The Hole Geothermal Loop 56K Warning!

[Naja002]

Months back, I was getting bored and looking for a project…a PC cooling type of project. I started seriously looking at putting copper pipe in a crawlspace and a number of other things. I ended up going with this seasonal dual car rad setup:

Yes, Another Car Radiator Thread....Major 56K Warning!

I ended up getting into “crunching” (Distributed computing) Boinc/WCG, F@H, GpuGrid, etc. Now I run 4 OCed rigs and 4 OCed Gpus 24/7. The entire setup sucks about 1,200-1,300 watts out of the wall. During the summer the heat is ridiculous—even with pumping a lot of the heat outside. It got bad enough that I purchased an 8K btu a/c right before the weather turned cold. Needless to say, my electric bill sky-rocketed. But now that system is providing my winter heat via the indoor rad…

...I have not turned my heater on so far this year (temps as low as 30F/-1C with highs in the 50’s-60’s, but as low as the 40’s). The ONLY supplemental heat comes from a 1K watt $20 space heater from wally-world. When temps outside start dipping into the 30’s it’s in the low 70’s in here ( ~72F with 30-35F outside). Additionally, I don’t run it everyday and only run it a few hours through the coldest part of the night…some nights. Not much at all.

So, what does any of that have to do with “The Hole” ? Honestly, it’s just a little background to help you understand what’s going on here and how the goals have come about…..With the way the economy is and has been, things have reached a point where I would have to choose: Suffer, shutdown the PCs or do without heat. Now I don’t have to choose. My electric bill last summer kept climbing and climbing and climbing. With this seasonal car rad setup, my electric bill (after doing the math) decreased ~$40. That’s with a 20% rate increase. Its winter folks—everyone’s bills are going up. Mine went down! And its going to go down further. With this Geothermal Loop I expect to not have to choose this coming summer, and with the way things are looking atm….I’m not going to have to choose!


So, on with the show:

This is what I did, how I did it, why I did and what I would change—Yes, If I had it to do over….I would probably do it a bit differently:

First I had to decide what kind of pipe to use for the exchanger and why: PVC, HD polyethelyne, copper, etc. Then what size for maximum surface area v. cost and all that good stuff. Ultimately, I went with 3/8” copper refrigeration coils…50’ x 4 = 200’ total. And that is one part that I might change, if I had it to do over:

My original intent was to dig up an old plant bed, dig deeper and install it into there. But after thinking for a while, I came up with a number of reasons why I really didn’t want to do that. So, I sorted out a new location and started the hole from scratch:

8’ x 5’

Then the Real work began….sheesh, what a job:

By the end of the first day I reached ~22” depth. Now keep in mind that this is clay and got harder and harder the further down I went. Some of it was like trying to dig through concrete:

 

[Naja002]

So, bright and early the next morning, once again all wired-up on caffeine, I went back to work. I dug out about 25% of the next layer and decided to check something out. At ~22” depth I drove the shovel in, cracked the ground and inserted a temp probe….this is what I found:

Then at ~28” depth I did the same thing:

The increase in ground temp directly indicated to me that I may have reached temp stability. The frost line here is only 12-18” (less then 18&#8221. Ground temp stability is reached shortly below the frost line….so, this makes sense.

Finally, after digging the hole I checked once again and several inches deeper….the temp was the same:

Understand that when I did these “tests” I just cracked the ground, inserted the probe as deeply as possible and got everything closed back up. Then waited for the temp to stabilize.

These tests indicate to me that I reached ground temp stability…at least for this time of year. Through out the winter this spot is shaded 24 hrs/day. In the summer it will get a few hours of sun, but not much. That’s one of the reasons that I chose this particular spot. As long as the temps don’t exceed ~68F….I expect to have almost free air-conditioning!!

So, by the time I got done digging the hole….this is what I had:

In the top left corner of the 1st pic above you can see where I dug out for the pass-thru into the crawlspace. That pass-thru is ~12” deep.

Just a pic of the final ground temp test and digging the pass-thru:

I had decided to put copper in the ground, but make the “run” with pvc. Part of the rationale of this was cost and part was thermal conductivity. Copper = ~401, PVC = ~0.19. I could have gotten picky about making the run to the hole out of copper (warmest water) and all that, but I decided to keep it simple, so I made everything out of pvc, except the heat exchanger in the ground (copper). Anyway, the hole and pass-thru where done, so now it was time to start the plumbing. Initially, I was going to do this a bit differently, but, ultimately, decided on what’s pictured below:

 

[Naja002]

As I’ve said in other posts….I like to build plumbing from opposite ends and from the middle to the ends….meet in the middle so-to-speak. Crazy? I don’t know, but that’s how I do it! So, I started at the pass-thru, but the first thing that I did was install a temp probe. I picked up this indoor/outdoor thermometer for ~$9 at lowes:

It has a 3m (10&#8217 “outdoor” temp probe on it…..so, I threaded that through some ½” pvc as follows:

Got out the construction adhesive/caulk:

Pulled off the sticky paper and stuck it to the inside of the pipe:

Then filled the opening with caulk:

Glued the cap on (all of the pvc was glued together):

And let it sit for a bit:

Then covered it up:

I dug out an extra 2-3” at that spot in order to insert the temp probe. Heat rises. I want the temp of the ground…not necessarily “the pit”. Plus the end where it is buried is the cool end return. Hopefully it will work out as planned….it should, but we will see this summer.

After sitting buried for a bit this is the result:

 

[Naja002]

So, on with the rest of the plumbing. Again, starting in the middle. 1 ¼” pvc pipe was chosen for reduced frictional loss:

At this point I had pretty much decided to call it a day. But I had too much to do in too little time, so I went a bit further.

I decided to take the pit plumbing to the manifold start:

Took a soil sample in order to determine the soil’s pH. Got it setup for later testing:

Distilled water….

The liquid test bottomed out at 6.0 (as low as it goes). The Ph controller/probe settled at 5.9, so the soil is acidic. In order to help balance that out I picked up some Lime from lowes:

I put 2 cups at 1lb 11.5 oz each on the bottom of the hole:

I covered that with about 1” of soil to help prevent direct contact with the copper tubing:

 

[Naja002]

Finally, after deciding what size tubing, how much, etc….I had decided before even starting this hole that I would run 8 parallel loops = 8x 25’, so I had to measure out and cut the 4 coils in half:

So, now it was time to call it a day.

So, now on day 3 things were getting rough—really rough. Pumping caffeine and popping max strength Tylenols (heavy duty drugs here, folks

) it was tough to get started. But I knew that I needed to get this done and the hole filled back in. I had today and tomorrow to reach that point.

I picked up a container at wally-world to house the indoor/outdoor thermometer. Cut a small gap for the probe wire:

And stuffed this up in a spot that’s +12” above the ground. As a note: the ½” pvc also rises +12” above the ground.


Now, I realized before starting the hole that I would have to build the manifolds on site (in the ground). It would have been really nice to do it sitting in a chair, but that wasn’t going to be possible. The problem was the heat from sweating the copper and the copper’s connection to the pvc. So, prior to digging I sweated together the initial copper to 3/8” connections….all 16 of them:

So, I then sweated those onto each end of the 8 25’ coils, then sat down and applied Teflon tape to all, a 1 ¼” tee to 14 and an elbow to 2. Determined my pipe length for in between the fittings and cut 16 pieces of pipe. Glued those into one end of each and things were setup to be completed in the ground.

So, after setting everything up….it was time to get back into the hole and start putting things together. Ultimately, it ended up looking like this:

Now, for those who are going to say” Oh, man, you shoulda done this instead of that or that instead of this”….as I said at the beginning of this thread: If I had it to do over there are somethings that I might do differently. Don’t get me wrong…I am open to comments and criticism…there’s no problem there. But it is done the way that it’s done and initial results are very promising!

The hole actually has a grade. The sloping pvc is the input (warm water), the elbow straight up is the output (cool water). The grade slopes downward from the input ~29” to the output ~32-33”. The purpose for this is simply that heat rises. It’s a simple effort to help keep the warmer ground from affecting the cool water result. Also the tubing coils rise 3-4” above the ground…they are not in direct contact with each other as it may appear in the picture. When I filled the hole back in…I made an effort to make sure that I did not compress the coils onto each other. In other words, there is earth separating the coil loops. Yes, there is some contact, but not near as much as may be assumed from the pictures.

 

[Naja002]

So, after getting everything laid out, and with the repeated insistence of my Buddy Hicks ( Thanx, Bro!) it was time to leak test. I wasn’t going to, but I did. Running out of time I decided to hook up a garden hose, fill the system and then use it for the leak test:

Well, I had a leak. One of the 1 ¼” x ½” tees had cracked at the neck. I had this issue before on my seasonal manifold and fixed it like this (pix borrowed from that project):

Put gobs of pvc glue on/into the crack and squeeze it together with a clamp:

Again, those pix are from that project. Same issue, same fix….just more glue used on this problem.

So, after letting it sit for about an hour I decided to try the leak test again. The leak was down to a drip, no other leaks noticed and I knew it would be a few hours before the one leak completely sealed itself. So, still running out of time…I decided to move forward. I put a couple of inches of ground on top of the coils/pvc and then added another +3lbs of lime. Then kept filling in the hole. All-n-all I used probably 15-20lbs of lime adding some every 6” or so. I ran water the entire time to help the clay settle in and fill the gaps. I soaked everything down very well. I managed to get the hole ~50-60% filled and it was once again time to call it a day.

Day 4 was fairly simple. Finish filling in the hole, replacing the sod, spreading seed and just getting everything situated to where it didn’t look like I dug a hole (as much as possible anyway) and get everything shut down for a few days.

The Hole:

The Pile:

As day 4 went on I continued to try to fill/leak test everything. From what I could tell I still had a leak. Seemed to be about 111 oz/hour…just under a gallon an hour with only gravitational pressure. Hmmmm, not good, but 24 gals a day was something that I could live with and build a work around if I had to….but I’d rather not have a leak. Hmmmmm…..

So, everything was shut down in preparation for a few days at work. I managed to sort out what was needed in order to now complete the run inside to finish connecting everything into a working system. Stopped by lowes and got what I needed.

 

[Naja002]

After spending my days at work I was now once again free to do as I please, so Day 5 consisted of fixing the leak and just generally checking things out. Now I know that Black pepper can be used to temporarily fix a small leak in a radiator. And I also know that digging the frickin’ hole back out is not an option for me.

So, black pepper it is!!! I set everything up in order run the system with an extra pump that I have. Amazingly that pump is only rated for 6’ of head, but did the job very well. I added the black pepper to the system and got everything going. The black pepper did the trick!!! Yes!!

So, I let the system run overnight. After checking things out on Day 6 it was now time to get it together and get this baby hooked up and operational….so that is what I did.

I plan on plumbing another rad into the living room, but I want to be able to isolate the system to just this room, if necessary. So, that should explain the tees and ball valves:

I added unions on each run of pipe for drainage and future disassembly:

Each pipe runs at an upward angle to the elbows where is goes through the floor:

While I was out there taking pix I checked the temp probe:

I had remembered that I had forgotten to reset the min/max after I got done filling up the hole, so I went a head and did that for future reference. If it matters, they read min: 52.?, max: 55.0…..they had not changed since filling up the hole, so that’s good.

Not as neat as the car rad setup but this is the geo-loop coming through the floor:

I may redo it some time in the future, but it will do for now at least. The gate valve on the left is the output from the hole (coming out of the hole and into the pump, rad, etc. Cool water). The elbowed one on the right is the input into the hole (warm water). I need to fix the upper connection to the seasonal manifold…I have an elbow. I may use that or fix it some other way. Right now it doesn’t seem to be having any negative effect on flow—with or w/o the geo-loop going.

A pic of the geo-loop setup with the Eheim 1262 pump:

Just a pic of the indoor rad setup:

 

[Naja002]

Here are 3 x 3min videos showing the Geo-loop effect. Keep in mind that this is after 3 minutes the temps drop another degree or 2 after a while longer. More importantly, this is with the OCed crunchers pumping heat into the system. Without the crunchers it will drop another +7F.

Video of the air temp coming out of the rad fan. The clay covered probe is inserted in the fan in the background of the video. Audio has NO value:




Video 2 is the water temp entering the water block on the vid card of cruncher #1…it is pre-warmed by an OCed NB. The water temp is the blue LCD:




The last vid is basically a video screenshot showing the cpu temp change of Cruncher #1 which is a Q9450 @ 3.584. Please keep in mind that the cpu water is being pre-warmed by an OCed NB and an OCed Gpu:



Also Please keep in mind with the water and cpu temps--this is not the only rig on this system....

 

[Naja002]

Ok, so my basic plan at the moment is this: I will continue to experiment and play with this setup over the winter. However, without a heat pump it really defeats my purpose of using my PCs to heat this place. When the warmer months arrive I will be able to experiment further. I plan on using this Geothermal setup for A/C type cooling….not PC cooling. I plan to add another car rad which will be located in the living room. I also plan on adding another car rad outside. The Dual outside rad system will be the PC cooling system, and the Geothermal loop will be the home cooling system. If it is insuffucuent to cool this place then I will have to determine whether to let it supplement the a/c or confine it to this room (this is where the PCs are). Either way, this will save me money this coming summer. The Eheim 1262 is an 80w pump. The push/pull car rad fans are running @ 5v and pulling 43w. 1 9” fan on high=45w which will more then adequately mix the air in this room. So, basically I am looking at 168w=$12.18/month. If it is sufficient to cool this whole place then 43w more will be added and 2 9” fans that are already running (on low)….for a total of 133w +168w= 301w= $21.81/month for cooling…A/C!! Either way, this system will pay for itself in no time.

Cost:

Copper tubing--$231
Pvc, Copper fittings, etc--$296.15

Now I have probably $30-$50 worth of returnable stuff. What I am going to return and why….I don’t know at the moment. Also, the $296 figure includes a set of pipe cutters, Teflon tape, etc….Its an all inclusive figure, so $250 is probably more accurate. However, it doesn’t include a few thing that I already had like some of the 1” barbs, blue tubing, etc. So, we’ll go with the $296 figure….if it’s a bit high that’s ok.

So, my guesstimate is $527.15 out-of-pocket.

Also, that doesn’t include the car rad $68, pump $109, fittings for that stuff, etc. I already had that stuff in use….but there’s the general value.

I hope Ya’ll enjoyed the show…still need to wrap the pipe with insulation and then cover with foam, so there will be more to come. Anybody looking to save money and go green—look into Geothermal…just google it. Nothing new. Been around for many years….pays for itself in 3-5yrs on average with a professional install. Check it out….


Oh, did I tell ya the part about how much I hate anything that involves a shovel…..

 

[Naja002]

Updates:

Managed to get the elbow installed in order to eliminate the creased tubing:

Turned out to be a real pita. Installing the elbow was no big deal, but I only have 1 set of drains atm, so trying to bleed the air back out of everything (both car rads, geo-loop, PCs, etc) was about a 3 hr job.



I've been trying to sort out how to set things up for this summer. Looking at buying a couple more car rads, trying to figure out placement, plumbing, etc.....just tossing ideas around in my head. Well, after playing around with this Geo-loop I know that it can handle the PCs with no problem, so I've decided to install another one....No, not the same as this last one. Basically I'm back where I started in May: Crawlspace......?

I stopped by Lowes last night and picked up 200' of the yellow pipe wrap insulation. I'm going to cover that with the black foam after I get the pipe wrapped. But while I was there....I picked up 90' of 1/2" straight type M copper pipe.

My plan is this: The air in the crawlspace stays about +/-60F during the warmer months. After I get the heat exchanger built I am going to drag it under there and dig the ground out ~2". Then install the exchanger and cover it up.

The current loop is ~19.6sq. ft. This next loop should be just over half of that at ~11.75 sq. ft. The PCs raise the air temp coming out of the car rad 7-9F, so I wouldn't expect it to increase more then double (14-18F) on a half-sized exchanger. The spot is shaded 24/7 and the air is also. There's some air-exchange of course, but it's not massive or anything. I think a shallow geo-loop will work out well. My plan is to use this next loop strictly for the PC cooling. It will eliminate the outside car rad and allow me to use that in the living room-- a nice, big triple-core rad.

$175 is what the stuff cost. By the time I purchased a car rad for the living room, another for the outside setup.....I would have spent over $175 for the rads, fittings, etc. So, I'm hoping/thinking/expecting this to not only be less expensive.....but I honestly think this will work better overall also.

After pondering the design for a while, I think that I am going to hit lowes and return a few pieces and get some others. I'll need to think on this for a bit. The cost is going to go up some, so I'll let ya'll know what the deal is after I get it sorted out completely....

Either way, another...PC dedicated loop...is on its way....

-------------------------------------------------------------------------------------

Ok, made it back to lowes, so it's time to up-date some $$$-figures....

$527.15 is what I had figured for the initial loop. Now it's time to add the isulation wrap and foam:

Wrap: $16.34
Foam: $12.78

New total: $556.27



PC dedicated Loop:

After returning some goods and getting others the new total is: $175.45

I am not expecting to insulate this piping. I may need to, but if so....I really don't think it would be much. Mainly toward the outside where it comes up through the floor to the manifold in here.

So, expected total for both loops is currently: $731.72

That of course is not including pumps, rads, etc......

Adding in those bigger items--prices are shipped:

Outside triple-core rad--$130
Inside double-core rad--$68
Eheim 1262--$109
Mag 9.5--$79

New estimate: $1,117.72

I will be putting together the new loop today or tomorrow....not sure when installation will take place though, but I am eager to get it setup....

Also, I had purchased 90' of 1/2" copper pipe. I was going to do an 8x 10' loop and use the extra 10' for the manifold. That's been changed. The manifold is going to be pvc and I decided to return the extra 10' instead of incorporating it. So, the loop is still going to be 80' of copper, but that works out to 10.5 sq. ft. of copper surface area.

 

[Naja002]

Update:

Started on the 2nd Loop for the PCs. Its built, just needs to be leak tested and installed. I expect to leak test it today, but installation won't be until at least next week. Ran into 1 problem during the build. Here's the problem and the straight-forward fix:

After starting to put one end and side of the mainfold together, I realized that the outside runs are not going to match up:

So, I used a straight edge to mark it as best I could:

Took off the elbow:

Cut the pipe:

Removed the fitting:

Sanded the pipe and inside the fitting (--just to try to smooth it back out a bit):

Resoldered them:

And put the elbows back on:

The manifold is completely built, but here are pix from before the actual completion--End 1:

In the pic above you can also see another split collar and pvc glue/worm clamp fix. Those fittings really suck. I would guess that ~10% of them split at the collar. I wanted to get a couple of extras, because I knew this would happen again. But they only had the exact number (12) that I needed.....

So, we'll see if it leaks during the leak test. Its had ~36hrs to set.....

End 2:

I'll try to get some pix of the entire manifold when I drag it out to leak test it. I've got several things that I need to do today, so hopefully more pix will be forthcoming....

 

[Naja002]

Update:

Gee, let's see....

Leak tested the 2nd manifold for 2 hrs at household psi, not sure what it is, but it's definitely 25psi or above. Average household psi is ~40psi or above. Either way, it's a whole lot more then the pumps we use will produce!

Loop assembled:

Filling it up:

I ended up having to actually glue this end-cap on and then cut it off later. Within about 3 secs of turning the water on--it would pop right off!

I jacked it up on some blocks to help check for leaks and let it sit for a couple of hrs:

Backyard phase and Bar-b-que anyone?

That's about what it looks like...

Started wrapping the piping in the crawlspace:

I wanted the overkill, so....:

I managed to get it all wrapped with the filberglass and plastic. I got it covered with foam, except for about 8-9'--ran out of foam. But I picked up some more, so just need to finish the job. It became a real time consuming pita. You can see in the pix below how the pipes cross over each other:

Lota fun trying to wrap those suckers....

But they are 100% wrapped. Now I just need to finish the foam and move on to installing loop 2.

While I was out there I checked the temp probe. There was a temp variance. Not sure what the cause of it is, but I have a few ideas:

Temp reading at the time:

Min:

Max:

1) the temp of the ground maybe varying....duh! that's the easy one. However, 2) without the pipes being insulated and the daily high's in the 30's and 40's, Low's in the 20's and teen's---it may have come from the ~50F water in the ground rising as the colder water in the crawlspace pipes sank. I'll know a bit more on that after I finish the insulation job. And, finally, 3) maybe the exposed unit was affected by the low temps. Even though its indoor/outdoor---the unit itself is ment to be kept indoors. It's a $9 unit, so maybe it's exposure affected the outdoor probe reading, so to help offset that:

I insulated it before putting it back.

I reset the min/max, so I'll check things again when I go to finish the foam insulation.

But the fun part is a head: Laying on my back or side while trying to dig out the "trenches" to install loop 2........Good thing I'm off for the next 2 weeks.....

 

[Naja002]

Here's an update:

I've gotten things setup and functional as of this morning. Right now, I'm just checking things out. It's supposed to be about 70F and sunny, so today will be a good little initial test. Tomorrow 77F and sunny. Yes, sunny matters.

I did things a bit differently then planned. The original plan was to use loop 1 (the buried one) to cool this room only....then possibly the other areas if it worked well enough. And use loop 2 to just cool the PCs. I went ahead and incorporated loop 2 into the entire setup, so now the 2 loops will be cooling this room and the PCs. I am going to leave the winter rad on the floor where it is. I am going to take the triple-core rad that I used in the out-building and bring it into this room also. It's going up on top of the PC hutch near the ceiling (cold air falls, hot air rises--max exchange).

The temps in the crawlspace last summer when it was sunny and in the 90'sF stayed around 58-61F. I did not "bury" loop 2....I just put it in the crawlspace on top of the ground. I also added a Mag Drive 9.5 pump in the crawlspace and will be swapping out the Eheim 1262 for another mag 9.5. So, here's the loop:

PCs to loop2 to loop1 to triple core rad to floor rad to PCs.

I don't know quite what to expect, but, yes, I am expecting/hoping to cool the PCs and this one room (~10x10) without the use of AC. We'll see how it turns out.

I am going to have to turn the geo part off tonight (in the 40's), but I expect to take some screenshots/pix tomorrow. Today is checkin' it out day and will give me an idea of what to expect tomorrow. Additionally, the triple-core rad is not in here or setup yet, so I am currently only using the "winter" rad.

 

[Naja002]

Ok, first screenshot:

On all of these Tjmax is set at 100, not 95.

This PC is cooled Cpu and NB only. It uses onboard video, so there is no separate Gpu.


Q6600 @ 3.502--this MB is messed up, but I have managed to get this stable OC out of it. This is my everyday rig that does email, surfing and all that stuff plus Boinc and Dimes 24/7. Minimum temps are "idle" temps.

Ok, rig 2. This one is a Q6600 @ 3.591 with 2x 8800GSs. This is a cruncher that runs WCG, GpuGrid and Dimes 24/7/365. Again, minimum temps are at "idle".

Ok, this rig is a Q9450 @ 3.584 and has 1x 8800GS. It's a cruncher that runs WCG, GpuGrid and Dimes 24/7/365. Not quite sure why the temps are so high, but I can understand why I may have at least 1 flow issue and what needs to be corrected. It will just take a bit of time and effort.

This is still 8-10C cooler then just the winter rad producing heat.

My 4th rig is another Q9450 @ ~3.58 with 1x (air-cooled) 8800GT. The Gpu is most likely going to stay on air. The Cpu and NB are WCed, but they are on their own system. I may incorporate them into this mega-loop....we'll see...

 

[Naja002]

Below are just some temp pix:


Temps of the actual water temp inside the rad. This is after an hour or so....it was an after-thought:

Mixed up set of temps here:

81.5-84.9F are the ambient temps about 10-15 minutes after opening the geo-loop for today. The 81.5F was moved up and squeezed in there for the pix....just a few inches, but it's not effect by the cardboard backing that the other thermometer is mounted against. Which temp is right? Not sure really. The 84.9F was originally 85.8F when I first opened the geo-loop.

The 87.6F is the temp under the PC shelf, which is where the HDs and PSUs are. There's a couple of pix back in Post 18 back behind the cardboard and the 4 80mm fans.

The 59.9F is the temp of the air coming out of the rad.

The suns been shining well all day....After 4 hrs and a 5F increase outside...current temps are:

Ambient: 85-86.2F
Air out of rad: 68.6

Not great, but there's only the indoor rad and it's on the floor. Also, the flow issue I mentioned is that the water is going to both the PCs and the geo-loop at the same time--like running parallel loops, not a single long loop. In probably a month or so I'll be able to change it to a single long loop. Right now I need to be able to shut off the geo-loop for the cold nights ahead. I may correct it before then.

I'm not overjoyed with the result thus far, but there are a couple of improvements to be made here.....so, we'll see what happens! :up:

 

[Naja002]

As you can see--I don't have mad paint skills,

but here's a current diagram that shows the parallel loop setup:

The gate valves are used to open and close the geoloop. Turning it into 1 big loop is not a problem. Being able to go back and forth seasonally and conveniently will take some thought...

 

[Naja002]

You may recall the temp probe that I buried in the hole....here are some pix after a long cold, cold winter:

Seems like the max inside the little tub was:

The min inside the little tub wasn't so bad considering that there were many, many days in the teens and single digits this winter:

In the Hole:

Current:

Max--yesterday was the first day over 70F in months--most days haven't even reached the 60's this winter:

Min--This one kind of surprised me at first. But it was 5F here about a month ago. I had a pipe crack (not in the geo-loop) a month or so ago and it was wrapped! Plus warmer water rises and colder water falls, so over the cold months--the above ground water in the geo-loop pipes may have cooled while the "warmer" water in the hole may have risen....not sure, but here is the low for this cold-azz winter:

This winter has definitely been a Great test to see if I'm below the frostline! Well, I guess I am!

It doesn't get this cold this long very often around here, so life should be generally good.

 

[Naja002]

Update:

Well, I managed to damage-beyond-repair the 3 row rad that I used last yr--left it with water in it during the freezing winter.

I know better, but what's done is done. Repairing the rad was not a possibility--too many pinholes. "Rebuilding" is was not a possibility either--new core was $300. I found another one ($98.38 shipped) and it should be here tomorrow. I also picked one up locally that is a 2 row version ($60 out the door) of the same rad. Not sure what I'm going to do--seems that nothing is final around here until it's final.

I am seriously thinking of doing a double rad in the outbuilding and putting the PCs on that loop. Or I may do the single 3 row again, and use the 2 row in here for cooling. Just not sure yet--benefits either way.

I definitely have a small leak. If I shut off the geo-loop--it will lose several gallons of water overnight/24hrs, but if I keep it running....everything seems fine. Go figure.

If it was just a matter of filling it back up--it wouldn't be a big deal. But it is a bit of a time consuming gurgling process. So, I'll be happier when I can just run it 24/7 in a month or so.....

If for some reason it becomes too much of a pita....I'll break out the sprinkler valve and controller, incorporate it so that refilling won't be such a headache.

I have an incoming water source in this room that I installed for some previous projects (aquariums).

Not there yet---shouldn't be too much longer though before I can run this thing 24/7 and get some straight answers.....

 

[Naja002]

Ok, let's try to do some type of update on this:

The car rad is still on the floor. I haven't moved it up anywhere, but I've been occupied with plenty of other stuff. I've got everything I need to move it--I just have to move a monster out of that spot. The rest will be elementary!

As is the air coming out of the rad is generally 10F or more cooler then ambient....so it is having a cooling effect. Free- A/C...uh, no, not quite. I just checked the in ground probe and the readings are:

Low: 47.3F
Current: 68.5F
High: 70.0F

Not so good all-n-all. My guess is that I got below the frost line, but still needed to go a foot or 2 deeper minimum.

I've been running the 2 copper loops--the Geo and the 1 placed in the crawlspace. I just disconnected the crawlspace loop and have started running on the geo only, so that will help determine whether or not the crawlspace loop is behaving like a heat sink....I think that it may very well be...

I do have a leak, but with the auto-refill of my Bong setup and the seasonal manifold--filling the geo-loop is a 60 second no-brainer. Luv it!

Once I get it moved up to it's spot...it should offer a bit better effect. I may also check out the fans on 12v.....noisy, but I may be able to live with it. This room sounds like an airport already....

 

[Ogre67]

Just gonna say pretty awesome... While definitely a bit more on the hardcore side I like it.

 

[soulesschild]

My hat off to you sir. You are truly the definition of [H]

 

[undertheradar]

Nice. I am also part of the 'massive radiator' club...lol.

http://www.hardforum.com/showthread.php?t=1399592

 

[Naja002]

Just gonna say pretty awesome... While definitely a bit more on the hardcore side I like it.

Thanx!

My hat off to you sir. You are truly the definition of [H]

Thanx!

Nice. I am also part of the 'massive radiator' club...lol.

http://www.hardforum.com/showthread.php?t=1399592

Very nice! I've considered that finned type tubing in the past, but always wanted to avoid mixing metals. Now that I've gone to a chiller...I use anti-freeze because of the aluminum on the evap. But I'll be fixing that in a couple months!

 

[haadij404]

amazing stuff here! my hat off to you sir.

this was a great read. thanks for posting.

 

[ThisIsTheNewSound]

Hahaha awesome. Truly a [H]ard setup! Thanks for posting your log!

 

[Shingoshi]

But for anyone else reading and considering doing something like this, I would suggest a different approach.

1.) Dig the hole as properly described here.
2.) Use black plastic sheeting to cover the bottom of the hole.
3.) Pour a thin 4" deep layer of concrete in the bottom of the hole.
4.) Let concrete set.
5.) Instead of using copper, use polypropylene tubing instead.
6.) Using 1000' of tubing, crisscross the concrete floor of the hole (For the cost of copper, you can afford more plastic tubing).
7.) Pour another thin 1" deep layer of concrete over your tubing (You don't want to compress the plastic tubing).
8.) Let concrete set.
9.) Pour another now 4" deeper layer of concrete over the first layer.
10.) Continue with the original instructions as posted here.

At step #2 you should mix black latex paint into the concrete. Black surfaces have unique thermal characteristics. Not only does black absorb heat, it also radiates heat better than any other surface treatment. That's why the condensers of refrigerators are black. Black is simply more efficient. So mix your concrete using black latex paint. You might also want to ask a paint store to sell you the black pigment that comes in the tube.

There's a possibility that mixing cat litter in with the concrete will help with the absorption of water during the summer months. During the winter, the ground may be frozen enough to resist water saturating the concrete blanket. But during the summer, the water soaked into and resting on the concrete blanket will greatly assist in the transfer of heat into the blanket. If you really want to ensure water reaches the blanket, you could install a drain tube down to the blanket and distribute that water over it.

I believe the benefit of doing this will even more dramatically increase the efficiency of the geothermal cooling system.

Shingoshi

 

[colinstu]

Holy shit... first [h] thing I've seen for awhile. thumbs up!

 

[DarkCow]

Shingoshi

But how well will polypropylene transfer heat from the water to outside surface?

Also doesn't concrete just heat up? I just don't see how the concrete can be cooler than the moist dirt around it.

 

[Shingoshi]

Shingoshi

But how well will polypropylene transfer heat from the water to outside surface?

Also doesn't concrete just heat up? I just don't see how the concrete can be cooler than the moist dirt around it.

1.) How ever low the heat transfer properties of polypropylene are, it will be offset by the cost of the copper you won't be using.
2.) You will never have to worry about the ph of the soil for the installation of polypropylene. Polypropylene tubing won't ever corrode.
3.) Polypropylene tubing being flexible won't split if the fluid ever managed to freeze.
4.) Soil is less dense than concrete, and therefore won't absorb heat as well as concrete. Soil will in fact act as an insulator. Just think, which would you rather sleep on in the middle of winter, A) Concrete or B) Soil. That should answer your question.

The choice is yours...

Shingoshi

 

[Naja002]

1.) How ever low the heat transfer properties of polypropylene are, it will be offset by the cost of the copper you won't be using.

For standard polypropylene you would need 2,000x the amount of pipe surface area--No, that's not a typo--and you definitely will not be saving any money.

For HD polypropylene you would only need 800x as much-- and you won't be saving any money.

2.) You will never have to worry about the ph of the soil for the installation of polypropylene. Polypropylene tubing won't ever corrode.

Not completely true and there are ways to setup copper against corrosion that will make it all just as durable.

3.) Polypropylene tubing being flexible won't split if the fluid ever managed to freeze.

Done properly--there will be no "freeze" so it's a moot point.

4.) Soil is less dense than concrete, and therefore won't absorb heat as well as concrete. Soil will in fact act as an insulator. Just think, which would you rather sleep on in the middle of winter, A) Concrete or B) Soil. That should answer your question.

Concrete has about the same TC as average "ground". There's a number of variables that determines the ground's TC....it averages about 1.5 though. So, for the purposes that you state adding concrete would truly be just a waste of time, energy and money. With the proper additives adding concrete can be beneficial, but not for TC.....



This is what happens when you think you can "guess" the answers....



The choice is yours...

Shingoshi

Inside quote^^^^^

 

[undertheradar]

He is onto something though. I didnt think of it until I read that last post. Several people in my other hobby use geothermal loops to cool their reef aquariums, and rather than set them up as some sort of radiator system, they instead bury large poly containers (200g or more) in the ground. Both the earth and concrete are about equal in insulation (it really depends on how moist the ground is), but they do insulate well enough that it would seem such vast array of copper tubing would be best left 'above ground' to act as a convective radiator system. Simply put, the ground is a better insulator than the water, so if you take away the tubing material, and just consider the transfer of heat to the surrounding environment, having the water transfer to other water does make more sense. Most of the in-ground 'sinks' I have seen are instead just large volume-heat-storage devices that just have a mass that is large enough to last the whole day when active, and then cool all night when off... large poly tanks, sometimes multiple ones in series sometimes as large as 800 gallons of water buried in the back yard. At that, depending where you live, you could have some tanks buried, and some left above ground. It does seem that doing that would yeild better results than a radiator buried underground.

How do you plan to circumvent the freezing anyways? Are you keeping all the tubing below the frost line and then punching holes through your basement walls? Or are you putting enough anti-freeze into the loop that it just wont freeze?

Here, these are the tanks they bury in the back yard...
http://www.usplastic.com/catalog/product.asp?catalog_name=USPlastic&category_name=35&product_id=3874

If you go to reefcentral.com, and use the search for 'geothermal' there are multiple threads that should pop up where people bury these things. And make no mistake... these things are set up to handle a good kw/h in cooling, so Im sure they would handle a computer.

 

[Naja002]

He is onto something though. I didnt think of it until I read that last post. Several people in my other hobby use geothermal loops to cool their reef aquariums, and rather than set them up as some sort of radiator system, they instead bury large poly containers (200g or more) in the ground. Both the earth and concrete are about equal in insulation (it really depends on how moist the ground is), but they do insulate well enough that it would seem such vast array of copper tubing would be best left 'above ground' to act as a convective radiator system. Simply put, the ground is a better insulator than the water, so if you take away the tubing material, and just consider the transfer of heat to the surrounding environment, having the water transfer to other water does make more sense. Most of the in-ground 'sinks' I have seen are instead just large volume-heat-storage devices that just have a mass that is large enough to last the whole day when active, and then cool all night when off... large poly tanks, sometimes multiple ones in series sometimes as large as 800 gallons of water buried in the back yard. At that, depending where you live, you could have some tanks buried, and some left above ground. It does seem that doing that would yeild better results than a radiator buried underground.

You're mixing apples and oranges. As I'm sure you realize--all of this is application specific. Some people turn their PCs off, some leave them on, some OC the shat out of them and crunch 24/7/365<--that's me. I push 900-1Kw 24/7<--that's what's actually being cooled by the water.

A 200 gal MDPE tank buried is not going to handle the load. Reef tanks don't absorb all of the electricty (heat) that they use. Plus as you said--the geo-tanks are allowed to cool at night. And just as importantly--so is the ground. On top of that most reefers are only trying to cool down a few degrees F....some only a degree or 2.

I understand your point, but once you switch applications--everything changes.

In my case, my limitation is the ground. That's the weak link.

How do you plan to circumvent the freezing anyways? Are you keeping all the tubing below the frost line and then punching holes through your basement walls? Or are you putting enough anti-freeze into the loop that it just wont freeze?

Below the frost line.

Here, these are the tanks they bury in the back yard...
http://www.usplastic.com/catalog/product.asp?catalog_name=USPlastic&category_name=35&product_id=3874

MDPE @ ~0.3 TC. Compared to the water and ground it's an insulator....meaning the weakest link in the TC chain. 1,336x less TC than copper.

If you go to reefcentral.com, and use the search for 'geothermal' there are multiple threads that should pop up where people bury these things. And make no mistake... these things are set up to handle a good kw/h in cooling, so Im sure they would handle a computer.

I may in the future. If you want to point me directly to a link or 2...I'll check it out. In reference to the application again....sure-->A computer. But not everybody only runs 1 PC. I run 4 24/7/365.

And make no mistake... these things are set up to handle a good kw/h in cooling,

That's not really accurate. A Kw/h....how many hours of the day? What about when the lights are off, etc. I run my stuff 24/7...that makes all the difference--same load....24/7.

 

[undertheradar]

If you are not trying to get a 'mass thermal storage' device, but rather want 24/7 cooling, then I wonder why you would want to put a radiator underground in the first place. Putting it underground means mass heat storage, but if you intend on using it constantly, then its not. I hate to say it, but aren't you blending two conflicting approaches to cooling then?

The thermal conductivity of your ground is going to reach a much higher equilibrium than if the whole radiator were simply left 'above ground'. You may be attributing the assumed cooler ambient temps of a geothermal setup, but if this is for constant use, then the ground will heat up... no longer being a sink, but a giant insulator for the loop. Your air temps might be higher, but air will also transfer the heat faster... even if its 110 outside, it will still shed heat from the computer, and do so at a constant that will be higher than anything underground. You are correct that the ground is your weakest link... so why use it? You could just use a radiator like I did, or maybe a couple... one outdoor, one indoor, so you can mix the two for the ideal cooling on any particular day. LIke I mentioned before, 5-6 50' coils of 3/8" copper coiled waterline will easily cool about 2000 kw/h indoors.

I just want to clarify... you actually plan on punching holes in the basement below the frostline to run the loop into the house?

As for your conparison of the poly tanks to the ground and copper... merely relying on the TC of the materials w/o respect to other values does not tell the whole story. The MDPE/poly material is not a higher insulator than the ground because the thickness of the container is very thin, while heat will have to transfer across many feet of ground to exit the system. Treating the copper pipe with some poly based coating to avoid corrosion would render the copper pointless otherwise. The large volume of the water also comes into play, but for 24/7 cooling... you would need even more volume, and then yes, attention to surface area comes into play more than just volume. But then you are defeating yourself by placing the system in the ground. Even if your ambient air temps are higher than in the ground, the ground will block conduction where as the air will easily convect.

You are right about the different application of the reef tanks vs. this... but by your own admission then, it makes me then question why you are going geothermal. Geothermal is best when used as a volume heat sink and not for constant use. Constant use undermines the whole idea.

You are right about my kw/h figures as well. They lack information on the delta between the processor and ground, as well as what the actual temps would be. I can get specific if you like... I am no longer working in the thermo field, but still remember how to do all the calculations. Just let me know if you want me to calculate anything. I think its safe to say though that I am talking about keeping a processor at about 100-110F if the ambient temp of air is 70-80F, maybe 20 degrees higher if you are overclocking. Those reef tanks are kept at 79-82F, and you are correct that they do not rely only on the geothermal for their cooling (then again, neigher will your computer if using external cooling). With 1200 watts of light over say... a 200g, I would bet only about 100-250 watts of that actually gets put into the water... the rest is convected into the room usually. But with a reef, there is usually a good 2 watts per gallon of energy going into the water due to the massive pumps required. The tanks themselves act as a sink as well as they heat up through their daily photocycle... but their delta must remain much smaller... smaller than anything we would use for silicon. Going more than 2 degrees in either direction on a daily basis will kill everything inside, and although there is some convection and evaporative cooling going on, the tank itself is being insulated by glass on 5 sides, and sometimes on top as well if not actively cooled. On top of that, the main display is kept in rooms that are also insulators (maybe not to the high intensity lights, but to the water) since the room temps are usually similar to what is in the tank, or 10-15 degrees cooler in winter. A processor handles a much higher delta to ambient temp. Most of my BTU and wattage calculations are assuming air temps of 65-75 degrees, and a liquid temp of 180F. Sure, these things might change, but overall, we are operating in a range that is rather forgiving... we arent talking about huge wattages or deltas that are more than 100F. We can start to get picky when the scope of the project is to eek out just that extra degree lower in the processor, but considering the magnitude and scale of the cooling solution compared to the desired application, debating deltas is a mute point. Debating the material you intend to use to conduct the heat from the water to the ground is also a mute point considering the relative mass of the ground in the system. You could have used PEX and most likely gotten similar results... unless you have water flowing through the ground...lol.

 

[Shingoshi]

I've been thoroughly BITCH SLAPPED!!

Shingoshi

 

[Shingoshi]

I thought about this, but then decided not or forgot to include this in my posts.

Remember the part where I said

2.) Use black plastic sheeting to cover the bottom of the hole.

What I was thinking was to effectively create an underground pool. Kind of like a cesspool, but for cooling the computer instead. If you have all of your plumbing installed in that artificial subterranean pool, the surrounding water will do everything we need it to. The pool would be constantly replenished by water seeping into the ground. I think all of us would be happy with that result. At least I think so.

Shingoshi

 

[Naja002]

If you are not trying to get a 'mass thermal storage' device, but rather want 24/7 cooling, then I wonder why you would want to put a radiator underground in the first place.

For the same reason people do for general heating and cooling. Finish your research.

Putting it underground means mass heat storage, but if you intend on using it constantly, then its not. I hate to say it, but aren't you blending two conflicting approaches to cooling then?

Mass heat storage is a possibility--if things are set up improperly. So, finish your research so that you will understand what when and why.

The thermal conductivity of your ground is going to reach a much higher equilibrium than if the whole radiator were simply left 'above ground'. You may be attributing the assumed cooler ambient temps of a geothermal setup, but if this is for constant use, then the ground will heat up... no longer being a sink, but a giant insulator for the loop.

Very true--again--if set up improperly. Geothermal loops are used all over the world for heating and cooling--you should know this, but apparently you don't....so, finish you research. Then you'll understand what reality is.

Your air temps might be higher, but air will also transfer the heat faster... even if its 110 outside, it will still shed heat from the computer, and do so at a constant that will be higher than anything underground.

Whatever. You are just"guessing" and thinking you know the answers.

You are correct that the ground is your weakest link... so why use it?

Right. So, WHY is that? Explain it. I know why, so let's hear it. There's only 1 reason--what is it?

You could just use a radiator like I did, or maybe a couple... one outdoor, one indoor, so you can mix the two for the ideal cooling on any particular day. LIke I mentioned before, 5-6 50' coils of 3/8" copper coiled waterline will easily cool about 2000 kw/h indoors.

Ummmm, you really have no idea about my projects, do you?


2000Kw/h, huh? 2,000,000 watts? Really? You are a fool. Do you even know what a Kw/h is?


I'm not even going to waste time responding to the rest of your post. You are babbling and just trying to be "Right" when you really need to finish doing you research. Go finish learning about geothermal. Actually learn about it....and not think that you can use your "common sense" to just "guess" the answers--because that is what you are doing.

 

[Naja002]

I thought about this, but then decided not or forgot to include this in my posts.

Remember the part where I said
What I was thinking was to effectively create an underground pool. Kind of like a cesspool, but for cooling the computer instead. If you have all of your plumbing installed in that artificial subterranean pool, the surrounding water will do everything we need it to. The pool would be constantly replenished by water seeping into the ground. I think all of us would be happy with that result. At least I think so.

Shingoshi

The plastic will degraded and become useless before long. My original intent though was to use a pond liner in the exact manner that you describe. However, after weighing the possible negatives and the added cost--I decided against it. This was never intended to be a really long term project. More of an experiment.So, a pond liner should have lasted for a reasonable length of time--in my case. But I couldn't really pin-down the reality of any negatives....and the cost is quite a bit considering.

 

[Shingoshi]

Hey Naj, would it be any better to first lay the liner and then spray that stuff they use to coat pick-up truck beds (Gorilla?)? I don't know. I'm just asking if it would be a long term solution. Then, you could do all the stuff that you've done here, and have your underwater pond as well.

Shingoshi

Edit: Does this kind of look familiar?
http://www.zurnpex.com/BuildersContractorsHome/tabid/222/Default.aspx

 

[Naja002]

You would have to spray it on both sides and then I doubt it would help. The ground is a pretty treacherous place....

 

[Shingoshi]

Did you understand that I only meant for creating a basin to catch underground water in? Why couldn't you apply it in layers to reinforce one layer with another progressively, until you had the strength and rigidity required? How large of a basin would be needed for that?

Hell, come to think of it, you could use a kiddy pool or two, to create the basic structure, and then coat that. Then you could fill the pool with sand. That might be better than soil.

Shingoshi