Watercooling RAM?

Associate
Joined
10 Nov 2009
Posts
118
Location
Doncaster
Theres a first time for everything, i've built a number of watercooled rig's over the years, but never have i chosen to watercool the RAM..
I've seen video's on the Tube about removing the headsinks attached to various makes and models of RAM..

I've got a couple of questions that i cant find a suitable answers to..

1) They say to test the RAM before doing anything to it, to ensure that it works.. Thats all being well, but how do you test something that cant be powered up due to it being in the same loop?

2) Am i right in saying, removing the heatsinks voids your warrante?

3) What are peoples opinions on watercooling RAM.. performance? purely aesthetic?

4) What makes and models of RAM have people used.. I'm looking at using:
https://www.overclockers.co.uk/cors...d-channel-kit-cmd16gx4m4b3000c-my-417-cs.html

There was another question but my minds gone, so ill add it later when i remember..
 
1) They say to test the RAM before doing anything to it, to ensure that it works.. Thats all being well, but how do you test something that cant be powered up due to it being in the same loop?
--Surely just test the ram in your system before building/updating your loop?

2) Am i right in saying, removing the heatsinks voids your warrante?
--Yup

3) What are peoples opinions on watercooling RAM.. performance? purely aesthetic?
--General opinion seems to be that it has relatively little advantage as far as performance so is only really am aesthetic choice.

4) What makes and models of RAM have people used.
--I've got corsair dominator DDR3 which doesn't require removal of the heatsinks to work with EKs monarch ram blocks. Unfortunately from what I've heard that no longer the case with the dominator platinum.
 
1) They say to test the RAM before doing anything to it, to ensure that it works.. Thats all being well, but how do you test something that cant be powered up due to it being in the same loop?
--Surely just test the ram in your system before building/updating your loop?

Only issue there, is its a total new build, the GPU's are coming with stock water blocks, the CPU dosnt come with a cooler, so will be water cooled from the get go..

I dont have or know of anyone with a DDR4 build to test them on..

If i had a spare air cpu cooler it wouldnt be a massive issue but im not dismantelling my i5 for it..
 
Have you the fittings to be able to run flexible tubing and do a temporary loop without the RAM blocks installed?

Really bad idea not to have tested the components before doing something that will void the warranty.

I had a similar issue once where I didn't have a cpu cooler only my block, so ran just the CPU and motherboard blocks on a temporary loop whilst testing my graphics cards on air. In the end glad I did as one of the cards was faulty and had to be RMA'd.

E-I
 
Only issue there, is its a total new build, the GPU's are coming with stock water blocks, the CPU dosnt come with a cooler, so will be water cooled from the get go..

I dont have or know of anyone with a DDR4 build to test them on..

If i had a spare air cpu cooler it wouldnt be a massive issue but im not dismantelling my i5 for it..

Well I think you know your answer, sounds like your just going to have to roll the dice and hope the parts work. Generally as noted when they say test your Ram, most people will check the ram and most other components work before putting them under water.
 
Watercooling RAM is utterly pointless. They don't even need heatsinks in most cases. It's expensive, complicates your loop, more risk of leaks and harder to upgrade.
 
Last edited:
Have you the fittings to be able to run flexible tubing and do a temporary loop without the RAM blocks installed?

Really bad idea not to have tested the components before doing something that will void the warranty.

I had a similar issue once where I didn't have a cpu cooler only my block, so ran just the CPU and motherboard blocks on a temporary loop whilst testing my graphics cards on air. In the end glad I did as one of the cards was faulty and had to be RMA'd.

E-I

I can relate to this, bought a980ti the other week, less than 3hrs use it pop'd so it's been RMA'd, but as again, the replacements are coming with water blocks pre-fitted.. So testing on air isnt physically possible, its a gamble they will last.

Watercooling RAM is utterly pointless. They don't even need heatsinks in most cases. It's expensive, complicates you're loop, more risk of leaks and harder to upgrade.

I think, i might just agree with you on this... It was one of them ideas, can i do it, should i do it... the mobo, cpu and both gpu's will be, so thought why not do the RAM aswell...
 
Last edited:
All it does is add restriction to the loop and add even more tubing. Lot's of expense for zero gain. My ram doesn't even have heatsinks yet is overclocked and overvolted and barely gets warm.
 
I've water-cooled mine purely for looks and because I've not done it before.

If your confident in setting it up and you have money to spare why not. I personally think it makes my system look better.

I did rip a load of ram chips off the pcb on my first ram kit trying to remove the heat sinks so make sure you pick your ram wisely.
 
At the risk of jumping on the bandwagon, I'm with Minstadave. However, I'll back it up with some numbers to show it's not just an arbitrary opinion. I'm running G.Skill DDR4@ 3600MHz on its default XMP profile. After gaming it will hit 40°C. Right now, it's at 30°C. I don't have a massive amount of airflow through my case - 1x120@800rpm in and the same out. Unless you're going to be pushing it to the limit voltage wise, it's just not necessary - frankly I'm far more worried about my SSD temps! ...and even those I'm looking at passive heatsinks.
 
Back
Top Bottom