Overclock.net › Forums › Cooling › Cooling Experiments › 12 Feet Under: 1000 Square Feet of Geothermal PC Cooling
New Posts  All Forums:Forum Nav:

12 Feet Under: 1000 Square Feet of Geothermal PC Cooling - Page 22  

post #211 of 340
you sir are a badass

with plenty of free time
Smee 2.0
(14 items)
 
  
Smee 2.0
(14 items)
 
  
post #212 of 340
subbed.
I was considering doing this sort of thing except with my pool but then university holidays ended and i have no free time
BLU
(13 items)
 
  
CPUMotherboardGraphicsRAM
fx-8120 990FXA-UD5 Dual Gigabyte GTX560 Ti  G.Skill Ripjaws X 1600MHz CL8 8Gb 
Hard DriveOSMonitorPower
120gb OCZ Vertex 2, 1 + 2 Tb WD Green Windows 7 Pro Samsung 2233SW 21.5" + 2233 21.5" + ASUS 20" Corsair HX-750 
Case
Custom server case 
  hide details  
BLU
(13 items)
 
  
CPUMotherboardGraphicsRAM
fx-8120 990FXA-UD5 Dual Gigabyte GTX560 Ti  G.Skill Ripjaws X 1600MHz CL8 8Gb 
Hard DriveOSMonitorPower
120gb OCZ Vertex 2, 1 + 2 Tb WD Green Windows 7 Pro Samsung 2233SW 21.5" + 2233 21.5" + ASUS 20" Corsair HX-750 
Case
Custom server case 
  hide details  
post #213 of 340
Thread Starter 
Quote:
Originally Posted by Slink View Post
Oh, THAT'S where the insulation is going. Oops!

P.S. Good progress/updates.


I know there's too much for anyone to keep track of or visualize. Myself included with all these fittings.

I've really made a mountain out of a mole hill. When it comes down to it, this project is just a hole in the ground with some piping as a radiator.

It should be finished within 36 hours though! I'll get a better batch of pictures collected for the initial completion write up.

Getting OWFS setup for the data logging and graphing isn't going to happen right away. I'll probably want that going before doing the the loop thermal capacity testing so I don't have to manually record the temperature for hours and hours. Next week.

I'm really looking forward to the temperature results. I'm now hoping the cooling is too good and causes condensation this summer. It'll give me an excuse to dump the heat from the spare bedroom into the loop water before it reaches the PC(s). That room's closet is right above the utility closet so it'll be simple to connect the two loops through plated heat exchanger and rock on. When the (probable) de-humidifier is running, or the dew point threshold looks safe, I still could turn that loop's pump off and enjoy the cooler loop temperature if I wanted.

Anyway, that would be the best use of too much cooling capacity. Unsightly heater cores and the fan noise won't matter up there either.

Plus my server's hard drives up there will thank me. No failures yet because good airflow, even if its hot, keeps the drives under 50c. I have a full mirrored backup just in case though.
post #214 of 340
Thread Starter 
I think I'm ready for the big day tomorrow. It might be a bit hard to sleep tonight!



Yes, that's another 100' roll of 160 PSI piping. The previous 100' for the supply/return line wasn't enough! If I'd known I was going to locate the ground loop so far down the hill I would've bought the 500' roll to begin with ($120 + S/H). The colder summer ground temperature should easily be worth the small pressure loss. (There's also room and should be pump capacity to install another buried coiled loop to the right of this one, if desired. )

Because of the higher than expected restriction, and the lower than expected power consumption of the Iwaki, I'll probably connect the PCs to the loop through plated heat exchangers. Doing that and cracking open the bypass valve should avoid all potential condensation problems.

My loop last summer was more than 5c higher than ambient when loaded so its possible to drop more than 10c off the intake water temperature and stay above the dew point. Going slightly below it isn't immediate doom and gloom either, but that area could be viewed as the safety wiggle room.

A couple pictures:

The pressure gauge and filter valve installed. Now I can remove the filter without spilling out the top water content of the reservoir.



Here's some of what the filter had in it. I wasn't very careful when working in the crawl space. Without the filter I would've had to cap all the tubing when moving it and have kept all the fittings in plastic bags.



Anyone care to guess what the ground temperature will be? 12 feet under, in wet higher water table soil?

The constant ground earth temperature is supposed to be 58 in the coastal area here. So I'm expecting to see a temperature slightly below 50 tomorrow based on these:





"A minimum delay of five days between loop completion and test start-up is
recommended in formations that are expected to have low conductivity [< 1.0 Btu/her-ft-°F (1.7 W/m-°C)] and three-days for other formations."

It looks like the final settled ground temperature will take a little while to get, but I'll have IR readings of the excavated soil tomorrow.

We're going to risk not using any sand as a dressing because of how wet the soils going to be. The coiled loop will need to be lowered immediately after temporarily scooping the muck out of the trenches finalized depth. Once it's down there, its going to float or potentially be raised by mud sliding in from walls. It'll need to be immediately backfilled from a pile of pre-screened soil to keep it in place. All the water should prevent air filled pockets of soil from remaining around the coils. This installation definitely won't need to be filled from a garden hose.

That's the expectations anyway. Only 14 more hours and we'll find out.
post #215 of 340
i will be so mad if the bucket falls off the digger after just one swing and this gets held up for another week wel ok i know the bucket wont fall off but piston o ring blow out are comon when pulling out supper soggy wet soile or trying to move huge bolders out that are nicely stuck

i know in one week i managed to blow out 2 seals on 2 difrent pistons on the boom but it was a crappy old 1/4ton bucket deer basicaly a farm tractor that should have been retierd 10 yrs ago and its still digging
Be
(13 items)
 
  
CPUMotherboardGraphicsRAM
PhenomII940 X4 3.0Ghz BE asus m3a78-t BFG GeForce GTX 260 Overclocked MAXCORE patriot 6400 
Hard DrivePower
hitachi 1TB ultra 750 heavy duty 
  hide details  
Be
(13 items)
 
  
CPUMotherboardGraphicsRAM
PhenomII940 X4 3.0Ghz BE asus m3a78-t BFG GeForce GTX 260 Overclocked MAXCORE patriot 6400 
Hard DrivePower
hitachi 1TB ultra 750 heavy duty 
  hide details  
post #216 of 340
Thread Starter 
The pits dug and the coils are in the ground. The final connections will have to wait until tomorrow though.

We started three hours late because of an emergency errand this morning. So I didn't finish laying, insulating, and trimming the return/supply lines. I did manage to clean out most of that little trench though. I need to head to Lowe's now to get some PVC pipe to repair a broken water pipe.

The ground was a quagmire so the depth varied from 3-7 feet. The highest point was at the far end where the warm water will go first, and the deepest is where the two trenches meet. Measured soil temperatures were in the low 40s Fahrenheit. My temperature probe unfortunately isn't responding. I checked it before siliconing the connections and securing it the pipe, but not after. Oh well, water in and out are all that matter anyway.



post #217 of 340
Looks awesome dude. Fail on the temp probe though.
X-Caliber
(13 items)
 
  
CPUMotherboardGraphicsRAM
Intel Core 2 Quad Q8300 @3.0 Ghz Gigabyte EP43 UD3L ATI Radeon HD3850 512MB 800/950 4x1 GB 800 mhz DDR2 (Dual Channel) 
Hard DriveOptical DriveOSMonitor
160 Hitachi HDD LG Super-Multi Drive with SecurDisc Windows 7 - Ultimate (x86) Samsung SyncMaster 2043SWX 20" 1600x900 
KeyboardPowerCaseMouse
Logitech G15 Gaming Keyboard rev. 2 460 watt Coolermaster Coolermaster Elite Razer Deathadder 
Mouse Pad
Steelseries QcK 
  hide details  
X-Caliber
(13 items)
 
  
CPUMotherboardGraphicsRAM
Intel Core 2 Quad Q8300 @3.0 Ghz Gigabyte EP43 UD3L ATI Radeon HD3850 512MB 800/950 4x1 GB 800 mhz DDR2 (Dual Channel) 
Hard DriveOptical DriveOSMonitor
160 Hitachi HDD LG Super-Multi Drive with SecurDisc Windows 7 - Ultimate (x86) Samsung SyncMaster 2043SWX 20" 1600x900 
KeyboardPowerCaseMouse
Logitech G15 Gaming Keyboard rev. 2 460 watt Coolermaster Coolermaster Elite Razer Deathadder 
Mouse Pad
Steelseries QcK 
  hide details  
post #218 of 340
i cant wait for the results from this badboy.
Lurking Ninja
(13 items)
 
  
CPUMotherboardGraphicsRAM
Intel Core i7 920 D0 3849B203 4.0GHz/1.164 Evga X58 Classified E759/Koolance X58/BIX 120 TRI SLI GTX 285s.** 6GB 3x2 OCZ Gold 1600mhz 
Hard DriveOptical DriveOSMonitor
2x WD3200AAKS (RAID0), 1tb AALS Samsumg Wrigtmaster Sb 204/182 Windows 7 Ultimate X64 Gateway HD2401 24" 
KeyboardPowerCaseMouse
Logitech G15 V1 Corsair HX1000W Modded Rocketfish Logitech MX Revolution 
Mouse Pad
soft cotton? 
  hide details  
Lurking Ninja
(13 items)
 
  
CPUMotherboardGraphicsRAM
Intel Core i7 920 D0 3849B203 4.0GHz/1.164 Evga X58 Classified E759/Koolance X58/BIX 120 TRI SLI GTX 285s.** 6GB 3x2 OCZ Gold 1600mhz 
Hard DriveOptical DriveOSMonitor
2x WD3200AAKS (RAID0), 1tb AALS Samsumg Wrigtmaster Sb 204/182 Windows 7 Ultimate X64 Gateway HD2401 24" 
KeyboardPowerCaseMouse
Logitech G15 V1 Corsair HX1000W Modded Rocketfish Logitech MX Revolution 
Mouse Pad
soft cotton? 
  hide details  
post #219 of 340
Thread Starter 
Quote:
Originally Posted by G|F.E.A.D|Killa View Post
i cant wait for the results from this badboy.
It'll take a few days for the pit to settle a bit, and get the most accurate results. Tomorrow will be both sunny and warm, perfect for starting to dry the surface out.

Thanks to the assistance of my neighbor, we were able to connect/clamp all the piping before it got dark. The return/supply piping still needs to be insulated, buried in the mini trench, and shortened a bit. I did fill and bleed (fun times) the loop to leak test. So far it's looking good.

I've installed the extra temperature sensor into a conduit pipe and will jam it 5-6 feet into the ground tomorrow. The grounds will still be so soft that walking on it isn't possible. I tested it this time and it's working fine with same ethernet cable. :up:

There's over 1gb of pictures to dig through, so more later... I still have things to finish and photograph.

Oh, it's kind of odd running furmark on the stock 5970 and watching the temperatures immediately jump to 13c, and then flatline.
post #220 of 340
Thread Starter 
Stock 920, 1.15v load. The is update removed my profiles. I'm going to run the OCCT GPU+CPU test overnight but the ambient air will cool the exposed pipes and skew the results downwards.



New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Cooling Experiments
This thread is locked  
Overclock.net › Forums › Cooling › Cooling Experiments › 12 Feet Under: 1000 Square Feet of Geothermal PC Cooling