Water blocks for amd 64

[TerraMex]

Assuming the other side equal and form the assembly, and the two wire pairs, it looks like it's a peltier sandwitch, at least. Double that is. One on each side. It would be interesting to see the middle plates without the neoprene.

Basically it's a chiller. The blocks cool each of the hot sides, both cool sides must be attached to a maze type block? (usually, for maximum effect) . Althou i am intrigued about the noticeably smaller size outlet (inlet?) barb.

( Well, you could just circulate water to a deposit for the main cooling rig. And you dont actually need a very large id (or very large flowrate) with a chiller setup (less restrictions when you eliminate the radiator). But i'm just guessing )

[BillA]

quite correct TM
the chilling chamber has 2 diamond pin matrix bps back to back
-> low flow is the unappreciated key to chiller performance

pHaestus has been sent a loaner to test

[TerraMex]

> low flow is the unappreciated key to chiller performance

That was my initial thought, seeing the barb size (and makes sense), but i wasn't sure .

Looking forward for the review.

And while we're at it ..

http://swiftnets.com/products/MCW20.asp

Looks very good. It was the missing piece.

PS: isnt it...well, big? Almost the size of some of cpu blocks.

[#Rotor]

those are just 4 exposed pads, for grounding the EM shield.... indeed they would have been handy, if they where holes... but they are not..... yet anyway....... SOOOOOO any mobo manufacturers lurking...... hows about it.... make them pads into holes.... your boards will become holy in more that one way

[nikhsub1]

I believe Cathar was experimenting in pelt based water chilling about a 7 months ago. You can see it HERE

BillA, I would have thought for sure that we would have seen COPPER tops out of Swiftech by now?

[siavash_s_s]

copper being the emphasized word *snigger*

[nicozeg]

Originally posted by unregistered

-> low flow is the unappreciated key to chiller performance

Something is missing here: The CPU loop is going to require a specific low flow block and pump that don't exist in the current Swiftech line.

[BillA]

this product was developed for an industrial application, though it can be applied to CPU cooling

you are correct that we do not offer a very small pump such as would be appropriate,
the instructions will address this when the product is announced
- the MCW5000 works just fine as the cooling does not rely on flow regime effects,
effective system insulation is quite important

re COPPER:
a copper top will provide no improvement in cooling
we are not seeing corrosion problems with our aluminum tops (we sell a corrosion inhibitor, eh ?)
-> a copper top would have a huge impact on costs

so where is the benefit to the user ?

#Rotor
boards have been shown with both 2 and 4 holes, and one at least with both (the 2 plus the 4)
in an Inquirer writeup several months ago

[Les]

Originally posted by unregistered
.........
-> low flow is the unappreciated key to chiller performance
.......

Certainly not fully appreciated here, that in a closed loop "low flow is the key to chiller performance".
In fact I am not sure it is correct, will have to think about that one.

[nikhsub1]

Originally posted by unregistered
re COPPER:
a copper top will provide no improvement in cooling
we are not seeing corrosion problems with our aluminum tops (we sell a corrosion inhibitor, eh ?)
-> a copper top would have a huge impact on costs

so where is the benefit to the user ?

Bill we know that a copper top will have no effect on cooling. The reason I ask is that you always seemed very against using aluminum and copper due to corrosion issues. You have voiced this many times. Just thought your influence may have brought a copper top is all.

[Since87]

Originally posted by Les
Certainly not fully appreciated here, that in a closed loop "low flow is the key to chiller performance".
In fact I am not sure it is correct, will have to think about that one.

Got me baffled as well.

I can see that reducing pump heat is a significant issue, but that's not the same thing as low flow.

[BillA]

were there no cost impact, copper or brass would be preferable - no argument from me
but such is not the case

as before:
always use an inhibitor

[superart]

Originally posted by Since87
Got me baffled as well.

I can see that reducing pump heat is a significant issue, but that's not the same thing as low flow.

You want low flow in a chiller, so as to give the water more "time" to get cooled off.

[nicozeg]

Hmmm, but as more time it is cooled, the same happens with the warming on the cpu side; balancing things out. My guess is that heat transfer is more eficient on the wide surface of the TEC's than on the small CPU, so the sweetspot is to the low flow side.

[Boli]

A low flow pump would be better if it were a "screw" or suction pump for these types of situations where water isn't flung around more but relies more on the slow powerful movements to pass the water around the circuit. In much the same way 7v fans work .

Shame most of the pumps on the market now are impeller style.

~ Boli

[JFettig]

I cannot beleve it has happened. Im extremely dissapointed in you.

[superart]

extremely disapointed at who...and for what?

[Althornin]

Originally posted by superart
You want low flow in a chiller, so as to give the water more "time" to get cooled off.

this is the same sort of thinking that says you'd want low flow through a rad, for the same reason. But it seems wrong to me. Higher flow results in same avg time for each "piece" of water to spend in chiller block. Delta T is higher, so more efficient heat transfer.

[Boli]

You only think that because most of the best water blocks today work on impingment where faster flow is best. Yet low resistance blocks such as swifttech and the maze3 will work as well with any flow (even convection as I found out to my surprise one day).

This is of course a never ending debate.

~ Boli

[BrianW]

Actually a restrictive impingment block usually works just as well at low flow, whereas a high flow block requires more flow to be as effecient at transfering heat to the water, and at low flows becomes a rather poor performer.

BrianW

[siavash_s_s]

cant argue with that boli he is an uber geek where as you are just a geek

[Boli]

*Bows to the Uber Geek*

*Beats up Noob for pointing it out*

~ Boli

[superart]

Originally posted by Althornin
this is the same sort of thinking that says you'd want low flow through a rad, for the same reason. But it seems wrong to me. Higher flow results in same avg time for each "piece" of water to spend in chiller block. Delta T is higher, so more efficient heat transfer.

good point, good point, but look at it this way.

In a rad, the deltaT is not very large, in relation to the deltaT of a cooler or instance, so the flow rate of water really doesn't make much of a difference since high flow or low flow, the efficiency will be more or less the same. The reason you would want high flow on a rad setup is because it is better for your specific block, since some blocks perform better with higher flow.

In a cooler, however, the deltaT is much greater, therefore the more "time" the water spends in contact with the cooler, the longer it will be exposed to the "greater deltaT", the more efficient and effective the cooler will be. Because of this lower flow rate, you would probably want to pair the setup up with a block that performs better at lower flow rates.


Am i thinking along the right track?

[BrianW]

Sounds right to me, And I would pair it up with a cascade....

BrianW

[BillA]

nooo, not at all
as it applies to the cooling chamber, so it applies equally to the wb

the cascade would not work so well, maximizing 'h' at a small area will not get it done
review in your mind all of the many many TEC chillers that have been made over the years
what did they have in common ?
how is this one different ?

superart is on the right track, as are several others