Preparing for Nuclear war – issues of inside shelters

In the 15minute shelter I assumed that the shelter would be outside. The reason for that, is it might just be possible to make a shelter in 15minutes outside. That is because the material (earth) is readily available outside.

But what about a shelter inside? In this article I explore the potential for building a good shelter with 1m thick walls. A wall 1ft (30cm) reduces radiation by 10fold, so a 1ft wall provides a lot of protection, but a good shelter using earth has at least 1m walls.

But, there are big problems building inside:

First, there’s the psychological issue. Nuclear war is so detached from our normal reality, that it is very difficult to fit in the concepts of living in a nuclear war, with the day-to-day of our normal lives. Few of us have spare rooms which can be turned over to a shelter, our houses are the biggest we can afford, and we cannot afford to turn over a room to something. Yet that is what we’d have to do to create a shelter inside. So, my guess is that most people just won’t prepare. They need the space and they will tell themselves “yes that is where it will be”, but they won’t be able to bring themselves around to doing anything about it.

Second, is the simple fact that building a shelter in a house is EXTREMELY dangerous, both before and during an attack. The reason it is dangerous before an attack, is that you either dig down into the earth under the house, potentially causing it to collapse (both before and in an attack), or you pile many tonnes of soil onto flooring never intended for such extreme loads.

Thirdly, you imagine that your home will be just as snug and cosy in a nuclear attack, but when the power goes out and perhaps the windows blown out by the blast, it won’t be much better than sitting in the garden. But, at least you are mentally prepared for being outside in the garden.

So, let’s look at the potential for an indoor shelter:

In a cellar

A cellar, surrounded by earth, is about as good as anyone is going to get for radiation protection in a nuclear war without getting something purpose made. But they are not immune from danger if the building in which they are sited collapses or goes on fire, both of which could happen in buildings around a nuclear explosion,

But they are good for radiation, I hesitate to say “it doesn’t need anything adding”, because you can still improve the radiation protection in a cellar, by e.g. adding an L wall in a corner. They don’t have a lot of protection from back scatter from the sky. So a roof would improve it. Generally a cellar is so much better than anything anyone without a cellar has in their house.

But there are issues similar to a general concrete floor.

Concrete floor

A room with a concrete floor, like garage, with thin or wooden walls, offers almost no protection by itself, but the concrete floor may be able to take the many tonnes of earth that are needed to make a shelter. I say “may”, because there is a risk of the added load causing the ground beneath the concrete to compress, and that could cause cracking to the floor and/or depress the whole floor causing cracking around the edge. Either could lead to water ingress and a hefty bill for the home owner. Also, unless the soil you use is bone dry desert material, you are bringing in a lot of moisture, and that also could lead to problems if you are “just trying it”. Which is why I have to advise against it … because I don’t want someone trying it and then blaming me for expensive bills to their house.

These risks are also why, as we approached a nuclear war, why people thinking of using a room with a concrete floor, would probably delay.

However, in an actual emergency where you don’t care about water ingress and “moisture” damage, it isn’t a bad place. If you were to do it, then you need to fill sacks with soil and pile them up. My guess for an “ideal” shelter with 1m thick walls for two  is that you want about 20 cubic yards, or 20 tonnes of earth for the shelter (see why there might be an issue!). Filling sacks and carting them inside – I’m guessing it will take 1 minute to fill the bag, 1 minute getting into your room and 1 minute to stack and get back. That’s 3 minutes for 50kg. That’s 60minutes per tonne, or 20 hours for the shelter. You also need 400 bags … and I’ve not worked out the details of how to add a door.

Why so long to build? When I said a shelter can be produced in 15minutes outside? The reason is that the 15minute shelter starts with only 1foot of soil removed, and that the earth being removed is moved only 1m to the side. And, the size of the inside of  a 1m x 2m shelter (2m²) is much smaller than the outside (10m²)

Wood/suspended floor – DON’T

There is no way a normal house and most other raised floorings can support the enormous weight of a good nuclear shelter (1m thick walls). It may be possible that your floors could support a much smaller shelter, perhaps using 1foot (30cm) walls of earth, however the bearing weight of suspended floors varies a lot. Some modern ones are not supposed to support a full sized bookshelf. Some older ones are so rotten, they wouldn’t support one. Before considering an internal shelter, you should consult a structural engineer, who will likely say they wouldn’t advise it.

However, if you literally have no other choice. Then making the shelter so the long axis is perpendicular to the joists (parallel with the floorboards) with low walls … that would be what I would try in an emergency if I had to build our house where I know the structure.

If you have 9″ thick brick walls, or brick cavity walls, that gives approximately the same protection as a single 1foot wide sandbag wall.  So, if you enclose a corner of the room (without windows or doors) with an L wall of sandbags, your joists might possibly hold. It gives slightly better protection than a room totally surrounded by solid 9″ walls. However, you also need earth under the shelter as well – particularly if you are lying down.

Under floor trench shelter

If this goes wrong with this, your house falls down. So, never to be recommended … except if you have no other option in an emergency. No one  considering this option (who is sane) is going to start doing it before an emergency.

But theoretically (and that is all this is not a recommendation) how would someone who knows what they are doing approach it?

They first have to get under the floor. They then have to check all the services to ensure that they won’t be digging into any or knocking any whilst working. Then you have to check the depth and extent of your foundations and consult a structural engineer to work out the critical angle. This angle is the region of soil that supports the foundation. Remove it, and the house will crack (expensive) or fall (very expensive and very dangerous).  And, the critical angle will be much shallower if the ground is subject to an earthquake which do happen in nuclear attacks and I doubt most structural engineers will know how to assess that risk. And, I don’t know what it is.

Also, before going down into the shelter before the emergency … you have to completely turn off any gas, because if the nuclear attack cracks any pipes going under your house, you are toast. Likewise, if your house goes on fire, you can’t easily get out. So there are some SERIOUS drawbacks.

But the theory is that you work out the angle from the foundation down to the bottom of the trench and so long as that is less than the critical angle, you can dig the shelter.

Realistically, even in the best conditions, you won’t build a suitable shelter in a space under the floor of less than 3m by 4m  … and any room much larger will have additional supporting walls under the floor, meaning the  space under the floor doesn’t get bigger.

The general problem of building a shelter in a house – fire & falling houses.

The nuclear attack will cause a shock wave, that will cause houses near the blast to collapse. It will also cause tremendous heat, which will set alight houses quite distant from the blast. A shelter in a house is at risk if either of these happen. A shelter outside is much less likely to be affected.


There is no safe way to build a good shelter with 1m wide walls in a house. A walled shelter is  such an enormous weight that it becomes a serious engineering project and a shelter dug into the earth under a house could cause it to fall down.

It may however be possible to build a 1m thick shelter on a concrete floor, but it still could damage your house, and it will take a long time.

Shelters with 1ft (30cm) walls provide a lot of protection, particularly compared to wooden houses. Some house floors may be strong enough to support such structures, but I can’t recommend it, certainly not outside the context of an emergency.

Even if you can build a shelter, there are serious risks from collapsing buildings and fire. A cellar, with a strong floor above it, will give some protection against the collapse of upper stories, but you could be trapped or killed. Cellars are generally resistant to fire, but a wooden floor over them is not, and even if the floor above does not burn, killer smoke could get in.

The general conclusion is that houses are not great places for nuclear shelters. Houses severely limit what you can do, and they add extra risks of building collapse and fire and so if you want a good shelter, you need one outside.

This entry was posted in Ukraine. Bookmark the permalink.