Have you seen this block?

[SiGfever]

Looks darn sexy...



http://translate.google.com/translat...language_tools

[Fragger56]

ooo shiny

too bad they don't have a x1950 version

also, wouldn't the steel tubes connecting the VRM cooler to the main block increase the chance of galvanic corrosion?

[Captain Slug]

Originally Posted by Fragger56

also, wouldn't the steel tubes connecting the VRM cooler to the main block increase the chance of galvanic corrosion?

Nope. Most grades of stainless steel are innocuous.

[BGP Spook]

There should still be some galvanic corrosion even if it is stainless steel.

There are ways around this but I tend to think they have not instituted any of them.

See here.

Galvanic corrosion occurs when two dissimilar metals are placed in contact with each other in the presence of an electrolyte, such as salt water, resulting in the unintentional formation of a galvanic cell and concomitant chemical reaction of the metals involved. There are several ways of reducing and preventing this form of corrosion. One way is to electrically insulate the two metals from each other, for example by using plastic or fibre washers to separate steel water pipes from copper-based fittings or by using a coat of grease to separate aluminum and steel parts. Another way is to keep the metals dry and/or shielded from ionic compounds (salts, acids and bases), for example by encasing the protected metal in plastic or epoxy. Another method, called "cathodic protection", uses one or more sacrificial anodes made of a metal which is more active than the protected metal. Metals commonly used for sacrificial anodes include zinc, magnesium, and aluminum. Finally, an electrical power supply may be connected to oppose the corrosive galvanic current. (see Impressed-Current Cathodic Protection)

Unfortunately my chemistry has gotten rusty recently and I don't remember where to find the data tables at the moment.

Give me a bit and I will see what I can come up with.

[ibmkg]

Keeping the overall potential to zero and chrome/nickel plating eliminates all risks. Chrome/Nickel can withstand extra ordinary conditions for years. Example would be:

A car's header (exhaust manifold, Nickel plated can withstand >100 C)
Bumper Chrome can withstand exhaust emissions, rain, snow, sun heat etc

Advanced methods include anodizing the metal. Example:

Military tactical equipment is anodized to withstand harsh conditions. Usually Al is anodized to a state that it changes its color to green (forms a protective layer).

[jman1310]

Originally Posted by ibmkg

Keeping the overall potential to zero and chrome/nickel plating eliminates all risks. Chrome/Nickel can withstand extra ordinary conditions for years. Example would be:

A car's header (exhaust manifold, Nickel plated can withstand >100 C)
Bumper Chrome can withstand exhaust emissions, rain, snow, sun heat etc

Advanced methods include anodizing the metal. Example:

Military tactical equipment is anodized to withstand harsh conditions. Usually Al is anodized to a state that it changes its color to green (forms a protective layer).

nickel plating is tedious, but affective
anodizing works but extremely well known for causing problems - just the tiniest scratch or imperfection...

it would be much simpler to use copper pipe in the first place... DUH!

[SiGfever]

Originally Posted by jman1310

nickel plating is tedious, but affective
anodizing works but extremely well known for causing problems - just the tiniest scratch or imperfection...

it would be much simpler to use copper pipe in the first place... DUH!

Since the tubes are the only support for the small block, the stainless (if it is very high grade) is a better choice unless they go to nickel plated hard drawn copper.

[BGP Spook]

There is another option.

They could use a nylon seal and sleeve to prevent the steel from every contacting the water, for example.

As I said, I doubt they instituted any of them as they are troublesome.

[ricecrispi]

Nickel chrome plating works well but how does gold plating prevent galvanic corrosion. I seen it on zalman blocks and recently in koolance. From what I understand it uses copper nickel gold sandwich. http://www.engineersedge.com/galvanic_capatability.htm

[bigben2k]

http://www.engineersedge.com/galvanic_capatability.htm

Keep the potential difference low, and it'll last much longer.

Nickel to copper -=> 0.05 very good.
chrome to copper -=> 0.25 mid-range, but should be good enough for this application.

Aluminium to copper -=> 0.40 bad

[jaydee]

I would be more interested in the performance. Sure looks like a basic cross drilled block to me. Not what I would want on a modern $500+ vid card.

[davidzo]

i gues its the basic heatkiller design, no need to find a new one since it is really good already:

The Heatkiller is a top CPU block with lower flow resistance than pindesigns such as the Aquaextreme MP05 oder Nexxos and lower than jetdesigns as the storm. It is on par with the newest NexXxosXP models, on bigger dies it perfoms even better. Really i think its probably one of the best solutions possible for the big G80.

Radical53 from X'S compared GPU-X fullcover Waterblock with swiftechs MCW60, EK fullcover block and alphacools offering and some tzoher all for the 7800/7900/x1800/x1900 a while ago in german hardwareluxx printed magazine. The Heatkiller came out on top. The pressuredrop was by far the lowest in the test (best flowrate) while performance was first place on par with the alphacool offering (which had a ten times higher pressure drop).

too bad they don't have a x1950 version

@Fragger56: Guess - they actually have a X1950 compatible GPU-X. Its selling over here in germany since a year for aroung 65€uros.

[Fragger56]

Too bad im not in germany then :P
The 8800 version looks much better though, I think, but the x1950 ver looks 'safer' if you know what i mean.

[ricecrispi]

Originally Posted by bigben2k

http://www.engineersedge.com/galvanic_capatability.htm

Keep the potential difference low, and it'll last much longer.

Nickel to copper -=> 0.05 very good.
chrome to copper -=> 0.25 mid-range, but should be good enough for this application.

Aluminium to copper -=> 0.40 bad

I understand how the nickel and chrome works to reduce galvanic difference but difference between aluminum > gold is .75 so difference is much higher. Gold > nickel is .35 and nickel >copper is very very small.

I'm surprised that block had lower pressure drop than mcw-60 but i'm not surprised it outperformed it and all the other blocks with that design. Not surprised with alphacooling being restrictiveness but DAMN!: