ASMET is not “ceramic” as mentioned on the forum; it’s reinforced concrete (so is Roofkrete) – but not as we know it. John knows about ‘ferroconcrete’ yachts from the 60s/70s, but we’re not allowed to use that ‘bad’ word – nearly all have rusted their reinforcement and crumbled, except for a few still-healthy survivors which used a special expensive kind of cement. ‘Ferroconcrete’ hulls were typically 20mm thick. The standard ASMET panel is 14mm, and Roofkrete 6mm – not bad for reinforced concrete!

For both, the reinforcement is 12 x 12mm square weldmesh, 0.5mm wire, usually dip- or electro-galvanised because that’s available, but ungalvanised is actually preferable when available (e.g. China). It’s multi-layered 4-deep in Roofkrete, 10-deep in ASMET. In 6mm and 14mm panel thickness respectively, the mesh fills the volume and grins through the surface here and there. And rusts – but that’s OK apparently because the alkaline cementitious matrix permanently prevents deeper rusting, beyond what’s visible. Various tools and routines are developed for neatly lapping, folding and edging the mesh layers.

I’ll talk about Roofkrete first, as it illustrates some of the technical features. Unlike moulded ASMET, Roofkrete is laid-up in situ. Mesh layers are lapped together and lightly stapled through a light polythene slip-layer to the support boarding and/or clipped together with wire pull-twists (any wire high-spots are taken off with a disc after all is set). Cementitious mix is trowelled on , sets in minutes, and that’s it! Entire factory roofs are done, jointless, 100m sq; it’s a walk-on balcony surface, you can drop barrels on it, it’s completely water tight, vapour tight, you name it! It can be laid in convenient sections without need for daywork joints – apparently today’s mix bonds fully to yesterday’s.

I’m thinking – green roofs? Edward Cullinan/Buro Happold used it on part of the Weald and Downland gridshell project, where it was accused of leaking but after costly investigation it was proved to be copious underside-condensation due to pinholes in the ‘total’ vapour barrier beneath. A factory valley-gutter 70m long x 1m up both slopes, laid jointless in situ, has performed faultlessly where several previous methods had failed.

A 40 wide x 300 long x 6 thick strip of Roofkrete feels like spring steel; resilient and springy. Such a strip can be bent, pushing its centre away 6mm, with little-finger pressure. Much more than that, it begins to crack up – but that’s amazing elasticity for reinforced concrete! So, with such tensile cohesion combined with elastic suppleness, one can see that huge jointless slabs of Roofkrete can remain intact, whatever the inevitable movements of its supporting substrate and whatever its own inevitable differential thermal stresses (one part in cool shade, the neighbouring part in baking sun). But to survive that, you’d think that carefully-provided freedom to expand and contract in all directions under extremes of temperature would be essential. Over 100m, I’d expect that max-to-min expansion to add up to inches, and if not allowed would create big compression forces in the membrane, leading to membrane buckling and/or tearing-apart of the supporting substrate. The nearest jointless equivalent, glassfibre roofing, certainly requires such provision. However, the mystery is that no such provision is made with Roofkrete and apparently no such resultant problems appear – apparently Roofkrete doesn’t display thermal expansion! Is that possible?

10 layers of 12 x 12mm square weldmesh, 0.5mm wire, gives 1 mile of wire per square metre of ASMET panel, weighing 2.5kg, the equivalent of 1m2 of 0.3mm thick steel sheet - less than half the thickness of a typical car body panel. The volume of its dense concrete is 0.014m3, weighing 30kg. Anyone know how ASMET compares with the usual measures of cement and steel content for conventional reinforced concrete; the quantity of each per m2 of best-practice precast concrete panel?

First of all ASMET is certainly intriguing, but unsurprisingly it does raise a number of design issues. Whilst I’m trying to avoid making a judgment call at this stage I have listed my questions to date.

Fixings, Penetrations And Load Bearing Characteristics
The experience noted at the Weald and Downland project suggests that to avoid condensation, and the on set of mold within insulation material, the “wall plane” of the ASMET must be used to form the internal lining of the envelope (i.e. warm side of insulation). This negates its weatherproofing benefit, furthermore the consequence of locating the “wall plane” in this location raises concerns regarding the potential for penetrating (airtight) membrane. These concerns include:
a) Services penetrations through the ASMET envelope, opportunities for the ingress of water. (Not a new problem but how does ASMET cope with addressing these details? I’m thinking differential movement, longevity of airtightness, damage to structural integrity etc.)
b) Distribution of services within ASMET envelope, presumably to minimize penetrations though the air/vapour tight membrane this leads to the need for a services zone on the inside face again, whilst not a new concept how does ASMET cope with receiving SW battens/spacer bars (Can battens be adequately secured? Will they work loose over time? Can the battens be removed easily when the building is demolished or if they have to be glued on have we created an unrecyclable mess?)
c) Can ASMET support shelves, kitchen cupboards etc. Again concerns regarding fixings abound. What details will be required to provide adequate support? (Wouldn’t want to be tearing holes in the airtight membrane/structure)
d) If the service zone of item b is used some form of dry lining would be required so as to create a decorative finish and allow people to hang paintings, shelves and cupboards. What fixing details would be required, anything out of the norm?

External finishes:
Given the “T” formation of the monocoque how do fix your finishes?
Given that the lining of the envelope is on the warm side traditional weather protection to the insulation is required (soggy insulation does not perform well and will break down more readily). Does this mean creating an outer layer of ASMET (this results in box sections of ASMET) so as to encase the insulation? To form these box sections ASMET would ideally be extruded. From Tom’s description this is not currently the means of fabrication.
If we have created a box section i.e. outer layer of ASMET again how do fix your finishes?

Cold Bridging
Claims have been made about the system being suited to creating housing to the PassivHaus standards. Given that cold bridges can account for 25% of heat loss in a super insulated building how does the system avoid cold bridging?( if memory serves PassivHaus has a Psi value requirement of 0.01) What thermal break details are required? What impact do these details have on other aspects of the construction? How readily can these be achieved?

Jeff,
I agree that these issues would not appear to be insurmountable, however, before the ASMET system reaches the production line these matters will need to be considered, tested and proven to workable. From the information that I have seen to date no real conclusions can be drawn on these matters, hence me choosing to reserve making a judgement on the technology.

No secret formula just ferrocement

John Manniex, inventor of ASMET, has emailed me:

>Just a few corrections to your figures.

1. ASMET is constructed with 8 layers of 0.9 mm (19 gauge wire) 12 x 12mm giving a thickness of 14mm for webs and skin.

2. 7.2kg wire per sqm = 1,344 lin metres + 27.54kg of RoofKrete compound, total weight per sqm is about 34.74kg

3. Web (beam) spacing is 400mm

The beam testing at Portsmouth University relates to beams of 10mm thick and 6 layer of 0.9mm mesh.

We have upped the thickness to substantially over-engineer the product and then we could have a fall back position for reducing prices in the future. I do not think we will have a problem with price though because the sqm price of the present 14mm wall thickness including webs is working out about £120 and this should reduce further with quantity scaling.<

7.2kg/m2 of wire (not 2.5kg/m2 as I said) is the equivalent of 0.85mm thick steel sheet (not 0.3mm as I said) - so about the same as a typical car panel.

Posted By: Jeff Norton (NZ)I am visualiasing chicken wire and self leveling compond

No offence Jeff but hippies have been building chicken wire domes since the dawn of time. These may suffice in wimpy Arizona Whole Earth Catalog land but in a 'real' climate like UK and I'd guess NZ, durability seems to require a lot more than amateur know-how! John Manniex has clearly gone into this deeply with ASMET.
The use of weldmesh and the way it's layered has an extremely strong structural effect that might or might not be obtainable with chicken wire.
I kept asking about the cement/aggregate aspect and he consistently pretended he hadn't heard, so I know that's where The Secret lies! I've been thinking not so much self-levelling compound which is probably full of dodgy additives emphasising workability over structural strength; but more about cementitious tanking treatments. Vandex has unique crystal-growing additives that explain its supreme performance; but the numerous others rely on carefully-graded aggregates (which I can understand) and probably special kinds of Portland cement (which are beyond me). The strongest concrete results from selecting and proportioning aggregate granules of various sizes and shapes, so that once wet-vibrated they pack together to almost entirely fill the volume with pieces of near-pure crystal, of enormous individual strength. The relatively weak cement slurry is then left to do what it does best - as a glue of minimal thickness, not as a gap-filler.
In a rough attempt to create this gap-filled mass, concrete used to be specified as e.g. 1 part cement : 2 parts sand : 4 parts aggregate but the result was as highly variable as the particle mix within different sands and aggregates; so now it's a performance specification e.g C24, of strength, workability and other parameters. Now concrete batching plants make the best of the aggregates and cements that are economically available, in any mix that produces the specified performance. However even high-spec structural concrete has to be cheapish, with imperfect gap-filling and plenty of voids filled with cement slurry. High performance cementitious mixes such as tanking and certainly ASMET/Roofkrete depend upon very careful batching using fine aggregates (e.g. sand) possibly expensively selected, sieved or otherwise consistently graded for both size and shape. Not requiring gap-filling, high strength can be accompanied by low cement content. It is possible that ASMET/Roofkrete contains much less energy-intensive cement than you'd expect.
John admits to studying aggregate grains ("sand" to you and me, "silica-the-commonest-substance-on-earth" to him) under a microscope in his garage/lab. Asked how he trained in such science he says he just picked it up while self-building several houses but I think there's more here than meets the eye. He buys the aggregates already batched to his spec, then bags it with the cement, in his lab. The nature of the Portland cement, I have no idea. He hasn't tried substituting lower-energy lime.

Why does the ASMET need to be on both faces? The main strengths of the product mean that it is better placed on the outside face, this then flies in the face of typical vapour barrier locations. I understand why Jeff says to make a sandwich construction but that just adds to the cost and pretty soon it's not a viable proposition. Why not accept that it's on the outerface and then think how a secondary vapour barrier can be installed in the inner face, for example fully fill the 250mm ish webs with insulation, fix a batten to the web, pin a reflective vapour barrier with taped joints to the batten, plant a further batten or mf batten to create a cavity to improve u value further, provide services void and ensure vapour barrier not damaged, then fix wall finishes (they don't have to be plasterboard). These are just 1st thoughts on the problem and would need a condensation calc done (if any body knows how to with ASMET!).
There's more than one way to skin a cat, any observations?

I don’t have a proper understanding of vapour control layers and dew points. John Manniex was always going on about a precise control of water molecules within the structure is this possible? Is this branch of science to far above the building trade and more into space technology?

Jeff

An ASMET house can be anchored down on each corner or buried for basements. An average house would weigh about 25 tonnes - would that move in a wind? Should be easy to calc.

Cheers Steve

And I'm thinking that Roofkrete is the answer for green roofs, which otherwise have such an onerous requirement for a tough, perfect-forever, damage-protected membrane.
A big disappointment with green roofs is the need for through-ventilation above the insulation, beneath the membrane. The whole green roof then becomes merely a pretty weather-layer; it's great potential as a massive heat sink is cut off from any benefit it could give to the interior.
Roofkrete's strength makes the 'extensive' (as opposed to 'intensive') roof a little bit more feasible. That is, 500mm or more of soil, flowerbeds, trees etc, as opposed to a minimum-thickness, minimum-weight layer of specially drought-tolerant alpines. Such a soil layer, if thick enough, can contribute enough insulation value to keep the membrane warm enough (with more insulation under it) to avoid danger of interstitial condensation on its underside. so not requiring that through-ventilation. The roof's massiveness then becomes an asset available to the interior; such indeed is the key to the Hockerton approach to zero-energy.
Covering with soil also solves the worry (which John Manniex says isn't a problem) about thermal expansion of large jointless exposed roof membranes - any membranes, not just Roofkrete. The membrane, protected from extremes of temperature, not to mention UV radiation, has a much easier time and should really last forever.

Ahh Yes, This I can visualise! an ASMT dome buried in the ground, interseasonal storage (no insulation required,)just a single skin of roofkreet in contact with the ground as per greenshelter.com

Earth tube heating and ventilation?

Test reports from Portsmouth University
http://www.roofkrete.com/html/test_reports.html

Yes, look what happened to multifoil

Posted By: steveleighThe questions that were offered from the conference floor after the lecture and the conversations with the experts (professors of concrete technology) made it plain that all the books have been written and there is no way a new material could be introduced

Steveleigh,

I would like to say somethink positive about ASMET but there is an element of smoke and mirrors about it which tends to make me think it is too good to be true. I think your latest post is very defensive and I disagree with most of it.

In particular, I know several experts in concrete technology, a very highly regarded professor amongst them, who continue to publish new material about concrete despite your statement above. The material they publish is however, mostly subject to international peer review. These people may indeed be very vain, but publications are their livelihood, and anything which truly stacks up under scrutiny is a potential goldmine for them. This poses the question: if ASMET is the panacea you proclaim, why is there no peer reviewed material available?

Mike George
Where is the evidence of ‘international peer review’ before the great builders of the pass built the great structures of Europe (which academia continue to use as examples of great achievement). They were built by hands-on people and not academics. Does this prove that hands-on designer builders can manage to build great buildings without the need for academics? I think track record is the only true prove of any product. Academia can sometimes cloud simple issues which need experience more.