Overclock.net banner
1 - 9 of 9 Posts

·
Registered
Joined
·
78 Posts
Discussion Starter · #1 ·
So I ran across this chart when researching corrosion inhibitors and I thought I would share it with all of you.

ASTM D1384 is a corrosion test that Dow Chemical Company conducts when testing products with their corrosion inhibitors. They use water as a baseline and control.



The radiators used in water cooling have solder inside them. This test shows that the solder in radiators are probably the weak point for galvanic corrosion inside our loops using water + biocide. I just found this interesting and thought I would share.
 

·
Premium Member
Joined
·
4,954 Posts
Per haps weak point in terms of galvanic potential on a typical brass/copper/solder/stainless/nickel/plastics type loop:
corrosion1.jpg?w=614


I did loose one radiator by letting water (without inhibitors) sit inside one stored for a few months.
corrosion4.jpg?w=614&h=340


However, nickel also has some potential.
corrosionplating.png?w=614


Aluminum has the greatest potential:
corrosion6.jpg?w=614


Historically though there is much more to it than potential, potential just means rate under ideal conditions and failure is not only affected by rate it is more affected by thickness and condition.

Nickel plating tend to create the most rapid condition and it also has the smallest material thickness. Because of coppers more nobel rating, the corrosion occurs under the nickel surface (crevis corrosion). Under that nickel surface the fluid is very stagnant and can much more rapidly raise conductivity levels (electrolyte). The very thin plating is also much more quickly impacted by corrosion than thick solid metals.

radiator solder is relatively thick, also since Tin is lower on the chart than copper, there is no pocketing. If a system is constantly moving fluids then the electrolyte building process is kept very slow. In my case I put it under the ideal condition by letting it sit for months. So yes it is possible, but as far as I know that is the only radiator corrosion failure example I've seen.

Aluminum/copper blocks used to be common, but corrosion was a big problem. Most manufacturers have since given up on aluminum blocks.

The bigger risk anymore is nickel plating because of it's super thin material thickness and also because copper is more nobel than nickel which results in crevis corrosion.

Adding an inhibitor and regular cleaning/flushing with new is recommended for plated items. Generally we see fewer issues since the abandoning of aluminum but plating thicknesses and quality of plating and electrolyte development potential is equally if not more important than the actual material galvanic potential.

Also there are other corrosion types other than galvanic that are affected by fluid ions. I don't understand that type as well, but silver as a biocide is one that I have heard should be avoided. Unlike galvanic corrosion which require metal to metal contact, that other type of corrosion just works through the fluid.

My rule of thumb recommendation:
-Avoid aluminum in especially in blocks. It can be ok in rads with a good inhibitor but I like to avoid any aluminum if possible.
-Use a corrosion inhibitor in any a aluminum or nickel plated loop (use manufacturers recommended coolant if warranty is important to you)
-Clean and flush frequently (every 3 mo) to reduce electrolyte potential and also to inspect.
-Unplated copper/brass can be run without an inhibitor in normal use, but regular cleaning/flushing/inspection is still a good idea. note that it does often void warranty though.
-Doesn't hurt to add an inhibitor regardless.
-Avoid silver.

-If you value warranty and RMA capability highly, use only the manufacturers approved coolant. Any home brew or third party fluid can affect your warranty coverage. You better follow their rules if you insist on their support later. They won't play nice if there is questionable home brewing going on since there are so many possible causes that are then outside of their control.
 

·
Registered
Joined
·
78 Posts
Discussion Starter · #3 ·
Great Information Martin. Thank you for sharing it with us. REP +

So silver coils would slowly add ions to the water acting as colloidal silver in our loops killing organisms.
Slowly because there is no direct contact with other metals and does so via ion exchange in the solution.

The super thin plating of nickel release ions at a much faster rate due to the thin nature and direct contact. The acceleration of this by silver ions is something I haven't thought too much about.

There was a company blaming there nickle plating problems on silver coils or copper sulfate in loop if I remember right.

Perhaps thicker plating would help with this problem. All the imperfections from manufacture or scratches would complicate things due to the thinness of plating.

Regular system maintenance to eliminate overly high levels of loose ions is probably the best bet to slow this process down.
 

·
Old to Overclock.net
Joined
·
1,110 Posts
Quote:
Originally Posted by Cataclysm PC View Post

The super thin plating of nickel release ions at a much faster rate due to the thin nature and direct contact. The acceleration of this by silver ions is something I haven't thought too much about.

There was a company blaming there nickle plating problems on silver coils or copper sulfate in loop if I remember right.

Perhaps thicker plating would help with this problem. All the imperfections from manufacture or scratches would complicate things due to the thinness of plating.
Not exactly. In terms of ion release what matters is surface area--the nickel could be four inches thick and would release ions at the same rate if the surface exposed to water is the same. BTW I'm not even sure whether nickel releases ions as readily as silver does, but let's assume so for this discussion. In any event, he's not saying that the presence of silver ions affects the release of nickel ions. Silver (or other metal) ions can combine with other impurities in the coolant to make the coolant electrolytic, and that of itself has an impact on corrosion processes generally. FWIW, release of ions in this way isn't corrosion.

The problems with EK's nickel a few years ago were believed to be caused by micropores in the plating, most likely due to improper cleaning of the blocks and/or plating solutions during manufacture. Some testing and analysis was done on another WC forum that showed pretty convincingly that the pores were the problem, but in fact that testing also showed that the plating was the proper thickness. You are correct however that they blamed the problems on the use of certain coolant additives, but if the plating didn't have those pores to begin with, it wouldn't have been a problem. Anyway, in the aftermath they did switch to a more robust plating procedure but again, I don't think thickness was the issue.
 

·
Premium Member
Joined
·
9,681 Posts
Quote:
Originally Posted by Cataclysm PC View Post

Great Information Martin. Thank you for sharing it with us. REP +

So silver coils would slowly add ions to the water acting as colloidal silver in our loops killing organisms.
Slowly because there is no direct contact with other metals and does so via ion exchange in the solution.

The super thin plating of nickel release ions at a much faster rate due to the thin nature and direct contact. The acceleration of this by silver ions is something I haven't thought too much about.

There was a company blaming there nickle plating problems on silver coils or copper sulfate in loop if I remember right.

Perhaps thicker plating would help with this problem. All the imperfections from manufacture or scratches would complicate things due to the thinness of plating.

Regular system maintenance to eliminate overly high levels of loose ions is probably the best bet to slow this process down.
Plating needs to be around 12-14 microns for corrosion protection,EK (for example) have a average plating thickness of around 4 microns....this results in a porous nickel strike,the pores become hot spots of galvanic activity resulting,as Martin has already posted,crevice corrosion.

Sliver and copper and not a good mix,the aviation industry has had endless problems with silver coatings on wire,they call it the Red Plague...

Thicker plating and more care to ensure tooling marks are gone before plating,you want the surface smooth and uniform.

Watercool, for example,sandblast the tool marks out before plating.
 

·
Registered
Joined
·
469 Posts
Im building a watercooling loop, and im researching corrosion.

To have corrosion protection, it seems you either need to use similar metals, or different metals with corrosion inhibitor.

Here is what im planning on choosing.
1)Aquacomputer copper radiator, which are solder free, meaning you gon only copper in the loop
2)Copper CPU block with silver bullet, or Silver CPU block without silver bullet.
3)Distilled water as coolant, as i drink distilled water myself, and have a machine that makes me water every day.

Now all i need to figure out is the tubes.

Untill choosing tubes, the loop will have copper, and silver, silver in very small quantities, if i will use a bullet, or in bigger quantities if i will go for the aquacomputer silver cpu waterbloc.

I also need to find fittings made of pure copper.

The BEST solution out there which doesnt exist, is to use all siver in the watercooling loop. THis however cant be done.

1)Silver cpu block, we have the aquacomputer all silver cpu waterblock
2)All siver radiator, dont exist.
3)Silver fittings exist
4)Silver plated solid tubings also dont exist.

The best solution would be to use all coper, distilled water, a bit of silver as bactericid, and this is what i think ill be using.
Copper tubing ill still need to research that !

Here is what I wrote in this post a long time ago!
 

·
Registered
Joined
·
469 Posts
What i could do is create an entire loop with only copper, and then check the tds rating over time, to see if copper ions are going into the solution.
That will restrict the usage of certain components, such a multiple splitters, or anything that isnt copper based. But the more im reading about the olygodinamic effect of copper the more i s tart to think that probably a full copper/distilled water loop will also probably do the trick.
 
1 - 9 of 9 Posts
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top