In this segment, we mark and place the front columns and the curved I-beams that form the framework for the entry and green house sections. Most of the time-lapse footage was lost some how, but I did have some pics…
The video
Surplus Steel
I bought the columns from the surplus steel place in my area. The cost was low enough that I didn’t mind a few imperfections. No regrets and I will probably do it again. I did put tape over the holes to keep wasps from moving in.
Trouble with the Forks
When I bought the skid steer, the guy who sold it to me said he also had a beat up set of old forks that I could have for 200$. New forks cost 3 or 4 times as much, so I told him to send those with the skid steer even though I hadn’t actually seen them. At first, I just noticed that the back board was a bit damaged. After using them, I also noticed that the two forks were actually different thickness (miss-matched set) and had bent slightly differently and I was having trouble holding things level.
We didn’t worry about the back board, but my father and I fixed the “uneven” issue with some torches (and lots of patience) to heat up one of the forks so we could bend it to match the other.
But all that time, I was using the forks to lift heavy things, so I didn’t notice the 3rd issue… When you apply loads the other way (pushing down on the forks), the locking mechanism is supposed to hold them in place. However, the top ledge that holds the locking mechanism in place had been slightly stretched upward and increased the tolerance by maybe 1/4th of an inch, and that was enough for the mechanism to actually detach when the load was pushed the other way.
While setting the second I-beam, The beam got hooked on the bent back shield and wouldn’t let me lower the forks. Since this flipped the load direction, it also shifted the locking mechanism down 1/4th inch relative to the forks and they detached from that top edge.
With the load direction reversed, the forks detached from the skidsteer
Those Forks are a few hundred pounds of heavy steel, so rather than just let them fall off and possibly damage something on the way down, we strapped them to the quick attach mechanism on the skid steer so we could still lower them carefully.
The final fix was to weld 2 pieces of angle iron across the top of the quick attach mechanism to remove the gap so it won’t be unlocked by a reverse load.
Final view. There will be windows under most of those Ibeams and a Front door under the left most one. Earth covered in grass, etc. will be above.
At the moment, the construction schedule has 3 main paths.
1) On the east side, we are getting the steel arches rolled. We can start to erect those after the excavator comes in and sorts out the grading issue, but they are waiting on the frost laws to allow them to carry their heavy equipment on the roads.
2) On the west side, I am getting the Quonset hut ready as a work shop to build the wooden forms for the precast concrete ribs.
I probably have a months worth of work on each of these sides before the paths converge on the middle.
3) In the middle of the house, I need to get in these steel posts that will support the ring beam that will eventually support the precast concrete ribs that will eventually hold up the steel arches and shotcrete that will form the radial vaults.
This weeks video is about setting up the posts for the center section.
The Video
The Story
I’ll put the story in here to provide a little more detail and to make it text searchable for Google ;^).
Working backwards:
In this region of the earth sheltered house, the load from earth above (not shown) is directed, by the radial vaults (made of heavy shotcrete over steel arches), down onto the precast concrete ribs.
On the outside edge of the house, these radiating concrete ribs (which weigh almost 5000 lbs each) are sitting right on 4’x4′ concrete pads made of strong reinforced concrete 1 ft thick. However, at the middle of the house, the high end of these ribs are set into a central tower made of shotcrete (which will carry most of the final loads) over a steel skeleton (which will carry all the loads during construction before the shotcrete is in place). This image also shows the QuadDeck ICF floor forms. When concrete is poured over these, it will help lock the steel posts into position.
The load of those 10 heavy ribs runs thru the ring beam and down the structural steel columns into the shotcrete basement wall below.
The 3D CAD model shows a nice flat surface where these pipes attach to the top of the basement wall.
The reality is that the shotcrete guys did not do a great job of squaring off the edge of this wall, probably because it was 9 ft tall and hard to reach. I needed to fix that. I setup some cardboard forms to the level the wall should be at and backfilled with hydrolic mortar. It was thin enough that it was pretty much self leveling.
The next step was to prepare the steel bases. I bought a box of scaffold bases for $4 each, but the (nail) holes were too small to fit the anchors thru. I drilled them out using the lowest setting on my drill press and some lubrication.
Drilling steel requires low speed and lubrication. I lubricated with thread cutting oil. The smallest container I could find was much more than I would need and only cost ~4$, the drill bit was 7$, so I used the lubricant generously in order to preserve the life of the bit. I found applying it directly to the bit was the best way. If the bit started to smoke, I would stop the drill and add more oil. If the speed/pressure and lubrication are correct, long spirals of metal will come off. This indicates that the steel is being cut and not just wearing away the bit.
Then it was time to head back out and mount the bases to the wall. I was using sleeve anchors that require you to drill a hole first, and then drop the anchor in the hole. It is important that the sleeve be completely below the surface. When you tighten the nut, the sleeve is expanded and presses against the sides while the nut pulls what ever you are attaching down toward the sleeve… I was in a rush and didn’t put one of the first anchors in properly. There was probably dust in the hole. But I couldn’t pull it out again either. With the sleeve too high, the nut tightened against it before pressing the metal to the concrete, so it would never be tight. Eventually, it will be under another ft of concrete, so I was not too worried about it, but I put the rest in more carefully.
I was also careful to make sure each scaffold base was level so it would be easier to plumb the columns later.
My steel posts were ready a couple days earlier than promised (a first for me on this build), so I picked them up in my trusty trailer. They weighed about 100 lbs each and put the trailer near its official limit, but it felt like it pulled easily.
The following weekend, my parents visited with my sister.
The mission was to move the steel posts into position on the base plates, hold them steady and level and then tack them into place with my welder. The first one worked pretty well. But the welder jammed on the second one. I couldn’t untangle it, so we had to cut out all that welding wire and re-setup the machine. From then on, I always checked the wire spool every time I moved the welder to make sure that it was not tangled.
This third post was the one on the less stable base, so we added extra bracing to keep it plumb until the concrete is poured around it. We had a lot of experience pluming steel last summer, so everyone knew exactly what to do and it only took a minute or so.
My sister and I both enjoyed the welding, so we took turns welding vs holding… Some times on the same post if it was tricky to access a certain point. The most time consuming part of the process was simply adjusting the scaffolding, etc. so we could reach what ever we needed to reach.
It took us about an hour to get the first 5 posts in, but they were on the side of the circle that was easy to reach. We were expecting the back side, were we had to balance on the wall, to be more difficult, but it was pretty easy too.
After all the posts were in place, I was worried that someone may give these a good push and cause them to shift… Yes, they are welded to steel plates bolted into strong concrete with 3 inch steel anchors, but an 8ft long 100lb lever would probably be able to pry those out.
I wanted to weld the ring beam on right away, but each half was 320 lbs and I couldn’t think of a safe way to get them into position.
So we decided to weld some rebar between the posts as a temporary solution. This was nice and easy and I really liked the look of it, so it will be sad to have to cut most of it off again when we need to put the door bucks in.
I was not sure how far my tank of shielding gas (Ar and CO2 for the MIG welder) would go, so we only did tack welds. That turned out to be a good decision because it ran out just before we were done. Without the gas, the final few welds looked pretty ugly.
After that, we put in some work prepping the shop for building the forms, and I will talk about that another time.
The next step in this center section will be getting in the QuadDeck ICF (insulated concrete form) floor over the basement. I am having a painful time getting that scheduled.
The rest of the steel arches are still being rolled and I will soon start on the forms for the ribs.
After completing the footings, the next step was to erect the steel studs and metal lath that we would shoot the concrete (shotcrete) against. You can think of the steel studs as fancy integral formwork. I had had some trouble up front getting the steel ordered, particularly the steel studs. So I went with another company to at least get the track (including Flextrack) ordered. This let me put the track down first while I wait for the steel studs to arrive.
Basically, track is needed to hold steel studs in place. The studs hold the metal lath and the metal lath will catch the shotcrete and shape the walls.
I originally had this part of the job in my gantt chart as a 6 days’ worth of work starting the first week of May, so I started last week about 7 weeks behind. To make matters worse, I ended up working on it over several partial days spread across a week and I haven’t even started on the vertical steel studs. I would guess that by the time I am caught up on putting in the steel studs, I will be about 8 or 9 weeks behind. Winter is coming, but I will resist the pun of saying that I am working to get back on “track”.
There is definitely a learning curve, but I consider the basement as practice for the main level and I was definitely faster by the end (as you can see in video). Knowing my velocity (per ft of flex track or straight track) lets me better estimate the time (and resources) I will need to complete the main level.
Here is the time lapse video.
On to the story.
Like an ant moving a mountain, I am just doing things one bit at a time and trying not to be overwhelmed or forget anything. For this past week, the first step was preparing a simplified version of my drawings (a basic shop drawing) with just the door buck and track dimensions labeled.
Measure twice, cut once… But first make sure you know what the measurements should be. And simplifying the drawing to show just what you need does help… But make sure you also have an idea of the other bits that will interact with that or you could get yourself in trouble.
The next bit was putting together the bucks. I decided to do that at home where I have a chop saw and a garage to work in. I cut and labeled all the pieces and loaded them (un-assembled) onto my trailer. The bucks are made from 2×6 treated lumber. They are 6 inches taller than normal because I am fastening them to the footing and will come back and cut off the 6 inches at the bottom after the shotcrete walls are in place. Then I will pour a floor to make up that 6 inches.
Others might have built the walls on top of the floor, but my “wall first” approach uses the floor as a shear plane against lateral earth loads. “Walls first” also makes the shotcrete installation much easier because it covers up the joint between the wall and the floor (so they can be less picky about it) and because they don’t need to shovel the “rebound” out of the basement. Rebound is the “pea stone” and cement that bounces off the wall during the shotcrete process. It can be up to 10% of the volume of the walls, which for my basement would be nearly 4 yards. Lugging rebound out of the basement would have been hard work for the shotcrete crew and would have cost me a lot of money.
I added my generator to the trailer (needed to power my hammer drill) and headed to pick up my supply of steel. The look on the guys face when I pulled up told me he was pretty sure it was amateur hour, they told me that there was no way they could load my steel onto that trailer. I quickly told them that, “This steel order weighs 660 lbs. I already have 332 lbs of 2x6s, 55 lbs of 2x4s and the generator weighs 163 lbs. My trailer can handle 1100 lbs and my car can two 2000 lbs, so we are all good.” I had just guessed at the generator weight, but my rapid fire numbers somewhat startled the loading dock guys for a moment and convinced them that I was serious. While they were recovering, I quickly took the back and side off my trailer and unloaded the generator so they could set the pallet of steel down with the fork lift. One of the guys helped me load my generator back on top again. Still a bit concerned, they asked me how far I was going. All I could say was, “All the way.”
Sherri said I should probably never show anyone this picture (above), but I think it shows that “where there is a will, there is a way”, even if you can’t afford to look like a pro and you end up being chuckled at by a couple of loading dock workers.
My Brother-in-law, John R., came out to help me assemble the bucks. With everything pre-cut, it was pretty straight forward. We would have screwed them into place, but I forgot my hammer drill. It took more than half an hour to get three Tapcon screws in with my regular drill. That was a waste of time, so we focused on leveling the steel storage container and getting ready for more productive days ahead.
To level the 4800 lb steel container, we used a hydraulic mechanics jack to lift one side and then we stacked bits of waste concrete (that the trucks had dumped on the sand) like a dry stack foundation wall. It looked pretty cool; I should have taken a picture before we covered it with sand. The process went pretty smoothly, but it was hot & tiring work, so I really appreciated John’s help.
As a consolation prize, John went home with a really bad sun burn.
I came back on my own with the hammer drill and got to work on the steel track. The footing is full of ¾ inch stones. It is a lumpy surface to work on, and depending on the underlying stone, the drill could go thru like butter or struggle and fail to penetrate at all. I often found that moving a few inches over was easier than trying to push thru a hard spot. Sometimes I had to try 2 or 3 spots before I found a good one.
After drilling the pilot hole, I would switch to my socket drill to drive in the Tapcon screw. I quickly found that the torque setting on my drill was also helpful. If I set it above 15, the Tapcon screw heads would just snap off. Still, it was often a frustrating struggle to get the Tapcons to screw into the harder spots. I worked until I ran out of screws and my cordless drill batteries were dead.
You may think I should have planned better and brought the charger, and you would be right, but at the time, I was tired and glad to have an excuse to go home.
The next day was my oldest son’s 10th birthday. I ended up taking him out to the property to help out. Having learned my lesson, I took my dwalt battery charger with me so I could charge one battery with the generator while using the other one in my driver. On the way, we stopped and picked up some more Tapcons. I also bought a box of smaller Tapcons (3/8ths instead of ¼), to see how they compared.
My initial plan was to use the larger Tapcons on the ends of the studs and the smaller ones between, but that was hardly necessary. The smaller Tapcons were just so much easier to drill and screw than the large ones. I ended up using up the box and buying another rather than keep using the larger Tapcons. The smaller 3/16ths Tapcons also cost 35% less than the ¼ inch.
I started out using ¼ inch Tapcons which have 1160lbs pull out and 900 lbs of shear resistance. Clearly that was overkill. The 3/16ths Tapcons had 900lbs pull out and 720 lbs of shear resistance, which should be more than enough to keep the wall in place while the shotcrete is applied. For the door bucks, I used several of the large ¼ inch by 3-3/4 inch Tapcons.
For the flex track, I bought very simple track. It doesn’t have the metal straps or locks or other “structural features”. It is simply cut so that it will stretch on one side. The shape is held by screwing it to the concrete. The fancy “structural” flex track is ~$2.70/ft and the simple stuff I bought is only ~$1.19/ft.
I had marked the footings at the right radii so I could lay my track. As a novice, I pulled the track into position, the stretching was uneven. I later learned that I could get a much more precise (and rapid) curve by inserting a screw driver and twisting to widen the outside side of the track. More experience and I found that two twists in opposite directions resulted in a more level track.
The close up install video is here.
I had to go back for a third day to put in a last few hours and get the job done before my son’s birthday party this weekend.
Next step is a small job of putting together the window bucks while I wait for the vertical studs and scaffold tower to arrive. Both should arrive next Wednesday, along with some family and friends to help erect it all.
