@gahz1, If your using all copper+ distilled, you will get all copper ions in the water rapidly at first (erosion corrosion), but after a few months you will no longer have distilled water, ie copper ionized... the water will be much less aggressive, and rate at some point becomes almost a steady state, ie very limited erosion corrosion. Which is why it is safe to run distilled in our loops since recirculates the same water, because it quickly becomes ionized. But it is not ok to pipe continually fresh distilled through copper piping, because after millions of gallons + years, it erodes the pipe and eventually can leak. And ideally you want all the same ions, so no galvanic corrosion starts.
When you have dissimilar metals, you get erosion corrosion first, but then for example you have silver ions +.8 interacting with nickel plating -.2, and nickel on plating will lose an electron to the silver ion, and now you have you have another type of corrosion as you have set up a galvanic cell.
But yes erosion corrosion always happens, but that is self limiting in a loop that recirculates the same water. And you want some erosion corrosion to take place, after all copper ions are antimicrobial.
@AMC, anecdotal reports help to determine possibilities but not probabilities, and with so many variables in each loop and in manufacturing, it will never be all or none.
For example say EK makes 300 nickel plated blocks, and under mass production keeping costs down, invariably the quality will be variable, some prepped better than others, plated thicker, etc. And 1/3 (100 each) randomly selected for 3 groups 1) inhibited water 2) distilled water 3) distilled water with copper/silver added. All ran in exact same loop (to eliminate the variability in our loops), same rad/blocks, etc, only difference was liquid used.
And say nickel loss with copper exposed happened in 5/100 of inhibited group, 20/100 in distilled group, and 50/100 in copper silver added group. And btw, in real testing, that is type of thing you see, rarely is it the all or none people seem to expect. Then you would claim 10x reduction in nickel loss with inhibitors vs copper/silver addition, and 4x reduction of inhibitors vs distilled only. And basically that is type testing EK did, and likely koolance, though on smaller scale.
Can you argue, why not plate the nickel 2x as thick with hardening outer layer or 5x thicker like that used in marine/saltwater settings...sure. That will solve the problem, you can use whatever fluid you want, but you will pay 3X the price.
I think you are left with 1) pay for custom thicker nickel plated blocks at 3X price and use whatever cooling you want. 2) buy mass produced relatively cheaper nickel plating, and either a) cross your fingers that you got one that was perfectly prepped/plated + your particular setup/use is the least harsh that allows nickel survival, since you cant rma it, or b) use inhibitors and pray you dont have the 5/100 that wont survive even using inhibitors under your particular use.
| CPU | Motherboard | Graphics | RAM |
|---|
| Ivy 3770k 4.7ghz 1.31v | Gigabyte Z77 UD5 | Titan | Corsair Dom |
| Hard Drive | Cooling | OS | Monitor |
|---|
| Intel 520 SSD 240gb | dual pump, 360 xspc + 280 sr1 rads | Windows 7 64bit | Samsung 26in LCD |
| Power | Case |
|---|
| Corsair HX850 | Lian Li A71f |
| View all |
hide details
| CPU | Motherboard | Graphics | RAM |
|---|
| Ivy 3770k 4.7ghz 1.31v | Gigabyte Z77 UD5 | Titan | Corsair Dom |
| Hard Drive | Cooling | OS | Monitor |
|---|
| Intel 520 SSD 240gb | dual pump, 360 xspc + 280 sr1 rads | Windows 7 64bit | Samsung 26in LCD |
| Power | Case |
|---|
| Corsair HX850 | Lian Li A71f |
| View all |
hide details