Follow along with the video below to see how to install our site as a web app on your home screen.
Note: This feature may not be available in some browsers.
Wow that's beautiful.
And no risk of corrosion either, seeing as the radiator's external![]()
i don't understand why the rad being external would cut down on corrosion
no offense .. i probably just don't get it
Because there's no common electrical ground connecting the aluminum and copper pieces.
The best way to prevent corrosion is to get rid of the ground... either by spacing your radiator out with nylon washers, or using nylon screws. If the copper block and the aluminum radiator are allowed any sort of metallic pathway, they will corrode. Without this path, they will not.
So you're saying that the external radiator is not grounded? How do you know this? It looks to me like there's a 4-pin molex connection on the PCI bracket, which could pass a ground to the AquaDuct very very easily - does AquaComputer specifically note that this ground makes no contact with any metals inside the AquaDuct?
Also, I'm no expert, but I wasn't under the impression that a ground was required for corrosion to occur - I had thought that all you needed was two metals (with different galvanic potentials) that are connected electronically in the presence of an electrolyte - which to me means that if you've got copper pipes and aluminum blocks with the same coolant running through them, you need to add a load of anti-corrosion additive.
Even if it used an external adapter or it's own PSU, isn't it common practice to pass a ground?Hrm... So the aquaduct uses PSU power does it? I was unaware of that. In that case, yes, there's almost certainly a ground.. I could have sworn it used an external adapter when I read the thread ~6 months ago.
You use absolutes too liberally. While the coolant is not the same as a metal rod connecting the two metals, the coolant can conduct electricity - so to a limited extent, you still have an electrical connection between the metals.And by "ground" I do mean electrical pathway connecting the two metals. It is essentially a ground. If the copper pipes and the aluminum blocks arn't connected by some sort of metal (or steps are taken to prevent this connection) nothing can occur.
Certainly not a bad idea, but again I have not been under the impression that a metal connection is required for corrosion to occur.Well, I'm sure taking the unit apart, and including some well placed rubber washers would prevent the rad from touching ground.
Even if it used an external adapter or it's own PSU, isn't it common practice to pass a ground?
Common practice... maybe, but if it can avoid the corrosion problem, I would have left this common practice out, in favour of increased reliability.
You use absolutes too liberally. While the coolant is not the same as a metal rod connecting the two metals, the coolant can conduct electricity - so to a limited extent, you still have an electrical connection between the metals.
Right, but by definition... Corrosion doesn't happen unless there's an anode/cathode pair connected electrically, and submersed in an electrolyte. There has to be an electrical connection between the components other than the conductive solution (from what I've read, I could be wrong, but most people seem to agree with this). The coolant is the electrolyte solution, it shouldn't act as the electrical connection as well.
Certainly not a bad idea, but again I have not been under the impression that a metal connection is required for corrosion to occur.
See quote below:
Australian Stainless Steel Coalition (lol) said:Galvanic corrosion can only occur if the dissimilar metals are in electrical contact. The contact may be direct or by an external pipe or wire or bolt. If the dissimilar metals are insulated from each other by suitable plastic strips, washers or sleeves then galvanic corrosion cannot occur. Paint is not a reliable separator from direct contact although painting all of the noble metal is quite effective. Painting the active metal causes drill holes at coating defects.
can you use any chipset block for the SB and or any NB block for the NB or do they have to be designed for this board. im looking into getting an evga but would wanna water cool it. is it worth doing the mofset as well?
You can still get a common ground through the water.....because you know.....water is conductive.....
Radiator > Water > Waterblock > Grounding ring on motherboard
You can still get a common ground through the water.....because you know.....water is conductive.....
Radiator > Water > Waterblock > Grounding ring on motherboard
H20 is conductive and I was wondering when someone would mention that.
actually H20 isn't
I knew that I should have posted this in AC thread.
I heart your aquaduct Seeker, I'm just trying to clear this corrosion bit up. I'm not 100% sure myself.
Please post this in the AC thread as we just want to hear about AC goodness and skip the BS.![]()
Have any temps yet?
woah! is that 2 8800gtx's and your cpu? wow! what rad are you using?
oh. i may need another then. my triple rad wouldnt cope. i reckon a triple and a single?
well my loop at the end of the year will look something like this:
Pump---Cpu block (quad core @ 3ghz i hope)---Motherboard blocks---Gfx blocks (overclocked but not overvolted)---rad---res---pump....
would it cope?
im happy to get another single....
cool. dont forget the mobo blocks. so if i go cpu---mobo---rad (small)---gfx blocks---big rad...etc...