Disgusting Loop

dancook · 23 Apr 2018 at 08:44

I wonder why my water is green, did I add some cleaner to it - I can't remember, or is it from the reaction of metals?

So I think it's a copper radiator, nickel/copper blocks and I put a silver coil in the pump when I first set it up - not knowing that was an issue, I assume that was the issue...

Now I'm going to move the system into a new tower, and I want more radiators - do I use the existing loop and add radiators without worry?

I want to upgrade in the future though, and I didn't fancy buying all new blocks.

Untitled by dancook1982, on Flickr

Untitled by
dancook1982, on Flickr

Untitled by dancook1982, on Flickr

dancook · 23 Apr 2018 at 08:51

Acme said:
How long ago did you set it up? Have you ever done any maintenance or changed the fluid etc?

Less than 18 months ago.

I have done no maintenance, except remove the foam in the resevoir when it disintegrated and blocked it - and topped up water.

It was so beautiful

https://danielcook.com/2016/12/03/custom-watercooled-pc/

dancook · 23 Apr 2018 at 08:55

Acme said:
Looks quite bad for 18 months but yeah definitely due a clean out...

This is why I never bothered with custom water. I couldn't be arsed with the faff.

What kind of fluid have you been using? I'd only ever use something like Mayhems. - Tends to stay cleaner for longer.

Was it just distilled water?

Yes just distilled water, I read later on EK's website not to use silver coil as it reacts with nickel plating, is that right? or does it react with copper and the nickel plating was too thing.. hmm

dancook · 23 Apr 2018 at 09:11

Acme said:
I couldn't say, but if thats what the website says its probably right.

I'd drain it, flush it, take the blocks off and clean if necessary, and put some mayhems coolant in.

See if the tubes are OK first though. They might be stained etc.

Since I was looking at moving to a new PC, it'll have new tubes, a new pump/resevoir and three larger radiators - I guess that just leaves the blocks.. so if they can't do any damage to the new stuff, I'll keep using them.

dancook · 23 Apr 2018 at 09:13

Acme said:
I'd just flush them through first, make sure they look nice and clean. Hopefully you won't have to take them apart.

What you see is the nickel plating that's been eroded away (by the silver, I checked) to reveal the copper.

I'm not sure about 'looking clean' anymore

but I can look into cleaning them anyway

dancook · 23 Apr 2018 at 09:28

Acme said:
Ah, in which case I'll let one of the people more experience with watercooling advise on this one!

Not sure if that is likely to cause issue. With remaining bits of nickel flaking away and blocking coolant channels etc.

Thanks, I also did what I should have probably done from the start - email EKWB

So I built this in December 2016

As far as I understand the silver coil I added to the reservoir caused the nickel plating to erode.

Later the mesh in the pump became brittle and blocked the pump and I had to remove it in bits.

I want to move my system to a new case (before later upgrading)

Will my current CPU and GPU blocks be ok in a new system with new radiators/tubing/pump/reservoir?

dancook · 23 Apr 2018 at 09:42

pastymuncher said:
Those blocks need to be stripped down and thoroughly cleaned. To be honest I would remove all of the remianing nickel plating that comes into contact with the fluid to rule out possible future problems. A lot of it is gone anyway so you may as well take it back to bare copper.

Ok cheers, I've found these resource - I have lots of white vinegar, so some how remove the nickel coating and then clean.

https://www.ekwb.com/blog/how-to-clean-water-blocks/

dancook · 23 Apr 2018 at 11:06

Bugbait said:
Odd about the Nickel given that the Anodic index:

Silver, solid or plated; monel metal. High nickel-copper alloys −0.15
Nickel, solid or plated, titanium an s alloys, Monel −0.30
Copper, solid or plated; low brasses or bronzes; silver solder; German silvery high copper-nickel alloys; nickel-chromium alloys −0.35

The message from EKWB I got in the early stages of erosion

"If you are using a silver coil in the loop, than this is probably the cause of this as it is not compatible with Nickel plating and it can lead to a corrosion. This are probably the early signs of corrosion because of the silver coil you are using."

I thought I'd find some online material about it, first thing is they say to refrain from using strip of silver

https://www.ekwb.com/blog/antimicrobial-algae-protection/

and some other related discussion over it
https://www.reddit.com/r/watercooli..._coils_and_ek_blocks/?st=jgc2sewj&sh=aa225ba3

dancook · 23 Apr 2018 at 11:15

I'm building my next PC in the Lian-Li V3000WX case for over the top amount of radiator cooling

and it'll go under my desk.. and not really a 'show pc' so I might just keep things non-fancy, no plexi + nickel

dancook · 23 Apr 2018 at 15:21

Acme said:
So possibly OP removed the silver when he found out it may cause issues, and then neglected to put a biocide in?

I don't think there's bacteria, just the nickel reacting with silver - I didn't flush the system after removing the silver, so probably the particles still in there?

What you see in the photos is just corrosion of nickel showing the copper behind.

dancook · 23 Apr 2018 at 18:24

Since I'm looking at a new case and new loop, I might get a new block for the GPU and transfer to new case.

Then I might sell the 6700k, mobo, ram with rad/pump/res and case all together. Get rid of all the tubing and start fresh there.

dancook · 23 Apr 2018 at 19:00

Habakkuk said:
Well, your loop/pond has convinced me not to be lazy with mine :/

Hah no silver and nickel! For a start...

I followed advice here on forum to add silver coil.. Didn't know there would be issues

dancook · 24 Apr 2018 at 10:39

Kei said:
Using a good quality coolant will pretty much prevent corrosion even in mixed metal loops. I went and retrieved my koolance 380i block from my previous intel build which ran the same coolant for almost 4 years straight. @Illuminist , might be of some interest to you as I only showed you a the condition of a block after 6 months. This was a mix of approximately 20% GM dexcool antifreeze and 80% distilled water. The loop contained 2x alphacool brass/copper radiators, a koolance cpu 380i nickel block, an EK Titan copper acetal block and all cheap nickel plated 1/2" barbs. Not only was there no hint of corrosion anywhere, there was also no significant signs of buildup either. The only disadvantage to the fluid was the fact it stained the tubing pink. I repeat what I've said before, if it can prevent a huge mixed metal cooling loop containing both aluminium and copper from corroding itself to dust over 23 years, it's easily good enough for my pc.

Thanks, so you're suggesting instead of typical coolant I should use GM dexcool antifreeze and distilled water in future loops?

When I google it, GM car users suggest it's destroying their cars..

https://www.consumeraffairs.com/automotive/gm_dexcool.html

https://www.consumeraffairs.com/news04/2006/08/gm_dexcool.html

but i'll be ok?

I might just use black tubing anyway

And you'd suggest buying this over, regular coolant and just not mixing your metals?

dancook · 24 Apr 2018 at 11:52

Kei said:
Indeed. Dissimilar electronegativities will result in galvanic corrosion. The smaller the difference the slower they react.

Brizzles sticky thread contains a great deal of info on this already. Well worth a read for those not in the know.

I just checked the sticky from Shayper, which is where I got the idea that distilled water and a silver kill coil was the norm......

https://forums.overclockers.co.uk/threads/custom-loop-guide-how-to-watercool-your-pc.18290835/

"Again, this is almost entirely up to you. The norm, however, is to use distilled or deionized water for the loop, and place a silver kill coil in the res to prevent algae and other little buggers from growing in your loop."