Copper and Alu

Discussion in 'Extreme and Water Cooling' started by lukepankau2002, Sep 14, 2006.

  1. lukepankau2002

    lukepankau2002 Member

    Joined:
    Dec 12, 2004
    Messages:
    79
    Location:
    Goldcoast - kingscliff
    hey dudes.... umm a question .. in my loop i have 2 alu oil coolers as radiators... and 2 copper gpu blocks...

    is this bad that i have both copper and alu in the same loop?

    why are they bad together?

    and can i still have my loop with copper and alu ?

    will i get like a chemical reaction or sumthin???

    help needed !
     
  2. OldnBold

    OldnBold Member

    Joined:
    Oct 5, 2004
    Messages:
    4,261
    Location:
    Sunshine Coast
  3. OP
    OP
    lukepankau2002

    lukepankau2002 Member

    Joined:
    Dec 12, 2004
    Messages:
    79
    Location:
    Goldcoast - kingscliff
    but can i use both alu and copper if my coolant is anti corrosive ???

    ... duznt the coolant act as a inhibiter or sumthin ??
     
    Last edited: Sep 14, 2006
  4. 2xCPU

    2xCPU Member

    Joined:
    Jan 21, 2005
    Messages:
    3,328
    Location:
    Saigon
    You'll get corrosion due to dissimilar metals when you have an electrical loop for induced currents to flow.
    Ideally, all metals used should be the same type. You need to be careful with different metals, but it not difficult. (think of a car engine - Alu head and water pump (steel impeller), cast iron block and copper radiator).

    What you need to do is:
    - electrically isolate the radiator from the case (so there is no return path).
    - use distilled water.
    - use a corrosion inhibitor at the suggested concentration.

    2.
     
  5. banshee

    banshee Member

    Joined:
    Sep 8, 2001
    Messages:
    1,766
    Location:
    Springwood, West of Hell (Sydney, actually)
    I use premix coolant with a copper block and alloy rad. Rad is mounted to case, but the block is isolated electrically. With the inhibitor in the coolant, I haven't had and problems, so far...
     
  6. 2xCPU

    2xCPU Member

    Joined:
    Jan 21, 2005
    Messages:
    3,328
    Location:
    Saigon
    Ok, that will achieve the desired result too.
    I was thinking of the Alu spreader on my xeons when I wrote that (which I assume is grounded).
    So long as one of them is electrically floating, it will be fine.fine.


    2.
     
  7. blakeyboysmith

    blakeyboysmith Member

    Joined:
    Dec 11, 2004
    Messages:
    379
    as long as you use appropriate (research what ones) inhibitors you shouldn't have any serious problem.. there are lots of ways you can prevent and inhibit corrosion but why stress when this is usually sufficent.

    you can get corrosion between the same metals based on different potentials (same sort of thing as dissimilar metals) but it is not really a concern for our application just a useless bit of information to add the the great collection we have on our forums.
     
  8. Kiba

    Kiba RIP Lest we forget!

    Joined:
    Jun 21, 2006
    Messages:
    685
    Location:
    Nadi Ali, Helmand
    afaik all engine cooling additives have a degree of corrosion inhibiting properties. anything ethelyene glycol should be corrosion inhibiting due to the fact that it makes water non polar, instead of it being (bi??) polar

    SAFETY EDIT do not mix any corrosion inhibitors that contain amines with ethelyene glycol it releases dangerous gases, toxic if inhaled.
     
    Last edited: Sep 15, 2006
  9. OP
    OP
    lukepankau2002

    lukepankau2002 Member

    Joined:
    Dec 12, 2004
    Messages:
    79
    Location:
    Goldcoast - kingscliff
    ok soolo how do i electrically isolate my components of my loop ?
    ... sorry this is my first time w/cing ....tad nooby
     
  10. tin

    tin Member

    Joined:
    Jul 31, 2001
    Messages:
    6,420
    Location:
    Narrabri NSW
    Mount them using something insulated... Screw them to platics or whatever. The water is one "lead" and you need to remove any way for the electricity to get back to where it started.

    Also, re Kiba's safety note, I'd add to that "If you can't pronounce and figure out the base ingredients, don't mix them."
     
  11. OldnBold

    OldnBold Member

    Joined:
    Oct 5, 2004
    Messages:
    4,261
    Location:
    Sunshine Coast
    This is correct. However anyone who has played with car radiators and face it WC setups are car radiators will know that particles drop into the bottom of car radiators and need to be flushed.

    The reason for this is that even though modern cars use coolants (the same as we do) the mix of metals in the systems produces deposits. These deposits various known as scale etc, are corrosion.

    Some metals "like" being together (brass and copper come to mind). A lot don't such as the solution the OP is currently using. If the OP insists on using it he should be prepared to bleed his system monthly, including a radiator flush. The radiator will be the first thing to go closely followed by the pump.

    For those that may post here saying I have been running copper and aluminiun together for years good luck to you. I say it is stupid.
     
  12. 2xCPU

    2xCPU Member

    Joined:
    Jan 21, 2005
    Messages:
    3,328
    Location:
    Saigon
    How often do you flush and refill your cars cooling system? Once a Month?

    I've never met anyone who changes their cars coolent even once a year - it's the same coolent, the same water and the same metals.
    A PC's environment is much less corrosive that a car.

    By all means, keep an eye on the fluid looking for scale and deposits (indication of a problem), but we're not talking voodoo here.

    The each of the suggestions above should stop corrosion. Combining more than one will make double or triply sure the OP doesn't have a problem.

    1. Distilled Water: No contaminants to produce scale and not electricly conductive - Result: No corrosion.
    2. Automotive Anti-corrosive additive - Result: No corrosion.
    3. Electrical Isolation: No return electrical path for an electro chemical corrosion cell. - Result: Greatly reduced corrosion.

    Can you justify your opinion?

    2.
     
  13. gelfling

    gelfling Member

    Joined:
    May 22, 2004
    Messages:
    153
    Location:
    sydney
    I'm in a similar spot to the OP, I've mounted an alu res and a copper rad IN my case. The res is anodised, I've got nulon ultra and demineralised water but the 2 are mounted in my case, so I would take it that both the rad and the res are earthed to my case yeah? Sooo if I used some plumbers tape and rubber washers on screws to the res so that there is no metal to metal contact from the res to the case I should be as OK as I can be?

    blocks are copper and brass btw
     
  14. Kiba

    Kiba RIP Lest we forget!

    Joined:
    Jun 21, 2006
    Messages:
    685
    Location:
    Nadi Ali, Helmand
    all this talk of electrically isolating compnents concrens me a little. first we need to know why we want to do it. reason one; to reduce the effects of ion transfer in the working fluid, and reason two; to reduce any sort of charge bluiding up on/in the water cooling components, which could possibly discharge to a pcb.

    using a car as an example, we do this by having an earth wire, botled to the engine block, which leads to the chasis. this will ground the engine block, mainly to allow the sparks plugs to fire but also making sure no charge is held in/on the engine block. this helps reduce the flow of electrons from a more reactive metal to a less reactive metal. in our case on the computer watercooling, we would want to ground metalic components to the case. this can be done using a wire connected to either the copper cpu block or the radiator, and connecting the other end to the case. it should not realy matter which one as if one is earthed as the ion flow will be reduced throughout the whole system.

    if you want to take this even further, you should then earth your case to the earth wire from the mains wall, as it is a more substantial earth than you case alone.

    this is all marginal however, and, as always, it is reading into the problem too much. using car cooling adative is more than enough to reduce the ion transfer effect for our home computer use.

    please correct me if i am wrong as the question confuses me a little.
     
  15. hugh50935

    hugh50935 Member

    Joined:
    Feb 3, 2006
    Messages:
    21
    Only if they're next to each other.

    Has anyone actually seen & photographed any major galvanic corrosion in cooling setups?

    I get the impression that it's over exaggerated.


     
  16. MMD

    MMD Member

    Joined:
    Aug 4, 2005
    Messages:
    107
    galvanic corrosion

    Quote's from the page :-
    "The problem is that it's anodized aluminum… and through the years I noticed my water kept losing the UV color and the hoses kept on getting black. I drained the system a few times and refilled it with distilled water and some UV pigment, but within a month it went black again, so that’s when a little light bulb started to show above my head."

    "The damage to the GPU block was extensive and not usable in a system anymore, I cleaned the older blocks with some Hydrochloric Acid, so those where still useable, so that just goes to show that different metals inside a watercooling system = BAD news."
     
    Last edited: Sep 18, 2006
  17. 2xCPU

    2xCPU Member

    Joined:
    Jan 21, 2005
    Messages:
    3,328
    Location:
    Saigon
    No, it's not to reduce the "effects of ion transfer ..", it to reduce the ion transfer.


    The static charge on a metal component will be insignificant with the water (however treated) able to discharge it easily.

    And the alternator wouldn't work .. and the starter motor wouldn't work .. etc.

    By providing the "return" path, this allows the flow of electrons.
    Most modern cars have Alum radiators that 'float'. i.e. there is no ground wire to complete a galvanic cell.

    It's ok to ground ONE component in the cooling system, but more will cause a potential ground loop.


    I'd hope the case is already earthed through the PSU. My PSU has 4 screws bolting it to the case making a good connection - are you suggesting this is not adequate?

    Probably yes, but knowing there are dissimilar metals, basic precautions are cheaper than replacing components.
     
  18. 2xCPU

    2xCPU Member

    Joined:
    Jan 21, 2005
    Messages:
    3,328
    Location:
    Saigon
    - What is your second metal?
    - How were the components mounted (electrically connected to the case or isolated)?
    - Why didn't you use an anti-corrosion additive?
    2.

    Edit:
    This makes it sound like the UV dye was part of the reaction and probably caused your problems.
     
    Last edited: Sep 18, 2006
  19. Kiba

    Kiba RIP Lest we forget!

    Joined:
    Jun 21, 2006
    Messages:
    685
    Location:
    Nadi Ali, Helmand

    woah woah, thats a big asumption. i never said that. for unwavering results to reduce inconsistancies in results, im suggesting to connect direct to the earth. but this is over kill as is most of the issues in the post. i dont 100% know if that guy who had his vga block eaten, was using a corrosion inhibiting product in his circuit. if he was using distilled water, than oviously he would get terrible corrosion. using glycol, i find that i personally have had no problem for over 6 months with.
     
  20. 2xCPU

    2xCPU Member

    Joined:
    Jan 21, 2005
    Messages:
    3,328
    Location:
    Saigon
    Yes, I understand your point - the single point earth ensures the components are at the same potential.
    This works well with similar metals, but may create the return path when dissimilar metals are used.
    The other situation the single point earth will help is when external voltages are applied to one component (I suspect this is what caused the water to turn black).

    I'm pleased to hear your system is holding up well, for interest, what metals are you using? (glycol is definitely a 'first choice' additive).

    2.
     

Share This Page

Advertisement: