Micro Channel "White Water" Block

[nikhsub1]

hey man, plug the center barb! try that out! i know cathar tried it, but see what diff it makes..

Not even an option. One of the biggest performance boosts is using the center barb with the nozzle that creates the "jet impingement". I'm KNOW it will be worse side to side.

 

[JFettig]

well thats obvious, but how much? alrite if you dont want to,

you should invest in an inline probe...

 

[nikhsub1]

I redid my plumbing Thursday night. I made the exit tubes long - like 18 inches each. To make the story short, my temps really suffered with long exit tubes, 2-3C. I remounted the block 3 times cause it didnt make sense. I just now put the setup back to the original way I had it and voila, temps back to where they were in the first place. Weird. Cathar if you see this, have you noticed this too? Could be that if I had a stronger pump (Iwaki) this might not have happened. Somehow I think that the pressure in the system is optimized when the exit tubes are as short as possible.

 

[Freeloader]

Trying to get accurate results using case and cpu temps that read in 1C imcrements is alot like trying to pull a pubic hair off a mosquitoe's balls, using boxing gloves. Difficult to say the least.

A 40C cpu and 28C case temp could be 40.9 and 28.0, or 40.0and28.9. That's almost a 2C difference. Compare that to another test with a possible 2C difference and you get the picture. Then there's fluctuating air currents in the case, T.I.M variance, voltage variance and who knows what else.

I think your results are clearly showing that your equipment isn't up to the task.

 

[Cathar]

I redid my plumbing Thursday night. I made the exit tubes long - like 18 inches each. To make the story short, my temps really suffered with long exit tubes, 2-3C. I remounted the block 3 times cause it didnt make sense. I just now put the setup back to the original way I had it and voila, temps back to where they were in the first place. Weird. Cathar if you see this, have you noticed this too? Could be that if I had a stronger pump (Iwaki) this might not have happened. Somehow I think that the pressure in the system is optimized when the exit tubes are as short as possible.

No, I have not noticed this. There are in fact a few people running with exit tubes 50cm (1.5') long each and still seeing the same 5C+ results.

I think Freeloader has the answer here. Trying to guage the improvement from case temperature alone is sorely stretching the realms of repeatable testing (and very big understatement).

I suggest that if you wish to "play" then you would do well to get a thermometer to measure the radiator air-intake temperature to around a 0.1C resolution (which will probably still have a +/- 0.2C accuracy), and as well for the water temperature.

Then using the CPU's 0.5C resolution (~+/- 0.5C accuracy) you'll still be doing tests with a +/- 1.0C accuracy, but "good enough" to detect the size differences we're talking about.

Sure, it's not BillA's test lab, but for a small investment it'll give you a much better idea of what's going on than relying on the case temperature diode which really is an extremely poor indication of the "ambient" temperature.

 

[Skulemate]

In theory it should be best to have the two return lines run all the way to the reservoir or rad, as this will reduce the amount of resistance caused by the tubing itself. Of course, doing this will make it harder to keep the resistance the same through each return, and will cost more in tubing.

 

[nikhsub1]

I know I'm not crazy. Maybe it was just the added tubing that contributed to the increased temps. Normally I run like 3ft of tubing total or less. Bahhh, maybe I need to just stop messing with things!

 

[JFettig]

lol, i hate it when i do stuff like that in your situation, less tubing would prob be best, if you had a dual inlet rez, longer tubing would prob be best...