• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

ENERMAX LIQTECH TR4 AIO Liquid CPU Coolers Review

FrgMstr

FrgMstr

Just Plain Mean
Staff member
2FA
Joined
May 18, 1997
Messages
56,775
ENERMAX LIQTECH TR4 AIO Liquid CPU Coolers Review

Enermax brings to us the first All-in-One coolers that are purpose-built for AMD's Ryzen Threadripper CPUs. We review both the Liqtech TR4 240 and the Liqtech TR4 360 using our overclocked 1950X Threadripper system and compare these to our XSPC RayStorm custom cooling loop. Yes, we are setting the bar high.
 
Very impressive value for the performance when dealing with the TR4 monster.
 
Enermax the Liqtech TR4 pump is rated at 450 L/h .. which is more than 10 times the coolant flow Asetek rated their 550LC pump, which as far as I know is same basic pump used in their other CLCs. Liqtech TR4 pump is rated 0.4amp .. Asetek pumps are rated 2.1 - 3.1watts (0.175-0.258 amp) .. again a fraction of what Liqtech TR4 pump is rated.

I suspect this increased flowrate is significant in Liqtech TR4's cooling ability. ;)
 
Thanks for noting specifically that the radiator is aluminum. No mention of 2 year warranty period when comparing AIOs? Surely this should be mentioned? Is everything 2 years?

Certainly seems to offer some value considering its price.
 
Nicepants42 said:
Typo last sentence

Thanks for noting specifically that the radiator is aluminum. No mention of 2 year warranty period when comparing AIOs? Surely this should be mentioned? Is everything 2 years?

Certainly seems to offer some value considering its price.
Click to expand...

yeah pretty much everything is 1 or 2 years these days in the US. warrenty period for computer hardware has been slowly shrinking over the last 5-8 years.. use to be lifetime for the premium companies, then 5 years then 3 now it's 1-2.

other note, 130/150 seems like an insanely good price for these things.. was seriously expecting it to be in the 160-180 range given how niche of a market they're designed for i figured they'd try to get every penny they could out of them.
 
A most useful article. I hope that Enermax will also do 140mm and 180mm variants.
 
Been using the 360mm model for over a month now. Absolutely love it. Quiet and effective with my 1950x OC'ed to 4ghz.
 
Balthazar2k4 said:
Been using the 360mm model for over a month now. Absolutely love it. Quiet and effective with my 1950x OC'ed to 4ghz.
Click to expand...
What vCore and vSoc are you running? Will yours run over an hour of Prime95 at 4GHz? Ambient intake temperature? I assume you are using a fan profile based on CPU die temp?
 
Good job Kyle! A review that makes sense for a change, after the utter nonsense the HardwareCanucks produced recently:

 
Just snagged the 280mm one for my Threadripper build that I am putting together this weekend.
 
I think the pump on my Enermax TR4 360 is dead/dying. Had it for about 5 months now - I've never dealt with a bad AIO before but one of the tubes was cool to the touch and half of the radiator wasn't even warm when maxing out the CPU. Just opening up a web browser would see temps skyrocket in the 60C range so it was throttling big time too. Guess I will have to RMA...

Reseating did not fix so wasn't a thermal paste issue. I went back to my Noctua TR4 NH-U14 and temps are back to ~30 C idle and 40s during light load.
 
If only outflow from block is hot and inflow is cold rather than warm pump is not circulating.

NH-U14S is a very good cooler.
 
It's been 6 months now since I built this rig and I am now noticing a jump in temps. My idle was 35-37c and now it's 43-47c. And my load temps that usually stay under 80c recently have been hitting the mid 80s. Dammit I hope it's not the pump.

Edit: Having a similar experience as x3sphere one tube is cool to the touch the other is warm (27c and 40-43c respectively). Not sure what to do if I have to RMA for warranty service, I won't be able to use the PC since I have no spare to mount on here while it gets fixed and all of my software licenses has been switched over to this PC now uggh. :( Now I really wish AMD shipped the TR4 with some kind of default cooling solution included even if I had to run stock settings while I got it fixed.

Edit 2: It seems it isn't the pump. :) I went to re-seat the AIO and to my surprise it popped right out. no vacuum like seal or anything. When I looked it was as if there was no TIM in the center of the heat sink. :eek: I mean I could easily read the top of the threadripper chip. It looked as if the TIM flowed away from the center. Not sure why it did that. Well either way problem is fixed. Idling at 35-36c as I type this.
 
Last edited:
I think the MX4 I am using (20G tube) might be defective. Since my last post i have had to re apply the TIM twice now. It seems every 2-3 weeks it's like the paste moved away from where it's needed. Today I was idling at 52c-55c that's why I had to reapply it. Now I am back to idling at 42c. So far the small MX-4 I had bought with my threadripper lasted 6 months after I put it in. But when I bought a new larger tube of MX4 it seems that it's not working so great. Not sure what to make of it but it is really getting annoying now. Anyone else ever had an issue similar to this with their TIM?
 
Ok I think the pump is bad now. I am now with my AC running and only a very low load (almost idle) and I am at 65c with an ambient of 75F (23.9c). I am guessing that if I remove the pump and go through the motions of cleaning it of thermal paste, I will shake it enough to get it working for a short time. Now that I switched from thermal paste to a graphite pad it can kind of conclude that it wasn't the thermal paste drying out unexpectedly. The sensor of the motherboard is still indicating that the pump is running but it doesn't seem to be circulating correctly. I tried contacting Enermax back in April but I never got a reply back. I contacted them this past Friday but since it's the weekend I am guessing I won't receive a reply till this coming week but since I never got a reply back last time I am not confident I will get one. Enermax only has some kind of email form for contact.

Anyone have any other good alternative to Enermax AIOs for a Threadripper CPU? I am not sure if I want to buy another enermax but I am going to need a replacement even if I can contact enermax to get this repaired. :(

Update: Removed the pump hit it a couple of times against my palm, reinstalled it and I dropped 20c lol
 
Last edited:
At a guess impeller is slipping on pump motor shaft or air bubble was creating an air-lock. Whatever the cause I agree it's time to get a new cooler. Most coolers now have AM4 mounts and there are several with bigger bases like you now have.

I would email Enermax again with all information including a copy of your April email. If your email has the ability to do delivery receipt conformation, use it so you have proof they received the email.

Here is a link to warranty
https://www.enermaxeu.com/service/warranty/
 
I'll call them tomorrow. I found their contact info here: http://www.ecomastertek.com/contact.php just in case anyone else needs to contact them. Hopefully I can get a human on the phone tomorrow.

The hitting against the palm thing only lasted for less than an hour then my temps started to rise again from 42 idle to 53 then to 60+. I had enough of it so today I bought another Enermax TR4 360. I just couldn't wait and Microcenter had it for $10 off. I'll get the other one fix and have it as a spare lol. Funny thing is the one I bought was leaking. Went back they didn't have anymore but a display model that was unused but inside a glass case, so I got that one. At least I am up and running again and everything seems to be working properly. I'll say it again the graphite pad is the best. It's so much faster than thermal paste. :)

I only briefly looked for a different brand and didn't find one I felt comfortable would cover the TR. I wasn't going to do air cooling anymore because I can't stand those giant heavy heat sinks sitting over my 1080Ti secured by only 4 screws on the socket. I couldn't stop worrying it would rip the socket off the motherboard while I was playing a VR game.
 
Sounds like the copper CPU block and the aluminum radiator causes major rusting issues. Your new unit will likely experience the same issues over time. This seems like a manufacturing defect, see newegg reviews of this unit. It stops cooling after 3-6 months.
 
Last edited:
Well that's not what I wanted to hear. :( If it fails again I'll probably just be pissed off enough to go full water cooling. But dam was this an expensive cooler. I am in $300 now since I have 2 of them. *sigh*

On the plus side, I got in contact with enermax finally. So it looks like I can get the broken one replaced.
 
If you can refund one of your coolers, you should:

https://linustechtips.com/main/topic/902933-enermax-360-mm-tr4-leaked-onto-2-gtx-1080tis/

When the loop gets clogged from rust, it also leaks, taking your system with it.

In the Gamer's Nexus video where they disassembled the unit, Steve already observed some rust. He commented on it might have been because he left it sitting for a while, but looking back I think it may be related to the general defect people are reporting:
I wish somebody could do a retrospective on these Enermax TR4 coolers. Having a $150 cooler that only lasts 6 month(at best) is not acceptable. This is exacerbated because this is just about the only viable AIO for TR4.
 
Yikes. Fortunately mine never leaked but I think I caught it failing early. This seems to be quite widespead - lots of people having issues with this cooler all of a sudden if you check the newegg reviews. Seems like a design fault to me.
 
I have two 280s I'm keeping an eye on, both from microcenter, no problems yet. I do wonder about the differences between the releases and surviving shipping seems to be a big issue.
 
I can only tell you that I am more concerned about the pump failure than leaks happening after some time installed. That linustechtips post above sounded like he didn't notice it was leaking when he installed it since he didn't have the monitor plugged in. The first one I got from microcenter was sealed in box and when I opened it I noticed condensation in the plastic bag with the radiator, then when I looked at the pump, there was corrosion on the bottom of the copper plate and a small puddle of coolant and the box around the pump was damp. There was no question it was damaged and given the box was sealed with no damage it might have been at the factory during packing. But there is definitely something wrong with their pump design there are just too many instances where the pump fails in 6 months. I guess if in December the pump starts to die on me I'll start planning water cooling solution. I really kind of don't want to go water cooling, for some reason the maintenance aspect always turns me off about it, but unfortunately I don't see another solution. There are exactly 2 made for TR cooling solutions that's not a custom loop, Enermax's AIOs and the Noctua Air Cooled, that's it. There are compatible AIOs but I'd never waste my money on one that didn't cover the whole heat spreader.
 
I've been pretty happy since switching to the Noctua air cooler, however it's not suitable for overclocking to 4Ghz (I leave mine stock) and it blocks the first PCI-e slot.

I wonder if we'll see some new coolers soon with TR2 coming out.
 
x3sphere said:
I've been pretty happy since switching to the Noctua air cooler, however it's not suitable for overclocking to 4Ghz (I leave mine stock) and it blocks the first PCI-e slot.

I wonder if we'll see some new coolers soon with TR2 coming out.
Click to expand...
Even after shifting the Noctua heat sink over it still covers the PCI-E slot? The video I saw on it's installation there were 4 screws that let you offset it.
 
You know what...

All this sounds like a valid excuse to build a custom loop.

You treadripper users can't halfass it with an AIO.
 
LurkerLito said:
Even after shifting the Noctua heat sink over it still covers the PCI-E slot? The video I saw on it's installation there were 4 screws that let you offset it.
Click to expand...


I saw that, but there's still not enough clearance. I think it is just an issue with the Zenith Extreme board that I have. Looking at other X399 boards, the first PCI-E slot isn't as close to the CPU.

Not much of an issue for me since you can boot from the third PCI-E slot fine and it is still x16. Probably would've went with a different board had I known originally though.
 
DTN107 said:
You know what...

All this sounds like a valid excuse to build a custom loop.

You treadripper users can't halfass it with an AIO.
Click to expand...
Yeah it is a good excuse, but most AIO users don't want to do the maintenance that is required of a custom loop. Honestly it's like the main reason I always stop myself from doing it. Just the thought that I'd have to drain, remove it all, clean the parts, reinstall and refill on a recommended every 6 months schedule makes me cringe.

In comparison I have an H100i AIO in my i7-3770k machine that has been working for 5 years with no maintenance but a few minutes of blowing off dust a couple of times a year. I was expecting the same from Enernmax.
 
LurkerLito said:
Even after shifting the Noctua heat sink over it still covers the PCI-E slot? The video I saw on it's installation there were 4 screws that let you offset it.
Click to expand...


Asus got sloppy and put slots too far inside the keep out zone, all the other current X399 boards are fine, even the micro atx.
 
I have the horrible feeling that there may be some design issues with their Liqtech coolers. Here are my thoughts:

  1. Mixed metal loop. These coolers use a copper cold plate and aluminum radiator & pump body.
    Aluminum and Copper do not like each other. It is not always a good idea to use them in the same cooling loop for the same reason you do not use copper and aluminum wire in an electrical circuit. Over a short time it is fine, but over the long run if you are careless about the coolant you use the two metals will react with each other. The Aluminum rad and copper heatsink can corrode via electrolysis if the coolant used is not designed to resist that chemical interaction. This is also why you need to be careful about the antifreeze you use in your car if it has an aluminum engine (the radiator is copper). The wrong antifreeze or even old antifreeze will eat away at the aluminum engine parts. Any good builder worth his or her salt will stick to the same kind of metal (aluminum or copper - not both) in all parts of the cooling loop.
  2. Manufacturing crap from the rad migrating to the cooling channels in the waterblock plugging them. As mentioned earlier Gamer's Nexus did a teardown of one of these coolers and saw such crap.

If you are out of warranty, and youj are willing/brave enough to do some work you can fix both of these issues. The cooler can be torn down, drained, flushed with distilled water and filled with a good coolant like Feser One or MCT-40. The 2 ports at the hose-side of the cooler will let you do just that.

For (2) you need to do some "open heart surgery" by unscrewing the copper base plate and cleaning out the base plate's microchannels. Be sure to smear a wee bit of Vaseline over the o-rings when you put the base plate back on - this will halt the possibility of leaks. You should be able to get at the impeller as well. A good clean out with some distilled water there will do the trick. The impeller is not attached to a motor driveshaft. It instead spins on a stationary axle and can be simply lifted out of its cavity. DO NOT lubricate the axle shaft when you put the impeller back on - you don't want to introduce impurities!! Finally put it all back together and tighten the screws a little at a time in a cross-star pattern to evenly until the base plate is tight.

As always after doing any kind of surgery to your cooling loop, test for leaks outside of your system by running it at least 8 hours with white kleenex placed under the waterblock yo show up any leaking coolant. If you don't know how to bench-run a power supply work without PC check out this article:
http://www.instructables.com/id/How-to-power-up-an-ATX-Power-Supply-without-a-PC/ A molex -> 3-pin adapter is needed to connect your pump.

Disclaimer:

I posted this as a help to those brave souls that are willing to tear down their expensive poor-performing Enermax TR4 AIO cooler to make it work again. I accept no responsibility if you totally bork your expensive Threadripper system after doing this. I am only passing on a little of my own knowledge and what I have gleaned by watching GamerNexus's teardown. What I have described here will void your cooler's warranty, so if you are having an issue and it is still in warranty, RMA it instead!

------------------- EDIT -------------------------

After a bit of googling around I found this:

https://linustechtips.com/main/topic/909855-enermax-tr4-360mm-aio-high-tempsproblem/

(A couple images from that thread)


(Impeller)
bE70sj4Z03g.jpg.e6ee877a45354d33b5df1e9b844700b9.jpg


(cold plate)
DrCbYEQRMu8.jpg.3212197dde0cfb952f93f73e85cdf85a.jpg


(sorry for the huge pictures - they aren't mine)

The organic buildup suggests that Enermax uses is a mixture of distilled water and possibly a wetter solution (i.e. Water Wetter or soap). And the corrosion suggests the aluminum and copper getting into a fight.

I'm not looking forward to overhauling mine :(
 
Last edited:
How much money could they have saved using an aluminum radiator? They have the audacity to charge extra for the AIO they should have at least put in a copper rad. Cheapskates.
 
The alu rad is not a huge problem. In fact I think most AIO coolers use an alu rad. Aluminum is way cheaper than copper after all. There are coolants that will block the chemical interaction. I used to use it as a matter of course back in the day. In fact some of the advanced coolants like Feser 1 already do that. My main shock is that they were not smart enough to put some kind of biocide in the water. That bio crap showing up and plugging the block/rad/pump is disgusting and inexcusable! If (and from what I have been seeing I am expecting this to be when ) I have to clean the gunk out of my system, you can bet I will be putting a much higher quality coolant in it - I only want to ever do that once!
 
cyberguyz said:
I have the horrible feeling that there may be some design issues with their Liqtech coolers. Here are my thoughts:

  1. Mixed metal loop. These coolers use a copper cold plate and aluminum radiator & pump body.
    Aluminum and Copper do not like each other. It is not always a good idea to use them in the same cooling loop for the same reason you do not use copper and aluminum wire in an electrical circuit. Over a short time it is fine, but over the long run if you are careless about the coolant you use the two metals will react with each other. The Aluminum rad and copper heatsink can corrode via electrolysis if the coolant used is not designed to resist that chemical interaction. This is also why you need to be careful about the antifreeze you use in your car if it has an aluminum engine (the radiator is copper). The wrong antifreeze or even old antifreeze will eat away at the aluminum engine parts. Any good builder worth his or her salt will stick to the same kind of metal (aluminum or copper - not both) in all parts of the cooling loop.
  2. Manufacturing crap from the rad migrating to the cooling channels in the waterblock plugging them. As mentioned earlier Gamer's Nexus did a teardown of one of these coolers and saw such crap.

If you are out of warranty, and youj are willing/brave enough to do some work you can fix both of these issues. The cooler can be torn down, drained, flushed with distilled water and filled with a good coolant like Feser One or MCT-40. The 2 ports at the hose-side of the cooler will let you do just that.

For (2) you need to do some "open heart surgery" by unscrewing the copper base plate and cleaning out the base plate's microchannels. Be sure to smear a wee bit of Vaseline over the o-rings when you put the base plate back on - this will halt the possibility of leaks. You should be able to get at the impeller as well. A good clean out with some distilled water there will do the trick. The impeller is not attached to a motor driveshaft. It instead spins on a stationary axle and can be simply lifted out of its cavity. DO NOT lubricate the axle shaft when you put the impeller back on - you don't want to introduce impurities!! Finally put it all back together and tighten the screws a little at a time in a cross-star pattern to evenly until the base plate is tight.

As always after doing any kind of surgery to your cooling loop, test for leaks outside of your system by running it at least 8 hours with white kleenex placed under the waterblock yo show up any leaking coolant. If you don't know how to bench-run a power supply work without PC check out this article:
http://www.instructables.com/id/How-to-power-up-an-ATX-Power-Supply-without-a-PC/ A molex -> 3-pin adapter is needed to connect your pump.

Disclaimer:

I posted this as a help to those brave souls that are willing to tear down their expensive poor-performing Enermax TR4 AIO cooler to make it work again. I accept no responsibility if you totally bork your expensive Threadripper system after doing this. I am only passing on a little of my own knowledge and what I have gleaned by watching GamerNexus's teardown. What I have described here will void your cooler's warranty, so if you are having an issue and it is still in warranty, RMA it instead!

------------------- EDIT -------------------------

After a bit of googling around I found this:

https://linustechtips.com/main/topic/909855-enermax-tr4-360mm-aio-high-tempsproblem/

(A couple images from that thread)


(Impeller)
View attachment 83246

(cold plate)
View attachment 83247

(sorry for the huge pictures - they aren't mine)

The organic buildup suggests that Enermax uses is a mixture of distilled water and possibly a wetter solution (i.e. Water Wetter or soap). And the corrosion suggests the aluminum and copper getting into a fight.

I'm not looking forward to overhauling mine :(
Click to expand...
I have sent this over to Enermax as of this morning. Thanks for posting. I will surely share feedback here.
 
Sure hope these were just bad batches and not a total botch job. I have a pair of 280s already and was planning on picking up the beefiest one available for TR2 when it comes out.

FWIW you can mix metal with the right coolant & design, it is not a deal-breaker by itself. Just about every AIO now does it anyways. Don't like it? Go make your own AIO and try to sell it, good luck with bullshit patents and a pretty well locked up retail market.

Of course once you go custom loop you have already given up on any sense of "cheaper" and full copper or bust.
 
I have 2 options if mine goes tits-up.

Choice (1) Open it up and clean out the sludge. Can't be too thorough because what builds up in your pump/waterblock also builds up in your rad. Not easy at all to clean without actually taking the hoses off.
Choice (2) is a custom loop. Got a handful of parts on their way from China - D5 pump, Bykski TR4 waterblock, full copper 360 rad (nobody smart does alu rads in a custom loop!!), bunch of 3/8" barbs & some nice soft pvc 8mm ID / 12mm OD hose. Still gotta score a decent bay-res for that pump. All told, about $500 CA (~ $370 US) in parts and I don't have to worry about galvanic corrosion. Custom loops ain't cheap! Even with Chinese parts. Since it is all copper or nickel-plated copper coolant will be straight distilled water + a drop or 2 of biocide (Dead Water) + silver coil.

What may happen is that if (when) the Liqtech 360 gives up, my system will get the custom loop while I clean up the Liqtech as a backup.
 
Last edited:
Hugely disappointed in this. I have seen two of these fail in under 90 days and I have read many customer reviews of people that experienced the same thing. The cooler works at first and seems to be a great solution. After a little bit of time, the performance degrades until it's unable to cool the CPU.

This cooler did not deserve the Gold Award. Enermax/sellers are really taking advantage of this review and people are being cheated.
 
I wonder if they responded to Kyle's inquiry. Maybe hardocp should retract the award and add more info to the review. Because there are literally people that bought a 100% failure rate product as a result of the review .
 
100% would imply they all failed, nobody intelligent samples at size=1. There has been no recall or supply notice yet, not to mention my pair still work fine every day so far. (data is not the plural of anecdote blah blah etc)

Maybe you should wait a bit before jumping to conclusions or internet hysteria hyperbole, having been in the computer parts world for awhile (among others) its more likely particular supply batches or production runs had problems than an entire line being defective. They would probably at least be off the shelves by now if the total fail rate was high enough to drive RMA costs through the roof or the first waves revealed a huge design flaw. Regardless of the outcome it takes time to play out.

With the various early reports there is definitely a closer eye being kept on these, I'm sure I'm not the only one logging pump rpm, cpu temps, ambient and load (otherwise temp is basically useless) full time now. With summer heat kicking in still no degradation yet.

Big picture this is the dice you roll with $100-range AIOs versus the gpu+cpu loop I want to build for TR2 will be $700+ easy, and thats just for a workhorse/man-portable setup not balls to the wall fancy.
 
branana said:
I wonder if they responded to Kyle's inquiry. Maybe hardocp should retract the award and add more info to the review. Because there are literally people that bought a 100% failure rate product as a result of the review .
Click to expand...
I am still waiting for response on this, but I did follow it up again today.
 
You must log in or register to reply here.
Back
Top