WaterBlock/Flow Question

[buttonmash]

Say you hooked up the CPU water block in reverse, say on a Danger Den TDX. Where your Push line was entering the "exit barb", and the Pull line was exiting thru the Accelerator Nozzle.

Would this dramatically affect cooling? Is this crippling the "channels" ability to cicrulate heat?

or

Is this comparable to the minimal temp difference of directional flow for the Rad?

I was in a hurry, we've all been there. I hooked the hoses up in reverse, but didnt notice any temp change. If you guys think its a bad deal, ill correct it.

http://img334.imageshack.us/img334/6255/amd2vx.jpg (56K Users Beware)

[Nebulous]

Actually you're killing the cooling perfomance designed for the block when set up in reverse.

[voigts]

It would be best to set it up correctly.

[Jas]

The TDX and RBX use a Jet to shoot the water onto the base of the block in order to increase turbulance and the waters ability to "collect" more heat. The sam eis true for the Storm blocks, and the Swiftech series of blocks.

Reversing the flow of those blocks completely goes against the manner in which they were designed for optimal heat transfer, and would defaintaley have adverse temperature diferences.

[buttonmash]

I installed the hoses back to 'normal'. In theory you would think there WOULD BE an adverse reaction to reversing the hoses. But, that just isn't the case. The difference of about 1c, and my ambients were the exact same for each test.

Updated Pic (http://img351.imageshack.us/img351/1131/p10100019xl.jpg)

[PoX Freak]

I think what happened is when you hooked it up backwards, you created a pressure differential between the pump, rad, and block, which in turn showed up as a 1º increase in temp due to flooding the block instead of "spraying" the block. My theory is, if you dont have enough pressure to "spray" coolant across the block, you wont see much difference in overall temps.
BTW, that pump will actually flow more if you put it higher up in the system (eg. above the radiator, but below the block). You could even put it higher, but not see much gains in temps.
PoXiE

[capneedle]

I think what happened is when you hooked it up backwards, you created a pressure differential between the pump, rad, and block, which in turn showed up as a 1º increase in temp due to flooding the block instead of "spraying" the block. My theory is, if you dont have enough pressure to "spray" coolant across the block, you wont see much difference in overall temps.
BTW, that pump will actually flow more if you put it higher up in the system (eg. above the radiator, but below the block). You could even put it higher, but not see much gains in temps.
PoXiE

You really think that putting the hydor up higher would be more beneficial? I'm curious as to why you would think that.

[buttonmash]

I got another one for you guys.


Water Stabilization.

How long til after filling/bleeding the hoses with Distilled water/waterwetter should one wait to guage for accurate temps and stability?

Meaning, should you wait til after the water is transparent, and the Waterwetter settled to run tests? Ive tried both ways and was never able to lock down a control.

Would adding "a little" too much WaterWetter have adverse reactions? Say 2oz's.

I recently did a Top Conversion for the TDX(From s478 to s939), I noticed the seals on the accelerator were kind of "chewed gum looking." I also noticed they didnt seem to seal the water in over the "channels." But I never really had this good of a view of the block in action.

Tell me what you know, im a Sponge.

[citronym]

I know that if you are using AS5 and wan't to gauge temps accurately, you should give the seat between the die and wb at least a few days of crunching to allow the AS5 to set. By then, the system should be bled anyway.

IMHO, adding more water wetter than what the bottle recommends is displacing water, and wasting cooling capacity. I honestly don't think adding extra would promote any better temps or anything else.

Just some thoughts...

[Otter]

I think what happened is when you hooked it up backwards, you created a pressure differential between the pump, rad, and block, which in turn showed up as a 1º increase in temp due to flooding the block instead of "spraying" the block. My theory is, if you dont have enough pressure to "spray" coolant across the block, you wont see much difference in overall temps.
That would explain the minimal temperature difference. But how do you reconcile that with the fact that most current jet impingement blocks perform better than standard blocks even with relatively low flow rates?

[PoX Freak]

First things first...
Cap, I say that because when you figure for head loss throughout the system as a whole, and the waterblock is the most restrictive item in the loop, I think the pump would flow better if mounted more closely (vertically) to the most restrictive item (water block). Gravity does pull down on the water flowing through, and i would want to keep the flow high when it runs through the system.

Otter,
Its all in the design of the jet. A larger jet will thus flow more water, allowing for a higher pressure pump, and coincidentally better "spray" over the impingement area.
Have you ever tried to stop the flow of water from your pump? If you did, you would see a pressure rise until the point of cavitation (the point at which the water will not flow anymore, and begins to "cavitate" inside the pump rather than flow through it.
Ultimately, you want to find the optimum pressure before cavitation and use the jet that comes closest to the flow rate and pressure you want to achieve, or just under the maximum the pump will flow.

Ultimately, you would want to use a resivor-type setup. Because in a closed loop system, the pressure would be counter-acted by the draw on the feed to the pump.

[Otter]

Yes, Pox. I should have said "pressure" instead of flow. Sorry for the confusion.

My point is that a jet impingement block is not well designed to operate without the jets. If I understand you correctly, your theory is that there really is no jet spray at low pressure, hence it doesn't matter much when the water goes the wrong way. But if there is no cooling benefit to the restrictive jets with the water going the right way, how does a TDX block kick sand in the face of a good maze with the same weak pump?

I don't understand Buttonmash's experience. It seems to me the block should perform very poorly with the jets providing lots of restriction and little cooling benefit. If that worked, you could make a good low-head block by putting a cover on a rough bit of copper and kinking the outlet hose.

...Unless it really doesn't work very well either way. Buttonmash, what are your CPU temps? Maybe your system needs a few tweaks to benefit from having the hoses connected the right way around.