oil instead of water..........???

Grammar **** up in my post, I meant the frictional loses will be higher and the flow rater lower. Neither of them good.

Corrosion in the sense it's normally met in watercooling, where someone has a piece of aluminium and a piece of copper bolted to each other, is easily understood. Electrons can flow from one metal to the other directly, and ions move through the fluid to complete the circuit. For every n electrons to move around, some fraction of n ions move from one to the other.

The next step is to move the aluminium and copper apart, and bolt both to the case. Corrosion slows, but does not stop. Electrons now have to navigate the resistance of the case in order to complete their loop, and ions have to move further through the water.

Jokester is suggesting taking that to the limit. If there is an electrical connection through the water but nowhere else then the electons-through-the-metal part of the loop is blocked, and corrosion will cease a very short length of time after it begins.

Why some metals are more electronegative than others is beyond me. There's either quantum mechanics or lies to fall back on, neither of which I follow. Regardless, a potential exists between dissimilar metals in electrical contact (also between dissimilar areas on a given piece of metal).


edit: Jokester knows what he's talking about here. Links are readily found through wiki, but simply thinking about it should be convincing enough. Pure water is an effective insulator. Pure water is also ficticious, it doesn't remain so for any length of time (that would be entropy). Air dissolves in it, copper does, even plastic does. Pure water will also conduct, as it tends to dissociate into H+ & OH- [I think that's a lie, I didn't get far enough with quantum mech to find out the reality. It might be H30+ and OH-, but same effect], and the ions then move independently. The effect is much exaggerated by CO2 dissolving in the water though, so pure water conducting is a fleeting effect.

What you're interested in is ions moving from one metal to the other. That happens through the water. However, without a route for the charge to flow through, the ions accumulate on the surface until the net potential is zero. Same idea as a capacitor.


edit2: While I'm on the topic, if anyone reading this knows things about alternating current electrolysis please give me a shout. I had a piece of steel near a piece of copper, only water and acetal joining the two. A largish induced voltage in the copper, but at a very low current, caused a hell of a lot of rust and I don't know how. It plated the iron onto the copper in traditional fashion, but I thought that would required DC, and apparently it doesn't.
 
Last edited:
sure, and that is a good example, but even though pure water has a low conductivity it will still allow the flow of electrons, ions will increase it greatly but it is still conductive:

"Pure water has a low electrical conductivity, but this increases significantly with the dissolution of a small amount of ionic material such as sodium chloride."

http://en.wikipedia.org/wiki/Water

the semantics are driving me crazy!
 
It's the transfer of charge via ions though, rather than electrons. If you remove the ions (such as pure water) then it becomes a lot less conductive as the only ions available to allow a current to flow are the H and OH ions. Adding salts and so on increases the amount of dissolved ions.
 
Jokester said:
It's the transfer of charge via ions though, rather than electrons. If you remove the ions (such as pure water) then it becomes a lot less conductive as the only ions available to allow a current to flow are the H and OH ions. Adding salts and so on increases the amount of dissolved ions.
Click to expand...

I agree with that Jokester!

Thanks JonJ678 for your lengthly explanation........it does help!

question is - after a year of running a water system with corrosion inhibitors and another with silicone oil, would I expect there to be no visible difference, will the oil one still look good as new while the water one slightly corroded or will the hand of god still have a role to play in the outcome?
 
Last edited:
I doubt you would be able to visually tell the difference and it would certainly not make a measurable difference to cooling performance.
 
JonJ678 - might sound off topic but how is the flow of charge in a capacitor explained across the dielectric (in a cap where there is a fluid as a dielectric)?
 
How much is the OP basing his plans on the photo in post #5? That to me looks like a combination of normal copper oxide you nearly always get (black stuff) and what looks like gunge from using a dyed coolant, probably feser blue and/or plasticiser. The OP seems to be basing this whole idea of vigourously preventing any type of blemish developing so that if he wants to sell his gear in a years time it will be spotless, is that right?

As you surely know the black stuff can easily be removed prior to selling, and persoanlly I don't think it affects cooling performance while it is there so doesn't need to be maintained by periodic cleaning. The gunge can be avoided by not using naff fluids. And galvanic corrosion can be avoided by easy means also. For me you are trying to solve a problem which isn't there.

Also, if you do come to sell up all your gear, you will have a rad, res, pump etc all exposed to oil which you will have to scrupulously clean prior to selling as no one is going to want to buy oil contaminated water cooling gear. So how are you going to clean an oil system down fully without leaving any trace of the oil?
 
Mr Gimp - Fluorinert looks good, shame it is fluorocarbon though, I'm always on the lookout for coolants ya see :P Expensive though.... :(

Jokester - thanks, do you know of a more specific example where the science is explained a little further? Granted it will be similar to what we have discussed earlier.

Bubo - thanks, the pic was only an example but your analysis was great info! I have never had a water system but have parts for one not put together, so excuse my ignorance on all this. When I do I would like to use a coolant that keeps it all squeaky clean even if performance is downgraded a little, hence my investigation into using oils...........plasticiser residues is a good one to think about......I am good at removing oil after all the work I have done on cars, but it will involve washing up liquid no doubt :P (oven cleaner also works a treat and is ok on seals for short periods)



From my rough estimations so far water is 20 times better than air and oil is 4 times better, which might be ok as long as I don't try to overclock to 8 Ghz+ :P
 
Last edited:
http://en.wikipedia.org/wiki/Capacitors

Have a look here, got links to dielectric as well, that bit completely bambooozles me even though I covered it in my degree.

I've always been interested in seeing if there was something else that could be used, liquid metals have always caught my eye, but they have other issues as well.
 
I personally think you are underestimating the phaff which will be involved with trying to remove all trace of oil from a radiator, seeing as you can only rely on filling it and shaking it, you can't directly clean it. That alone would put me off the idea.
 
Bubo - I agree with you, resale was only an example but I like to keep my equipment in good order......for cleaning if ever needed could even connect the loop with dispersants etc and run the pump changing the fluid every so often, as long as the pump can take it, but some good soakings of the rad might help, only an idea.....even simply running fresh water through would help........

Jokester - what degree did you do? are you talking about using liquid metal in a cap or cooling system?
 
I did a Masters in Electrical/Electronic engineering, there was a bit of interest a few years ago in CPU cooling with liquid metals but they never really lived up to the hype as they just turned out to be not as good as normal air coolers, were heavy and ridiculously expensive.
 
Yep, as I said, an inordinate amount a phaff I trust you'll agree :)

What gear do you already have? Is it all plexi topped stuff such that you can actually see the surfaces involved?
 
You must log in or register to reply here.
Back
Top Bottom