Self-Flow Castable in Practice: Flow Value, Water Addition and Formwork Sealing

Self-flow castable fails in two predictable ways: extra water when it will not move, and slurry leaking through formwork when it moves too well. This note fixes the three hard limits — flow spread 180–220 mm, water at 5.0–6.5% within ±0.2%, walers at 600 mm or less — and lists seven field checks plus the locations where self-flow is the wrong choice.

Self-Flow Castable in Practice: Flow Value, Water Addition and Formwork Sealing

Self-flow castable usually fails on site in one of two ways. Either it will not move, so someone adds water and halves the strength, or it moves too well, leaks through the formwork and leaves a honeycombed surface once the mould is stripped. Neither is a material defect. Both come from treating self-flow mix as ordinary castable with extra water. It spreads under its own weight at a low yield stress instead of being compacted by vibration, and that single difference reverses most of the field rules.

Self-flow is not "thinner castable"

A vibratable castable only needs enough water to be placed, typically 4.8–5.5% for low-cement systems, because the poker does the work of rearranging particles and expelling air. Self-flow mixes use a continuously graded particle distribution with small doses of superplasticiser and a workability retainer so the paste yields and flows on its own. Water demand is therefore higher by design, around 5.0–6.5% for low-cement grades. The trade-off is slightly higher apparent porosity and slightly lower bulk density. Self-flow is not formulated to be denser than a vibratable mix; it is formulated to reach adequate density where vibration is impossible. That is the frame for every rule below: if the area can be formed and vibrated, a vibratable mix is the better buy. Self-flow earns its price where anchors are dense, geometry is awkward, or no one can get a poker into the space.

Flow value: work inside 180–220 mm

Flow spread is the one property you can check in minutes without waiting a day for strength. A flow cone on a jolting table is enough. Treat 180–220 mm as the working window. Below 160 mm the mix will heap into a cone and bridge at corners, leaving voids behind the formwork; correct it with a small addition of water plus a little extra superplasticiser and two minutes of remixing, never with one big splash. Above 240 mm segregation has already begun: aggregate settles, paste rises, and the pour ends up with a lean bottom and a rich, weak top. That batch is better scrapped. Test every batch, or every five tonnes, and retest whenever the supplier batch or the ambient temperature changes. Superplasticiser response is temperature sensitive, and morning-to-afternoon spreads can differ by more than 20 mm on the same mix.

Water: half a point over the limit starts segregation

Weigh the water. Eyeballing it, or running a hose straight into the mixer, removes the only control that matters. Dry mix for one to two minutes, add the weighed water in one go, then wet mix for three to five minutes, keeping total mixing under six minutes. Longer mixing drags in air and consumes the workability retainer before the mix reaches the formwork. Keep water between 5.0% and 6.5% of dry weight with a field tolerance of plus or minus 0.2%. Going half a percentage point past the upper limit typically costs 15–25% of cold crushing strength and adds two to three points of apparent porosity, which also means deeper slag penetration. The more common mistake is a mix that stiffens in the bucket after forty minutes and gets "revived" with more water. The spread recovers, but hydration has already advanced and the placed material reaches roughly half its design strength. Discard it.

Formwork: self-flow pushes sideways almost like water

This is where most failures start. In its fluid state a self-flow mix exerts lateral pressure close to hydrostatic, while a vibratable mix sits semi-dry and pushes far less. Reusing a vibratable-mix formwork design for self-flow has produced bulging, blowouts and shifted panels. Space walers and supports at 600 mm or less, wedge the braces tight, run foam gasket plus tape along every joint, and close corners and penetrations separately. Release agent goes on thin and even, with no puddles. Leakage is worse than it looks: paste escapes, aggregate stays, and the stripped surface is honeycombed. Such defects usually sit inside the lining where patching cannot reach them, and they spall in sheets after a month or two of thermal cycling. Before pouring, walk the formwork with a torch held against the outside. Any light coming through a joint will pass slurry.

Placing and stripping: no poker, no cold joints, 24 hours

Space discharge points 1–1.5 m apart so the free flow distance stays under 1.5–2 m. Beyond that the leading edge has lost water and arrives aggregate-rich. Place continuously, keeping the interval between layers under 30 minutes to avoid cold joints. Never insert a poker: it drives aggregate down and paste up, which is segregation by another name. Release air by tapping the outside of the formwork with a mallet, working from the bottom upward. Stripping depends on temperature. Above 10°C, side forms come off at 24 hours; load-bearing and suspended sections at 48 hours. Below 10°C, add 12 hours for every 5°C drop. Cure damp for at least 24 hours with plastic sheeting or a curing compound, because low-cement systems that dry out develop a dusty, weak surface. Dry out at 15–20°C per hour to 150°C and hold for four to six hours to release free water, then raise to 25–30°C per hour. Rushing this last stage wastes all the density gained earlier.

Where self-flow is the wrong choice

Skip self-flow for linings thinner than 60 mm, where the mix neither flows nor stays; for slopes steeper than 30°, where the whole mass slides unless you build stepped retainers; for severe impact and abrasion zones such as chute bottoms and separator target areas, where the higher porosity shortens life compared with a vibrated or silicon-carbide grade; and for large emergency repairs where sealed formwork is not realistic, because leakage and blowout cost more than the labour saved. A checklist to take to the site: flow spread 180–220 mm tested per batch or per five tonnes; water weighed at 5.0–6.5% with ±0.2% tolerance; total mixing four to six minutes with no second water addition; joints gasketed and taped with walers at 600 mm or less; discharge points 1–1.5 m apart and layers within 30 minutes; no poker, mallet only; side forms at 24 hours and load-bearing forms at 48 hours above 10°C. If any one line cannot be met, self-flow is the wrong material for that location.

Related Articles

Winter Castable Placement: Insulation & Anti-Freeze

Low temperatures slow cement hydration, cut early strength and can cause frost damage. This article covers material temperature, water temperature, curing insulation and anti-freeze measures for winter placement.

17513376789
WeChat QR codeWeChat · 17513376789