Underground dry "Bunker" using sustainable materials - Triston Line

Hey Everyone!

I'm new to the permaculture movement, I work in IT, and I love the natural environments of the world. Kind of ironic really, my industry uses far too much packaging. If you have any better suggestions for the category, please let me know.


Anyway, I'm ambitious, and plan on building an underground datacentre (a "room" full of sensitive and powerful computers). I'm not sure how yet, but I have a pretty good idea of the concept and that generally involves making an underground structure, that process is well documented online from start to finish (the most amusing example could be Colin Furze's:   https://www.youtube.com/watch?v=ZQy89tZ-mRU)  

What bothers me about the general methodologies employed by conventional underground structures, is that most of them utilize highly refined/processed, and ultimately unsustainable materials. I understand that in recent years, we've begun to learn that concrete actually carbonizes and thereby reabsorbs a lot of (or most of) the CO2 that it emits in the setting process, but it's also quite energy (and water) intensive and I'm wondering if there is a better material that I can make or purchase.

The other question that I have, is whether or not I really need concrete or an "alternative" with similar properties. Maybe the real answer is layers of multiple materials, rammed earth, or a new concrete and graphene mixture that would reduce the required concrete and remove my need for steel (Example of newly produced graphene with concrete being used in construction: https://www.youtube.com/watch?v=n82_NsFjj_ Maybe the simplest idea of simply submerging shipping containers is the best solution, not because it creates demand for unsustainable materials, but because it uses an oversupply of shipping containers in North America (https://www.youtube.com/watch?v=b1JlYZQG3lI) Worth noting that I am somewhat opposed to the shipping container idea due to rust/metals leaching into the ground, but I do like the idea of using surplusses of material or objects that have no better use case (such as shipping containers right now).

Thoughts on this idea would be welcomed, I am so new to permaculture/sustainable building of underground structures that anything will be something to me, even if it's an off-topic penny of wisdom.
Cheers, and thank you for your time.


Triston

I see a couple of alternatives here:

1) You move your data center to a climate and geologic area where you do not need a structure at all. If you sited it in a desert with very little rainfall and stable rock you could simply mine out a cavern and use it just like that. Check on youtube about a place called Sandland. Some sandstone is soft enough to dig with hand tools.

2) If you cant move to better ground conditions, think about just offsetting the impact of your little endeavor. You are going to be filling this room with energy hungry electronics anyway, so getting to net-zero is going to be as good as it gets. If you want it to be sustainable, then you will need an on-going carbon-offet program to cover the operational emissions. Just amortize the construction emissions and add it into the overall offsetting goal. Natural materials will be hard-pressed to keep a space dry if the ground naturally contains moisture, so steel or concrete is likely going to be required.

I am not sure why your "data center" will need to be underground, but if you want to cut down on the use of unsustainable materials, building it above grade is going to be a lot less energy intensive.

Also, burying anything that is not designed for burial (i.e. shipping containers) is not a good idea.

Welcome to Permies, Tristan!

Regardless of your final approach, that fact that you've thought about the impact of your project environmentally is awesome.

I suggest you start with this article by Paul Wheaton: https://richsoil.com/wofati.jsp
Particularly note the issue of keeping an "underground" building dry and comfy. Sometimes the best approach is actually to build a pond and use the pond dirt to bury your building so that water flows away from your building on all sides - or some version of that!

You may find more ideas in our Wofati forum?  https://permies.com/f/75/wofati-earth-berm

Many of the "natural" building systems aren't recommended for underground use (or reuse in the case of the shipping container - I've read *many* sources that suggest burying them is a bad idea). For example, "earth bag" building is often suggested for round basements or footings, but adding concrete is considered essential to keep moisture from causing problems. This is still an improvement over all concrete and metal, and could be much better over the long term than a conventional stick building that is inefficient  for the entire length of its relatively short lifetime.

If you give us some idea of the size of the project and your ecosystem, you may get better informed replies. In my geological location, earthquakes and lots of rain are two key concerns, so that will bias my suggestions. Electronic equipment doesn't  like being drowned!  

Triston Line wrote:
What bothers me about the general methodologies employed by conventional underground structures, is that most of them utilize highly refined/processed, and ultimately unsustainable materials. Triston

Triston, welcome to the forums!

Here are some threads with some concepts that might interest you:

https://permies.com/t/164291/earthbag-cellar-double-survival-shelter

https://permies.com/t/165961/Root-Cellar-Fire-Shelter

https://permies.com/t/162602/Storm-shelter

https://permies.com/t/61403/Rainy-PNW-Underground-Root-Cellar

https://permies.com/t/32701/Earth-bag-root-cellar

Hi Triston,
Welcome to Permies and thanks for the link to Colin Furze's construction, certainly an unusual thing to have in one's garden shed! I', wondering whether it was strictly necessary to cover his steel structure with concrete? I suppose he must have done the maths for the loading required, maybe he's expecting to drive on the roof? It certainly puts a new meaning to 'man-cave'! I can't really add anything to the posts above, but please keep us informed on your project.

Hi, Triston! I really can't speak to the rest, but I'll second (3rd?) the advice against submerging the shipping containers, for your project - or much of anything else, for that matter. Their construction is designed to keep stuff in, not out, thus, they're not built to withstand the external pressures inherent in being buried, and have a wicked tendency to collapse - and, the deeper they're buried, the faster they collapse.

Hi everyone!

Wow so grateful for your replies! So soon too!

Hey Carl, I probably won't be able to house it in a desert, I live on Vancouver Island, the best I could do would be Lilooet (pretty much a desert town) and I'm not sure the commute would be worth it, but they have a lot of rail, so old metal and shipping containers could be abundant. Fear not regarding net zero, I will make it grid tie, but the above land will likely house a photovoltaic generating field (yes, a minimally efficient photovoltaic field because of the rainy location). Business objective will also be to sustainably manage the land and have renewable energy production elsewhere to off-set my carbon emissions. There will not be a diesel generator unlike most datacentres for example, and I would like to enable natural ecosystem growth.  

I'm hoping to reduce footprint by taking advantage of all three dimensions and I'm hoping to make it highly earthquake resistant. Part of earthquake proofing is about knowing what's under the earth, so I will also be avoiding liquifaction zones and nearby fault lines. Any other advice here would be welcomed of course.

"Also, burying anything that is not designed for burial (i.e. shipping containers) is not a good idea." Fair enough


Hey Jay, thank you very much for the welcome, and thank you for agreeing that environmental risk analysis and permaculture thinking for this project is a worthwhile idea, I'm so glad that there is an available community to help out

Re: Pond dirt - Ok so bury something not quite underground (make it a hill kind of)? Not a bad idea as it keeps me above recent rainwater, but my region has excellent drainage so I'm not sure how much to be worried.
Re: Earth bag - Sounds like earthbags really hate UV and are best used underground, checks out. How does this compare to rammed earth for moisture and strength? Are you suggesting that I pour concrete over earth bags in a form? Another trend that I observed on my brief browse of the web, was rounding of earthbag basements for their strength, which you mentioned briefly. The shape matters little to me, but my machines will be running in a squared formation, if that makes sense. https://i.pinimg.com/474x/ac/f7/25/acf725a5e94ca44cef96acb195c6ca41.jpg for reference. The arrangement matters little to me other than where cold air is brought to the machines and where warm air is exhausted. Normally for minimal ducting and moisture I would consider a heat pump solution, but perhaps there is a better method here?
Re: Project and Ecosystem - Let's pick a volume of one 20ft shipping container and the ability to add a new shipping container's worth of space every ~5 years. I live in Victoria, BC, right now, but I'm certainly not going to bury anything here. My best estimate is up island near Campbell River / Courtney, but I am still exploring other options (which may be influenced by my future tiny home location). The whole island gets a lot of rain, frequent earthquakes, and is built of stone (a lot of limestone however, which is pretty brittle).


Hi Anne, thank you very much for those links! I had read one or two but missed the storm shelter, and last two cellar articles! I'll read them right away.



Hi Nancy! Colin is a lunatic, but I absolutely love his creations and projects because he delivers impeccable results from highly creative ideas. His usage of concrete was two fold: 1. Moisture. 2. Weight (possibly permitting too from his UK municipality). Happy to keep updates on project thoughts and ideas.


Hi Carla! Ok duly note, I'll really dodge the shipping container idea. It was in my mind for a long time simply due to convenience and having witnessed "mobile datacentres" (shipping containers full of expensive computer equipment). The other draw was my fascination with Tiny Houses, and not that I want a shipping container for a house, but many have done it rather successfully. You know, I hadn't thought too much about the common sense of a shipping container's purpose being quite the opposite to my goal, maybe I had ignored it because of how well they stack on cargo ships. However, you are quite right, stacking corners is very different from a heavy load on top and all sides with uneven forces on the walls/panels. There was a thought to pour concrete below and on the sides, but even that would have required reinforcing (come to think of it) not to mention the amount of concrete required.
Thanks Carla, I'll consider something else.


Thank you all for the first round of replies, really set me straight on the shipping container idea, I may come back to it if I don't find satisfactory alternatives. I've marked my replies in different colours for accessibility purposes, if you'd like me to change any of the colours, or to try something else, please feel free to let me know.


This project's characteristics will change a lot depending on what property I am able to find, that part will have to be a waiting game, but you may be able to influence my purchasing criteria!
What would you look for in a property with the following region and requirements:

Situated Vancouver Island Region between Port Alice and Nanaimo

Houses moisture-sensitive electronics

Produces significant heat (5-15kW) which may require exhausting (how would you utilize this heat? The cooling solution will likely be a heatpump, is there a better way to do this?)

Is in a geologically young and seismically active region.

Will live underground.

Why do I want it underground? 1. Allows surface to be used for other purposes (renewable energy production). 2. More efficient space usage. 3. Natural cooling. 4. Protection from certain elements (heavy rains, sea salt).
Why does it have to be Vancouver Island? Well, it doesn't but I live here an am only going to put it someone else if I can drive to it within a few hours. Keep in mind that it will need to be physically serviced from time to time.


Thanks again everyone, happy to see some interest in this idea and already some great advice, have a wonderful day and if you have a moment, please feel free to leave your thoughts!


Triston

Why do I want it underground? 1. Allows surface to be used for other purposes (renewable energy production). 2. More efficient space usage. 3. Natural cooling. 4. Protection from certain elements (heavy rains, sea salt).
Why does it have to be Vancouver Island? Well, it doesn't but I live here an am only going to put it someone else if I can drive to it within a few hours. Keep in mind that it will need to be physically serviced from time to time.  

Haha, if you really want an underground bunker because it is cool, I am certainly not one to say its not a viable solution. However, here is some food for thought on your above points:

1) Instead of putting the servers underground and the solar panels at ground level... Put the servers above ground and put the solar panels on the roof.

2) Instead of digging down, you can build UP. Footprint is the same, just keep adding floors and moving the solar panels up too. If you dont want to shade any land, put it on your northern property line.

3) Natural soil temperature underground will likely be closer to an ideal range for the servers, but soil is likely not going to be an infinite sink with so much heat being produced. You will probably eventually saturate the soil with heat, and will still need some way of pumping that heat out. Also, outside temps will be warmer underground in the winter when you have less solar to offset the cooling loads. On the surface, higher summer temperatures will coincide with more solar power to run the AC, and lower winter temperatures will offset some of your cooling. Recapturing the waste heat is a noble thought, but I suspect that it will prove to be too expensive to be practical. Maybe you could dehydrate fruit with the exhaust from the heat pump :)?

4) A roof and marine paint on the walls should be sufficient.

It seems like if shipping containers are being used already to house servers, then you could just copy that. It will likely save you a lot of headaches with permitting, as you might be able to call it a temporary structure. If it is bolted to the foundation piers, I would think it would be about as seismically secure as you could hope for. Also, whoever is bankrolling all of this might be happier with a tried and true solution over something more ... fringe. That said, I am pretty sure there are engineering firms out there that specialize in environmentally sound practices.

This is the closest example I could find of my recommendation.

Some have used shipping containers

The main thought I had was that the underground cooling tubes dry the incoming air as the moisture condenses on the tube walls and drains through the perforations int the ground.  Your air circulation would be driven by the heat from the processors instead of solar gain.
What ever the construction material the important  thing is the water proof membrane that separates the earth around the structure from the surface earth that can become saturated by rain.
A steep gravel hillside is usually considered a poor building site but could be ideal for your purposes and less expensive. If south facing ideal for mounting solar without the horizontal footprint to avoid shading.
Filling the corrugations of a shipping container with aircrete is a possibility.  Aircrete is made with cement and air bubbles instead of sand and gravel as in concrete.

Two more replies, yippee! But before that, I have been presented with a new idea, one that I've spent some time thinking about and researching.

British Columbia is home to over 3000 mines, ~600 of which are on Vancouver Island and of those, ~300 are underground. I've retrieved an outdated spreadsheet from the provincial government reporting on all of this, but most importantly it also lists Mine names, ore types, and their locations!
So here's an idea presented to me by a friend, what if I refurbished the entrance section of an abandoned mine? Well here are some of my thoughts:

They're quite wet, even with "designed drainage". Not to mention, that drainage is often towards the entrance for horizontal mines as they dig up usually...

It would be a LOT of work, but there is a very very small chance that it wouldn't be as much work as digging a new hole!

Almost guaranteed to be far from anywhere my fiance would like to live, so it would be off of my property...

Mines that are "abandoned" and have a low sale price are likely to be very hazardous.

There is potential for multiple businesses... E.g. mushroom farming in a branch or simply further into the mine.

The environmentalist in me likes the idea of using a company (even my own) as a means for some rememdiation of a polluted area. If there is a suitable mine available, I would definitely factor in tailing remediation via systems such as an artificial biological blackwater filtration system and the removal of old and dangerous equipment (think: old unstable dynamite and rusting minecarts).

Unlikely to be within city limits... No building permits required!

Carl:
1. Extra structures cause shading, require me to mount the roof in order to service the panels, and require racking for the structure (more difficult than DIY racking on the ground).
2. Ok northern property line or up hill, I fear the fiance may dislike these additional buildings even with a good coat of paint.
3. Oh now you're digging into the good stuff. Importantly, where I live, winter is much cloudier than summer, when there is basically blue sky for three months straight (except the forest fire smoke caused by the three months of what is essentially a summer drought every year). So yes, my solar production capacity will be reduced, but I don't see the ground becoming "warmer" during winter, our winters are mild in any case. Your point about ground temperature saturation is valid, but that doesn't mean the saturation point will be the ambient temperature of the room. The ground is essentially a massive heat sink, and Vancouver Island in particular has some of the heaviest rain, and the best drainage. Maybe the right thing to do then, is put the servers in a hillslope so that water is draining over and past them underground, and to add gravel around them to promote that drainage. Alternatively, as Hans is suggesting further below, perhaps simply plumb empty pipes out of the hot-aisle (server exhaust facing) walls at ceiling height, plumb it a few meters away from the underground structure, and then loop it to the cool-aisle (inlet facing) walls at floor height. No fans required. So long as the air is able to be passively cooled in the pipe, air will flow. (Otherwise it will stagnate or worse it will create a positive feedback loop.
4. For an underground structure? Also isn't marine paint somewhat toxic?
5. Ah yes, but those shipping container datacenters are on truckbeds, being hauled around. No permit required for them... My only addition to your note about seismic structuring would actually be to increase motion/bending, instead of increasing rigidity as that can cause the machines inside to topple instead or worse, if one of the piles decides to break, the entire structure could tear. Excellent resource for understanding earthquake proofing quickly: https://www.bigrentz.com/blog/earthquake-proof-buildings and on the topic of using flexible materials for earthquake proofing: https://www.constructionexec.com/article/engineering-better-wood-structures-for-earthquakes-past-experiences-and-future-trends

Hans:
1. Thank you for those videos Hans!
2. For moisture inside of the datacentre, I would likely rely on either a rackmount AC or a dehumidifier that operates only when necessary to reduce humidity to below 70%. Getting rid of the collected water would be another issue, I don't know what is the best solution for emergency draining, a drain into gravel? That would require some careful planning to avoid underground water saturation from flooding the datacentre.
In addition to heat output of the machines, each machine has a set of static pressure fans, which if positioned correctly, could push air through pipes that leave the container from the "hot-aisle" and return at the foot of the "cool-aisle".
3. Do you have recommendations for such a membrane?
4. I'm not opposed to it either being burried in mostly gravel on a slope, or burried on a slope with a retaining wall and gravel on the retaining wall side of the structure. I like the retaining wall idea because then I can utilize even more flat surfaces on an otherwise steep slope. I'm not certain that this will have a positive effect on waterflow on the property however, because water that runs quickly is often better at eroding than water that is forced to move slowly through soils and vegetation. On the other hand, most of Vancouver Island is again mostly stone and drains exceptionally well, given how much rain we receive.
5. Although Aircrete appears to be an interesting material, my brief search online lead me to understand that it is not very water resistant or environmentally friendly other than utilizing less concrete per volume because of the air bubbles. What are your thoughts?


Thanks Carl & Hans for your replies!


Triston

It just occurred to me that solar panels on an unoccupied property might be a magnet for people who decide that their need is greater than your need, so I suggest you will need to consider how to manage security?

Real life here: The Hertzberg Institute of Astrophysics (renamed since) decided to put up new signage at their entrance at the base of Little Saanich Mountain 12-15 years ago. They decided to power the sign with solar panels on a large pole structure (tall trees and a big hill blocking sun down low.) It disappeared over one weekend relatively soon after installation. I'm not aware of the panels ever being found or anyone being charged.

Panels are cheaper now relatively and more common, but I'd still think they might attract unwanted attention.

I was hoping the engineer in the first video would be local enough to be available to you.  
Thinking about your proposal while traveling past new housing tracts being developed. Just a few thousand feet away was a triangle shaped hillside lot left isolated by freeway construction.   So as you look for housing look for such parcels. The local tax authority will have a searchable data base to locate the owner which is often a developer that would like to get them off their books.
Corrugated drain tile pipe is inexpensive and flexible so that with planning can be installed during the excavation and covering phases. As you observed they can be open ended drawing cool air in at the floor and exhausting hot air from the top or a closed loop using the earth as a heat sink.  If there is no hydraulic pressure bringing moisture into the tubes the net effect is dehumidifying because humidity is condensed in the cooling and drains from the slits in the corrugations. The machines do not need oxygen and the entrance can supply fresh air when servicing so a closed  loop would be acceptable and potentially safer from intrusion by biological contamination.