It's a wall!

After the boys’ swimming this morning we all headed up to Tallarook to finish ramming the wall panel. First job was mixing up a new batch of soil, but since I had half the last batch from last night left over I started with that and mixed in more sand, soil and cement. I’ve decided that with the materials I have on site, a 50:50 mix of sand and soil works best - more sand and the mix is too dry and won’t hold together, but it’s important not to get it too wet because then it doesn’t ram very well - it stays a little spongey, rather than compressing rock hard. There’s part of one layer at around the 1.6m high mark which got a bit too wet during an insta-storm this afternoon, which might cause trouble with cracking later as it dries and shrinks. We’ll see.

One of the things I was worried about and wanted to verify with this test wall was the colouring of the final product given the source materials. My little test block which I rammed a few months ago was a little dull in colour, but that appears to have been the fault of the grey GP cement as the off-white cement I’m using in this wall doesn’t dull the colour at all, and the wall has taken on a darker, rich tone from the intense red-orange of the local soil. I’ve not yet really seen the naked wall in sunlight, but what I have seen looks pretty promising.

Where the form panels join there’s quite a bit of patching required. Because the first panels bowed out in the middle quite considerably (20mm each in total, impressive considering how strong they are), I was unable to bolt the two courses of forms together as I had intended. This caused a step in the rammed surface of the wall which crumbles when the panels are removed, so when I do the next walls I’ll use a bolt through the middle of the panels top and bottom, with a length of half inch diameter conduit permanently embedded in the wall to prevent tear-out when the bolts are removed. This will keep the panels straight along their length, at the cost of a little imperfection where the through-bolt hole is patched. A fair bargain.

By around 5pm I had the third course of forms rammed to the top, so it was time to leapfrog the lower panel up to the top. This proved more of a logistical problem than I anticipated - not least because the two panels each side are bolted together, and each holds the other in place so when the bolts are removed, both panels will want to fall to the ground. Minor problem.

I solved this by stacking leftover concrete blocks (these things come in very handy) underneath the form panels, so three of the 200mm high blocks perfectly matched the 600mm high panels (there is a method to the madness ;) ) and they rested on these until they were lifted into place. Thankfully I had help for this operation - the neighbour Craig lent a hand to lift, and they slotted into place above the third course easily (and more importantly, safely).

I had expected to have to lift soil into the last course using buckets, but luckily the bobcat bucket goes just high enough to clear the top of the forms. At full reach though I can’t tip the contents into the wall, but since it’s a 4-in-1 I’m able to open the bucket and (most of) its contents ends up in the forms. Win.

Ramming continued into the darkness (again) until lightning began flashing around the hills and I chickened out, preferring not to be standing in the clear on a 2m high target. The final height before the rain set in was about 2.1m, and that’s where this wall will remain I think. I had intended to ram it to 2.4m, but that was an arbitrary choice so I’ve arbitrarily changed my mind :)

So this test wall has indeed taught me a few lessons about the process.

• First and foremost, the forms worked pretty well - I don’t think the columns moved at all, and although the panels bowed out impressively that’s easily remedied with a minor compromise.


• Metering the sand and soil to get a consistent batch is important, and my metering box did the job well.


• The bobcat did a very good job of blending the ingredients; it’s not fast, but it’s faster than either I or a cement mixer would be and the result appears to be good.


• It’s important to get the moisture content right from the start, especially when using the Plasticure additive as it’s difficult (if not impossible) to wet the mix further once it activates. This is especially important when mixing the first batch of the day, since it dries out considerably with warm-ish weather.


• It’s very important to set up an efficient work space ahead of time. Having everything ready to go, including forms, scaffolding and platforms on both sides, tools, etc. is crucial since the clock is running once the batch is mixed.


• I really, really need a filter/dryer on King Kompressor. The amount of water it produces in the air is incredible, and the rammer’s exhaust ices up considerably which I don’t imagine is very good for the tool at all.


• Finally, this is a really tough job to attempt solo. Setting up, batch mixing, filling the forms and ramming is a lot for one person to cope with. I’m not sure I could possibly complete a single wall panel in one day, even with everything set up in advance. It remains to be seen how well layers rammed from different batches on successive days looks and works.

All in all, I’m very happy with the test. It’s going to be far from perfect, but I’ve learned a lot of lessons in the process so the walls which count should be much better. One thing’s certain though - ramming a wall with reo mesh down the middle is going to be a royal pain in the arse…