Basic watercooling questions

Zarathustra[H] said:
I'm not sure how you are insulating them electrically, but the current loop occurs via the fluid.
Click to expand...

Negative, ghost rider.
The fluid is only one half of the current loop.The aluminum and copper have to be electrically connected (aka share the same ground), for galvanic corrosion to occur.
In my case, the aluminum reservoir and the copper rad can both be isolated using plastic mounts.
---
I certainly do appreciate the concern though.I'm going to make sure I watch things carefully for the first while.

:D
 
In most cases though the radiator is screwed directly to the case and the waterblock to the motherboard which has some grounding to the case for electrical safety. Once upon a time we used push thru plastic standoffs, but the ports on the back of the MB always contacted the metal shield that was attached to the case. Nowadays we use brass standoffs that contact the motherboard's ground plane. I don't know very many people who bother to electrically isolate the rad or motherboard from the case. Thus in just about every system that mixes metals in a cooler loop a small circuit exists.
 
BrainEater said:
Negative, ghost rider.
The fluid is only one half of the current loop.The aluminum and copper have to be electrically connected (aka share the same ground), for galvanic corrosion to occur.
In my case, the aluminum reservoir and the copper rad can both be isolated using plastic mounts.
---
I certainly do appreciate the concern though.I'm going to make sure I watch things carefully for the first while.

:D
Click to expand...


Well, you've obviously done your resesrch, and you do you, but if insulating it is as easy as you suggest, it seems odd that the caution against using dissimilar metals would be the number one rule of water-cooling.

I hope for your sake that you are right, but I also do feel I need to reply with caution in case any other beginner comes in here, reads this and gets themselves in trouble.

Let's be very clear about this. The number one rule in the water cooling community is still that you never ever under any circumstance use copper and aluminum in the same loop. Disregard with great caution and at your own risk.
 
Hehe.

I totally agree with your caution, I would not want to wreck someone else's rig.
-----
If you look at the picture of my reservoir , you'll notice the only thing that physically touches the aluminum are the peltiers, so the electrical insulation is a gimmie there.
The rad is a bit harder , it needs a rubber gasket, and nylon screws.

---

If I turn out to be wrong and demolish my cooling loop, well, I'll come and post pics of that too.

(y)
 
BrainEater said:
The rad is a bit harder , it needs a rubber gasket, and nylon screws.
Click to expand...
Does the passage of coolant through tubing, and the inherent friction between the two, not generate a minute charge and function as a pathway between the metals, unifying them, permitting the exchange of their differing charges... despite being isolated in the way mentioned? (similar to the static electricity generated with air flowing quickly through tubing)

Sincerely asking, not being a smartass :)
 
I'm going to say no.Either way, that's not galvanic corrosion....galvanic corrosion is like putting acid into your loop.It dissolves things.

Galvanic corrosion is the same as a battery.

Take a piece of copper, and one of zinc (or aluminum) and jam them into a lemon.Connect a LED across the two metals , it will light up (current flow).
If you remove the lemon, no current flows.If you remove the LED, no current flows.

Basically , I am suggesting I will remove the LED(electrical connection) from my watercooling loop , and prevent galvanic corrosion.
----

This isn't to say I have not missed several critical factors, and might blow the whole thing to smithereens.It's my first WC rig.
This is also [H] so I'm diving in right off the handrail of the tallest tower !

:ROFLMAO:
 
I guess that wasn't my understanding of the process at all...
Wiki said:
Galvanic corrosion (also called bimetallic corrosion) is an electrochemical process in which one metal corrodes preferentially when it is in electrical contact with another, in the presence of an electrolyte.
Click to expand...

Which kinda just says what's already been stated, that the Copper and Alu already have their own electrical potential and base charge, so when connected by coolant (acting as an electrolyte), whether there's actually a current present (as far as I know) is irrelevant due to that. Granted, the process will be significantly slower than it would be in the presence of an induced voltage source, but it's still there. Just as why iron/steel rusts in salt water. IIRC, in those instances they use a "Sacrificial Anode" to inhibit the corrosion process on the steel.

That's why I was thinking about the friction generating an added electrical presence in the system, and the fact that the liquid would be conducting, I was speculating it may mean your isolation (rubber and nylon) could be completely in vain.
 
Hehehe....
Friction has nothing to do with it.

The Wiki is totally correct , but I'll change the emphasis so it reads like it should.

Galvanic corrosion (also called bimetallic corrosion) is an electrochemical process in which one metal corrodes preferentially when it is in electrical contact with another, and in the presence of an electrolyte.
That means both an electrical connection, and in the presence of an electrolyte .
I've just finished electrically isolating both the reservoir and the rad....
Final assembly on the reservoir is almost done :



IMG_6725.JPG

Still have to finish the wiring on the reservoir , then its re-assembly time for the whole rig.

I have 5 days till ski time.(I have a lot of work to do.)

:eek:
 
Two questions about coolants - (1) any recommendation for premix and where to buy? I'm leaning towards XSPC EC6; (2) if use distilled water, I will need biocide and anti-corrosive. Any recommendations for these two including where to buy? Anything else I need to add to the water?
 
mosesbrodin said:
Two questions about coolants - (1) any recommendation for premix and where to buy? I'm leaning towards XSPC EC6; (2) if use distilled water, I will need biocide and anti-corrosive. Any recommendations for these two including where to buy? Anything else I need to add to the water?
Click to expand...
A lot of coolants that are available as premix are also available as a concentrate, containing biocide and anticorrosive additives already. It's hard to go wrong with a simple clear premix or concentrate. I'm fond of Aquacomputer Double Protect Ultra.

Performance-PCs, ModMyMods, Amazon, etc etc.
 
VanGoghComplex said:
A lot of coolants that are available as premix are also available as a concentrate, containing biocide and anticorrosive additives already. It's hard to go wrong with a simple clear premix or concentrate. I'm fond of Aquacomputer Double Protect Ultra.

Performance-PCs, ModMyMods, Amazon, etc etc.
Click to expand...

Yeah, there are plenty to choose from. I tend to like EK's Cryofuel. They also come in both 1L premix bottles or 100ml concept tested bottles to be mixed 9:1 with distilled water to make a liter of coolant.

In general most coolants are fine. I'd stay away from the opaque or shimmery looking ones though, as they tend to have microparticles that can gunk up the tiny microchannels in modern blocks. They are typically considered "for show only"
 
I am not done quite yet,
but I have a sneak peek before I throw this back to my work log ;


IMG_6728.JPG
I am just buttoning up the wiring on the reservoir/cooler, then I can get it in and run the machine.

(y)

man I hope I didn't fuck anything up.
 
Hee hee...

~500 days since this rig started. Time for an update on this.

This rig has processed several petabytes of video.
I don't use the peltier cooler units, it doesn't need it.

There is no evidence of galvanic corrosion.The glycol is still almost clear.I have not done anything to it....no changes no nothing.
It's flawless....I'm going to top up the glycol in the next month....100ml mebbie.

I do realize I can't say "I told you so" without totally disassembling the whole loop and photographing it....
I'm going to go the [H] way.
It'll run till it dies....I'll photograph it then...see ya in 10 years.

I will photograph the glycol sample.

:D
 
Last edited:
You must log in or register to reply here.
Back
Top