700x cpu. thanks to you Steve Burke for the recommendations.
A patent in China? 🤣
A patent is meant to protect innovation. Simply putting two items (that aren't patentable) together is not innovation and *shouldn't* be patentable. I guess Acetek has some good lawyers! This is the equivalent to patenting the location of a fan in a case.🙄
this is how every company operates
you think any of them give a shit about the consumer? the ceo's aren't even majoring in their business' field half the time, they just are there for the money.
it's why there won't be too many more cell phone companies too /it is difficult to get new tech onto them because the second you add anything all the other companies come with their patents suing them.
microsoft also used to do that iirc, they patented other people's designs then sued them because the other people were too slow to patent or didn't want to patent it themselves
The worst part is that CoolerMaster took them to court and in the US, they upheld this bullshit patent.
While a minor reason, this a reason why I avoid AIOs. I see Asetek as a patent troll that doesn't deserve my money. So, I don't give it to them.
@@bananya6020 Other companies patent things that actually deserve to be patented, though. And when they don't, like Mayhems and the word "pastel", I have as much scorn for them as I do for Asetek.
@@bananya6020 giant investment corporations want to know your location
@@mjc0961 Mayhems was never able to get a patent on the word "pastel" as it existed in the dictionary prior to Mayhems existing as a company.
Arctic is high quality and reasonable prices for all of their products ..i have 3 Arctic fans & Arctic freezer 34 duo (cpu cooler) better than other overpriced brands
noto moto They are like Xiaomi of PCMR, but with way better quality
@HAWK Aldo arctic have time in this cpu coolers Game they are not new brand
@HAWK Bosch is top tier though.
I agree with your statement 100% here. I recently completed a build with 2x 140mm for intake and 1x 120mm arctic freezer bionix f series case fans and the arctic freezer 34 esports duo for my cpu cooler. Arctic makes great cooling products and temps in my build have been great. I think a lot of people are sleeping on Arctic tbh. They are a great value product for the price.
While using this cooler in my newest build i lost a screw and they responded to my email in 24 hours and sent me a new one within the week. During covid that was pretty amazing customer service imo
The round cold plate has higher density, but on the square (Arctic) cold plate the fins look taller. This may be why performance is better?
Pretty sure it's down to the impellers and pump, the flow rate on this cooler is going to be far higher than the Aseteks.
Yeah, I thought the fin stack looked taller too, although whether the fins are cut deeper or just sit a bit higher is anyone's guess. Not sure which would actually result in more surface area, although I'd assume the lower density on the Arctic would be a little more clog resistant.
Fingers crossed that these hold up better than enermax's coolers in the long run.
Even if it's a couple of mm higher I'm not sure in terms of surface area that will overcome the fin density. I'm not about to calculate the difference, but when your're adding height you're adding more surface area on 2 surfaces(the sides of the fins), when you add fins you're addding basically a whole new 3d space.
For the increase in height you'd basically have:
Surface area increase=increase in height * length of fins* number of fins
For the increase in surface area by making fins denser you'd have
surface area increase= [surface area of fin( area of each side including top)-difference in space between fins- difference in top surface of fin]* number of fins added. Given you're losing only a little bit of space on the top and between fins by reducing thickness and spacing, but gaining 4 additional faces you're gonna make up that number pretty quick by only adding a couple of fins.
That being said, I'm wondering if the extra height is a way to keep some surface area while keeping the lower flow impedance of the looser fin density.
@@asthmaticrhino Long version of my thoughts. Would love to see flow rate testing. And impeller / motor combinations data versus loop size (total volume). We may be splitting hairs but I'm curios if it would make a difference. Insert a flow meter in line with one of the hoses, test in multiple installation positions.
@@iankemp2627 The Freezer II has a thicker radiator.
i got 1 the first day they come in on amazon for my brother mounted on a Ryzen 5 2600x great resoult very quiet and very nice price !
Thx for this tear down just confirmed its a good product
Never clicked on a GN Video so fast!
Artic is such an underrated company. I once bought their F series fan as it was the only thing in stock locally and everything from the price to packaging to quality was excellent. On par with noctua quality and silence wise and way cheaper.
Impeller designer here, bigger is better. No need to thank me.
Dat Skyline pfp doe. 👌
What's your opinion on plastic impellers build into internal water-pumps on vehicles then? My old 2003 sebring with the notorious 2.7L in it had this and was one of it's major fault points. Newer versions of it's water pump replaced it with a metal one. Such a stupid idea to put plastic in contact with such a hot fluid and it being one that's constantly moving too. Man chrysler can be dumb with design...
"It is different, but not a big deal"
Western Digital has entered the chat.
shingling intensifies
5400rpm intensifies
Overvolt, now 6500 RPM
Have you ever thought of doing fan teardowns? It would be interesting to see the bearing mechanism and quality of the motor.
Yeah they already came out with a V2 as listed on their website I think I got the V2.
Pretty sure that's the one we took apart.
@@GamersNexus They call it the "Rev 2" - IMO you should do a teardown of your Rev1, as well as some basic tests on the Rev 2, to determine if the cooling performance is still the same. (I'm not accusing Arctic of pulling a bait-and-switch, but it has happened in the industry, and as GN is the channel I rely on for authoritative, thorough and honest testing, I would hope you'd be interested in doing this comparison - as I would hope Arctic is happy for you to do.)
Of course it's the same we're talking about. "II" stands for V2..
@@iankemp2627 I'd assume if the rev2 cooler they received scored differently to their rev1 in any significant way when they benched it then it would've been mentioned.
@simoSLJ89 There is a Liquid Freezer II and there is also Liquid Freezer II v2
You can reach out to automotive turbo manufactures such as Comp Turbo, Precision, etc. for an understanding on impeller theory and design.
Yea, but thats air. In an automotive application air is being compressed with an insane flow rate, whereas water is simply being moved with low pressure. And plus, the method of rotating the impellers are completely different. Honestly, an aquarium pump is much closer, though Acetec has a very specialized design.
@@jacobrzeszewski6527 Remember, air is a fluid, so, the same basic rules apply. The only thing you have to worry about in liquid vs air is cavitation, which is a similar issue to air when considering flow separation.
Great tear-down...I got the Arctic Liquid Freezer II 360mm version at launch and I am now using it on a 10700K overclocked to 5GHz all core and the temps are just exceptional. I think the radiator being thicker makes a difference and after watching the tear-down, the fin density which seems slightly lower may allow for better water flow through the cold plate...For those who are looking to buy this, definitely take the radiator thickness into account as you may struggle to fit this in some cases.
Now we want to see you put it all back together again!! :P
the only reason i came here was to see how reassembly would go.. but he didnt mention it :(
Another sublime explanation! Thank you Tech Jesus!
Hey Steve! So what you're seeing with the screws being so tight in the block might be another example of a mixed metal reaction. The screws and copper block are different metals and a reaction can take place between them at the thread interface or where the screw head meets the block. This happens a lot with aftermarket aluminum car wheels and their steel studs/bolts. They make an anti-seize compound that gets applied to the threads and keeps this from happening.
Shady97342 well I see the electric connection, but where is the ion transport between the copper and the aluminium? In a AIO its the Liquid and the electrical connection is the Common ground. Also i think the screws are anodized, so the ion transport is much harder/impossible.
@@herpderpson4712Google "thread galling"
Shady97342 interesting phenomenon. Though that’s more abrasion and not galvanic corrosion.
@@herpderpson4712 I didn't mention anything about galvanic corrosion, just that it might be another example of a mixed metal reaction. I think the galling issue is seen more commonly between metals of different types due to their differing hardnesses.
Toujours appréciable de voir ce genre de choses pompes d'AIO ou cartes graphiques démontées !!
Merci pour cette vidéo très intéressante !
Thank you: I learned some interesting things! I really enjoy your in depth reviews.
Nice work man! I have been looking for a week now at reviews for a nice liquid and i think this one is the best for me!
That Arctic cooler just looks better built overall. Definitely higher quality to the motor.
I got this exact cooler & I swear by Arctic for all my cooling fans as well. I can tell you, that's quality at very reasonable pricing. And I'm not paid to say that.
@@erich3784 Their customer support is the best i ever experienced
I have an arctic fan I got for my pc back in junior high, oh i dunno, 11 years ago
Is currently still ziptied to to my case for VRM cooling .
Agree. In my country the Arctic P12 costs a mere 4€ and is high up there in terms of quality and performance. And you get 6 years of guaranty on top of that.
Erich Can you speak for the noise level of the fans?
1:45 "case by case basis" referring to actual cases. Mindblown.
Thanks for the detailed info. I'd just picked up the Liquid Freezer II 240 version last week. First AIO for me. Makes me even happier with my choice for a first AIO.
Where
thank you Steve for all this good videos ! very good info !
Absolutely amazing content. Thank you for all the deep-dive videos and info for PC builds and general knowledge.
Using this in my 3900x!! It’s amazing
Same here, hope I won't run into leaking, as at the moment I couldn't afford to replace my System if things go wrong.
That's my only concern pretty much.
I definitely have to say, that the Ryzen 9 3900x stock cooler is just not really fit for the task, even at idle it ramps up at higher room temps.
Positive side effect I noticed:
My GPU runs significantly cooler with the heat being expelled out of the case by the radiator.
For removing very tight screws, use an electric screwdriver. The high initial torque is much less likely to strip a screw than the wiggling and inconsistent pressure you get with a hand screwdriver. It's the same reason a breaker bar is much less likely to strip a tight bolt than a small wrench.
Hi im from Poland good work Steve !! Arctic LF2 number 1 in 2020 I think you are the first big youtuber who shows what this cooling can do, let the truth come to light, maybe it will mobilize other companies to try a bit: /
I mounted this in my NZXT 700 case, in the top based on the performance numbers. I SHOULD have re-watched this video first. I should have mounted the rad "inside" the top, and put the fans under the top. I don't have any more thermal paste, otherwise I'd take it all apart and re-do it. I am probably going to move this whole build to a new chassis, and will re-do it then. I'll be ordering ANOTHER one of these for my new Ryzen 9 5900x build - if I can find one in stock ever.....
Thanks for this coverage. Really appreciate the detail, when I remember to watch out for it :D
Ordered the 360mm version, should be getting it this week. Excited to replace my evga clc 280 with the shitty rgb implementation, terrible proprietary software, and loud fans.
I'm looking into getting the 360mm version as well. Can you please let me know how you like it?
I just installed the 360 in my new PC. It's a fantastic AIO. You won't be disappointed. You may, however run into issues mounting if up top in your case if that's where you want it. The rad is considerably thicker than other AIOs and you could run into clearance issues...I put mine in the front of my case which is where I wanted it anyway because cooling performance is better in the front rather than the top. I also replaced the fans with Corsair ML 120 fans because I already had them. Together, the performance on this is better than any other AIO on the market.
@@GSP-76 I actually have a thermaltake core p3 so I'm all good! :P
I have had the 360mm version since launch and it is just stellar especially for the price. I have owned AIO's from the beginning going through various Corsair ones and this is without a doubt the best I have ever used though take into account of the radiator size as it is quiet a bit thicker so make sure you have the space....
Any chance I could get an update on this? I'm looking to upgrade and wanted to have a look into AIO's. This one seems like a steal but I've heard that CLC's in general are less reliable. No failures or anything to report? I'm kind of concerned since they only offer a limited 2 year warranty unlike most other AIO producers who offer 5-ish. Are you supposed to replace them when the warranty expires just to be safe or can you just leave them around without them leaking/breaking?
Awesome video, I appreciate the detailed content on this CLC. I am an air guy but the quailty build and your A+ reviews have given me a reason to switch. If my new 3700X needs more cooling, this is what will cool it instead of the Noctua I was planning to upgrade to. You guys do phenomenal work.
Thank you for this. I made a mistake and use 2x AMD screws (there are 4 in total) to mount steel bracket to pump body (should have use 2 slightly smaller screws). They are just slightly larger diameter. Screws made some "thread" in plastic (hole with small fins inside) and pushed steel nuts inside. I can hear nuts making noise when I shake it. I figured out my mistake, tried to use proper screws but it was to late, nuts are not in place any more. :) I'm using it with Intel LGA 1200 brackets, so at this point I don't need 4 AMD screws, and "wrong" screws are holding Intel bracket tight, there is about total length of thread they made of about 3/4" in plastic fins. :) Friction and difficulty to make few last turns with screwdriver was serious, I have a blister now :) So, long story short it is nice to see watching your tear-down that I need to remove just pump cable and 2 small screws to remove plastic cover not risking any leak and reposition steel nuts where they should be. Thank you!
I am impressed. I think I found my next cooler.
To comment on impeller design you also need to know rpm of the motor. Smaller high rpm impellers can equal the flow rate of large ones but are usually designed for higher back pressure loops. Being a mechanical engineer and seeing the impeller comparison in this video, I immediately decided that this was the clc for me. I honestly wanted to design and open loop but couldn’t justify the $300-$400 usd of the systems after seeing this review
I had a recent thought watching penguinz0 "we did it" video, you guys are surprisingly similar in ways I would have never expected, I think I've discovered the two best youtubers
Well rewatched it and nevermind, I guess I heard him mention his charity stuff and love for metal in a later stream
@@teddygoboom1 penguinz0 is an annoying douche, Steve isn't.
@@leeadkins1360 Have you even watched him? Lol he's the complete antithesis of a douche
Lee Adkins why? I kinda like the guy’s attitude and I’m allergic to douches :)
Ok I'm gonna take this one back, checking another one of his vids he aint so bad :P
Thanks! As always - TOP quality content!
Really enjoyed this tear down. This looks like a well made product. I don't design impellers and the two look similar to me except for size. The bigger impeller almost certainly flows more for a given pressure than the smaller one. At least this is true for compressor impellers on turbochargers.
When you are talking about the flow across the block at 16:16 I think you have it backwards after having a look at the impeller and housing at 23:25. The liquid will be forced to the outside of the impeller fins due to centrifugal force and will flow through the two passages on the outside of the housing and then across the block and through the hole in the middle of the housing.
Awesome content. Love seeing the process and hearing about dinner if the high level details. Thanks for the great video!
Ahhh, I was just reading up on those. Been wondering if it's possible/makes sense to add another row of fans on the other side to have push-pull configuration.
Push/pull is typically not really worth it, but we haven't tested it on this one.
I was thinking about setting mine up. Will have to check if I can fit a push pull config in my case though since this is already so beefy.
If you have the space a shroud can add 5-10%, but it depends on the size of the fan hub. You get more with large hub fans like noctua nf-a12x25 and vardar, and less with small hub fans like the provided ones. On thin radiators push-pull isn't any better an twice the noise, unless your case is very restrictive and you need the extra static pressure.
Lowest effort improvement is sealing up all the gaps between the fans and the radiator (and fan-fan gaps) with the kind of foam adhesive tape you use for doors & shit. Which is why dumb open frame fans like the corsair ones suck at radiator.
Looks like Arctic has a small gap between the tanks and the fans - basically all radiators have this - you might gain a degree just by sealing that up with 50c of material.
On a 38mm rad I would say no. It makes sense on thicker rads.
I would just buy some silicone rad/fan gaskets. They seal your fan right up to the rad and provide a marked increase in static pressure through the fins, especially since many fans do not seal to the face of the rad AT ALL
just bought my 360 :D im soo excited :D love you steve
How is it? Wanting to get the 360 as well
@@completionofanexorcism1195 Its the best aio i ever had👍👍
@@florijanftw Good to know! thanks!
I've got the 360mm on my 3950x, works great. I wonder if they will go up in price now that GN so highly of them. I assume they will sell a lot better now. One thing that made me hesitate was the lack of in depth reviews, which is now no longer the case.
Temps/clock and case?
@@MsChupar Clocks are a bit... hard to answer, with AMD boosting clocks and whatnot. But I will say that with PBO on, PPT 395w TDC of 225A EDC 225A, A cinebench R20 run pulls about 166.5W according to ryzen master resulting in a single run(not reaching equilibrium) of 79.5C and clock speeds of 4032. With Prime 95 V29.8 build 6 I am pullin 195w, equating to 87.8C after 15 minutes with a clock speed of 3950Mhz(ironic, yes?) Keep in mind I normally keep PBO off as I don't need the added performance and i pulls nearly 90W more power\heat, But what better way to test a cooler than push it as hard as I possibly could with my setup?
Case is a Thermaltake Core V71, and the rest of the setup consits of ASUS tuf x570 WIFI, seasonic X series 1250W powersupply, Sapphire Nitro+ 5700xt, samsung 950PRO nvme, Crucial P1 NVME ssd, and a cheapo muchkin sata SSD.
@@cazandmal First of all, thanks for the brief reply!
My focus will be on video editing and I found several things, very useful in the answer. One of them is the case, which supports several hdd's (I have been making a spreadsheet with hardware to buy, and the one that was there was NR600 (but it sinks in storage and limits the radiator on the top to 240mm)
Another is the radiator in the case(your case), operating this CPU at very nice clocks (I don't intend to overclock, just use the maximum performance at a pleasant temperature, and yours seemed to me well under stress ...)
Now my sheet will change, since your case seems to be the right choice (for me), each case is different...
Future pc now:
Thermaltake Core V71
GigaByte X570 Aorus Master
AMD Ryzen 9 3950X +
Arctic Liquid Freezer II 360mm
4 x Kingston 16GB HyperX Predator DDR4 3200MHz CL16 DIMM -HX432C16PB3 / 16
RX 5700 XT Nitro +
Corsair AX1000 1000W 80 Plus Titanium Full Modular
MasterBox
Corsair 1TB SSD M.2 2280 Force MP600 3D TLC NAND NVMe - CSSD-F1000GBMP600
and some of my HDD´s ...
Once again thanks!
Depending on what software you intend to use, a nVidia card may benefityou. It is an area I know little about with the majority of transcoding being done with handbrake. I used to try to use Sony Vegas pro but it is not something I have a talent in.
I have a Arctic LF2 and am in automotive industry. Looking at the impeller, you can see how much longer and taller the blades are, it will move more water at same rotational speed. It being metal and looks to be coated , it will last a long time and withstand the extreme heat of the cpu's.
I see too many plastic impeller waterpumps fail and fall apart in the modern car designs. I believe they are trying to save money but also maybe mass and energy required to spin it but , bits of plastic thru out your cooling system is a pain in the ass to try to flush out. Some things are better left alone.
I believe the LF2 pumps are supposed to vary in rpm as well, based on heat load, so in order to test flow rates, that needs to be put into consideration.
PS- I think they have good potential to improve even further if they upgraded the copper plate fin density/height.
Steve, are the fittings removable so that a semi custom loop can be built from one?
I'm missing literally 1.5mm of clearance, preventing me to mount this at the top of my Silent Base 801 case, Imma go and cry, just 1.5mm of bullshit stopping me from getting this(not fan on front mounting)
or mount the fans on the outside of the case :D :D
Depending on what's in the way you could file down the shroud of the fans, or the case, or if the case has enough flex you cam force it in but that might be detrimental.
Put it in the front of the case...AIOs belong there anyway. Putting the rad up top will always yield less cooling performance because of the heat released from GPUs and their backplates. Bitwit did a in depth video on this and the numbers were astonishing. He saw temps 5-10c higher when the AIO was up top vs the front.
@@GSP-76 It's a tradeoff - front mounted rads achieve better CPU temps, but hurts the GPU temps.
@@Saltssaumure well it depends on how you have your fans situated. I'm running my fans in a different way now vs the traditional way. The three fans on the rad are pushing through the rad out the front of my case. The three top fans are blowing down into the case and the rear fan is blowing in. My temps have been great for the GPU and CPU this way in a Thermaltake A500.
When I had no mixer for vodka I used the liquid from this cooler. Out of all my AIO fluids I have drank, this made me equally sick for a week.
I should try the arctic silver III fluid to compare
Great video. Thank you
Just a few remarks this time:
7:13 Propylene glycols (including polypropylene glycol) are odorless (so is Ethylene glycol used in cars). the smell is likely from the corrosion inhibitors, which are volatile aromatics, or a bitterant added to prevent you from drinking it. Note that since Propylene Glycol is non-toxic, they have to add biocide to it, but you don't need to do that with Ethylene glycol since it's toxic. I looked it up just now, it's pretty cool.
21:54 "checking if it's metal." Not sure why you are checking if it's metal or not with a magnet.... The rotor is definitely plastic. It contains magnets so the stator can move it. That's why it's magnetic. You will notice that when he drops the magnet on the side twice it sticks at exactly the same spot, it even moves forwards a bit the second time. Your magnet test did show that the spindle is stainless steel (not very magnetic) You can clearly see the mold marks from the ejector pins on the impeller face thanks to the zoom at 22:32. You can see the 2 marks from the gate on the back at 21:58 (play at 0.25x and freeze it as he flips it over).
24:56 "the electromagnet" is called a stator. it's the part of a motor that stays in place. The part that spins is the rotor.
Otherwise excellent content as always! Keep it up!
I've had an Arctic 360 for a good while, it works great, is cheap, and came with a gross amount of fans, more than I could put in the PC. I'm very happy with it.
This was really cool. Love teardowns like this.
What I would like to see from you when you compare such coolers is the flow rate!
The impeller choice is significantly less telling than actual flow rate. What pressure is used, how much water is pumped, what is the power consumption of the pump.
After that you'd look at the inside and you'll probably see that the ptfe/teflon type tubes will have a better flow rate (less drag) which is another benefit to it.
The higher density of copper fins is also not that telling, higher density means more surface area and more drag, so you need a better pump to actually make use of it or it will be a disadvantage.
"I can smell this video."
- Anyone who was water cooling back in the days when propylene glycol mixes were the norm
Cool, I'm going to buy an Arctic Liquid Freezer II, thanks.
They changed their internals completely Steve. I've seen the inside of mine and it is different half willing to take it apart again just to show you. It is cooling my gfs PC so I'll have to do it tomorrow. It is only a 120 but looks to be the same as your older 280. The VRM fans are fragile as well.
it isn't just the spacing of the fins, it is the thickness, you still need to conduct heat from the plate into the fin and then into water, so if the fin is paper thin, then it can only receive the heat from the plate through that cross section of the fin that is making contact with the plate (I know they are skived, so technically not making contact, since it is all one solid, but you get the point), which is why thicker fins actually take more heat from the plate and into the water
if the flow rates are even (which they aren't), then I think the thicker fins is the reason why Arctic is performing better
Why do you not measure height of micro fins? It is important for surface area comparison.
Would be great if Arctic either did a complete liquid cooling system for a whole PC or an AIO for GPU's for a low price just like their current coolers.
Just build a loop, much more fun :)
Sadly Arctic completely missed the opportunity to fix their mounting hardware with the Rev 2 :(
What's the rev 2 changes?
I had no issues with mounting hardware.
I"m a machinist and tool designer. I can say with 100% certainty that the impeller + coil + pump housing on the Arctic Liquid Freezer II Cooler is about 5 -6 times stronger and more efficient than NZXT cooler. which means it can recirculate the liquid approximately 5 times quicker. and in the long run, can outlast the NZXT cooler. just by seeing the two designs.
however, the coper plate on the Arctic cooler needs 8 screws instead of 4 to prevent leakage. but overall made really well.
this pump design could destroy any competitor .
I will let you know if mine ever leaks XD
@@dighawaii1 As long as you like the look of the cooler. you should be OK for a long time. I just ordered a NZXT KRAKEN X63
What if I wanted a mat with copper power and stuff
You'd have to buy one of the charity auction ones! Maybe you could harvest the copper powder and sell it.
@@GamersNexus maybe. Keep the good work up man. I hope the community appricaties the power of tech Jesus
I bought the 360 version of the Arctic Liquid Freezer II 3 weeks ago. While the performance was very good I'll return my unit (probably had a bad unit):
The first issue I noticed: After unboxing and moving the unit, I noticed that there was a considerable amount of air trapped within the assembly (you can literally hear the air move etc.), which is quite noisy after cold starting my PC as well (air had been settled on places, where it doesn't settle while actively running the cooler). The unit doesn't leak though.
The second issue: Having screwed the mounting plates the wrong way my first time, the disassembly caused one of the internal mounting nuts to become loose. I had to disassemble the case (the 2 small ones near the pump case and the 2 larger screws holding the top cover) and glued the nut in place to be able to mount the unit on my CPU.
Very rarely there is a very loud and high pitched grinding noise (plastic on plastic) coming from the pump at >= 1400 rpm. Reducing the rpm to < 1000 and going back up strangely fixes that temporary...
Apart from that, there is nothing wrong with this CLC.
I don't know if I'm just a unlucky guy concerning water cooling (4th time trying and failing) or there is something generally wrong with these units, but I'm going to reliably air cool my 3950X again.
Did you install with the tubes up? That's a common mistake and causes a lot of those air-related noise issues.
@@GamersNexus I installed it the recommended way: VRM fan pointing up to the VRM heatsink, radiator on top in push config with the tubes coming out at the front part of my case.
So you expected the air noise to happen in CLCs? It is my first CLC. Before it was only custom loop without any trapped air noise...
19:35 Steve " What is this? ( points finger ) that's a gummed up hole! "
19:43 Steve " That's a fill port that they jammed up with rubber!!! "
Steve is doing the hard but much needed work of informing us and protecting us here on the internet.. Letting us all know that if you happen to come across a hole that is " Gummed up! " Before you jam it all up you need to use your rubber......
Awesome video. Exactly what I was hoping for after the review video. Now, where is my MetallicGear case review video? lol
Tech Jesus out here doin’ God’s work! ❤️
360mm version of this is cheaper than most other brands' 240mm AIOs -_-
Yep...picked mine up from Amazon at launch on sale for £69.99 for the 360mm version!!! and it is rock solid, great cooling especially for overclocks. Though as Steve has mentioned the radiator is thicker so do take that into account with your case!
the 420 model just released and its still cheaper than most other companies 360 models lol
Pause @22:57
The impeller design on the right looks better designed with my understanding on fluid dynamics.
The left one has holes on the base of the impeller (same base holding the fins), whereas the right one, the hole/gap is around the centre pin, but raised, so as for liquid to flow unimpeded, in one direction.
The left one has to flow at an angle.
Centrifugal force, blade design and coupled with the spacer/gap/hole on the right fin woul appear to be a better design choice but probably quite a significant level of difficulty to design.
Would lose to see a super closeup of the centre pin and how it is held in place because it has the gap around it.
Thanks!
This is why I am about to buy a bigger mod mat. I specifically am trying to avoid Acetek patents and thanks to you I can do that and get a good idea of how the device works! It is between this and the EK AIO for me.
dud what is your background?? you know a LOOOT of behind the scenes of industry!! that's amazing
Good video guys 🔥👍🔥👍
Bought one cause I see it's serviceable! eventually, I live in the tropics and I'll probably will need to do maintenance in a few years :)
How do the tubes pull off please? I want to be able to lenghten them. Thnks.
I wish to see Helor 360 Cougar open like that, it will be awesome Steve ;)
For flow rate testing, it would be rather easy.
1. Cut the radiator off from the tubes
2. Hook to 2 different containers
3. Decide how much fluid rise you are going to have.
4. Fill one container with water
5. Run pump for a set duration.
6. Measure fluid moved.
So long as fluid rise is a fixed function, then you would have a standardized test.
Preferably, the lower water tank would be wide to reduce the effect of water pressure from fluid depth.
Just got this one today(240 rev.3), first time on liquid cooling, damn tho, 5800x is a beast to cool.
thanks.
I'm curious to know if there is a performance difference by changing the fan with some even more quieter.
like a standard silent wings 3 (not high speed, too many noise)
even if the fan is not optimized for radiator.
Not really. The P14 is incredibly good for the price. The only fan where you could actually notice a big enough difference would be the NF-A12x25, but those are 30$/€ a pup. You could get 3 Arctic P14 PWM for that...
@@herpderpson4712 I just bought 5-pack of P14 PWM PST (basically the best they have) for 32€, and you can have 5-pack of cheapest P12 (120mm, no PWM and no PST and shorter warranty) for as low as 15€ here. Insane value.
I like those fans, they're a derivative of the Cooler Master Silencio fans, they're a static pressure fan that also throws a ton of airflow, while remaining reasonably quiet when against something dense like a fan filter or radiator. Great fin design.
You are doing things as an adult that I did as a child. I lost my curiosity somewhere. To tear down things and make money at the same time, what a dream.
14:28 - A professional Jay move right there. :D (No offense, both of you are awesome!)
Also Arctic's tubes are rubberized or teflon lined?
3:08 - We're unable to read the pump speed?! So we have no monitoring in case the pump fails... deal breaker for me.
(I mean I know you'll see sky high temps but still.. I like to monitor with Aida64 and get an alert if I get a failure so I have a chance to save it!)
Jason Gaunt Your pc is not gonna die from 1 time overheating man, if the pump is dead, it will just shutdown, and after changing the cooler, everything gonna be fine again
any idea if there's a way to measure the actual flow rate of water inside an aio cooler, say by opening the loop and adding in a relatively low flow resistance flow sensor into it?
flow rate not only affects the final temperature of water coming off the coldplate but also the actual behaviour of heat transfer between water and fins so it would be quite relevant to know in order to compare how good diferent coldplate designs actually are
I notice you took the fans off and placed them in a pull configuration - I would need to do this to fit in my case (mount the fans externally with the rad inside the case) - did you manage to do this without needing the cable extensions?
I have the older version which came with 4 fans with a push and pull configuration for the same price.
Still works today after about 4-5 years. Got 2 of them, one in a 3930k and another one i moved to a ryzen 3600
Awesome to see you call out the true thickness of the actual radiator at about 28mm... think they are getting away with a bit of "false" advertising claiming a 38mm tricking some thinking it will be a much better rad because its thicker... that said, still think its an amazing radiator, especially for the cost.
This is super informative, thank you! I'd love to see a comparison between the NZXT vs the Arctic with the fans normalized to see how much they contribute to the difference. That would be very useful for those of us who want a CLC but are willing to swap out fans (I'm planning a small build at the moment with a 240 but with the fans replaced with Noctua A12x25s).
Arcric p12's perform the same as the noctua a12x25's in both cooling and noise for 1/5 of the price, £4.99 per fan compared to £25, watch optimum techs video on it
@@danieleasthope5517 Thank you. Yeah I saw that Optimum Tech vid. He did say that the quality of the Noctua noise was a little more pleasant, I think, but the Arctics are amazing value for the money 👍🏻
4:00 my AIO will only fit tubes up into my P400, and a large air cooler wont fit.. oof
There's plenty top tier large air coolers that are
Thank you for explaining the loop being on the bottom. I changed mine to the bottom months ago when you mentioned it but didn't know why. I do think my temps went up a little bit when I did that, but it's having to run by my GPU now so that could explain it. BTW this radiator make the Kraken look like a child.
Steve, have you ever thought about measuring the amount of liquid in the cooler in grams? since 1g of water = 1ml, it might be an easier/more accurate way to look at that! or use both!
If you were to take apart a water pump when working on cars like guys did when I was a teenager, you would see a very similiar design to the "new" impellers that Asetek are using now. Its a time proven design for almost 100 years now. A gas engine powered water pump used to pump water during a flood uses the same type of pump impeller design as well.
I recommend parts to quite a few people, and build PC's for people often.. I am SO glad to finally not have an Asetek option, I will definitely be pushing these into more builds I do.