Twenty-Six Hours of Warning and One Number That Decides Everything
The gauge eleven miles upstream is rising at about 40 millimetres an hour and the forecast crest lands sometime before dawn tomorrow. Standing at the bottom of the garden with a laser level, the water is going to come over the low point of the boundary by something like four hundred millimetres. That is the whole brief: 400 millimetres of water, a line that has to climb to high ground at both ends, and one working day with whoever you can get hold of.
What decides whether that day succeeds is not technique. It is arithmetic done in the first twenty minutes, because a sandbag barrier is one of the very few pieces of construction where the quantity is also the schedule. Bags do not arrive filled. Every one of them is a shovel, a pair of hands and roughly forty seconds, and the count you arrive at is therefore a direct statement about how many people you need standing in a yard by nine o'clock. Get the count wrong and you do not end up with a slightly undersized barrier; you end up at four in the morning with a gap.
The trap is that height compounds twice. Raise the design level by 300 millimetres and you add three courses — but a barrier that is going to stay upright and stay watertight has to widen as it grows, so those three courses arrive on top of a base that is itself wider. Working it as a straight single stack, the way most people first sketch it, produces a figure somewhere between a third and a sixth of the real one. That single mistake is why sandbagging so often stops half-finished with the sand heap empty.
The Line Is Set by the Contour, Not by the Building
Barriers of this kind almost never fail by being overtopped. They fail at their ends, where the water walks round a barrier that stopped short of ground high enough to hold it, and then arrives behind the thing you spent the day building. So the first job is not measuring the frontage. It is finding, at each end of the run, a point where existing ground sits above the design water level, and committing to reaching it. A staff and a level does this in ten minutes; a laser and a pole does it in five; a garden hose full of water does it slowly and perfectly well.
Length measured that way is always longer than the elevation you are protecting, and often by a lot. The line steps out around a bay window, returns up a driveway to meet the kerb, doubles back at a gate that has to stay usable, and picks up two side returns to reach the rising ground the survey found. Thirty metres of house frontage is comfortably thirty-eight metres of barrier once those are added, and every one of those extra metres is bought at the same rate per metre as the ones in front of the door. Walk the line with a tape or a measuring wheel and write the figure down before anybody starts filling.
The other half of routing is deciding what the barrier cannot do. Water that arrives through the ground, up a foul or surface-water connection, or through an air brick below the design level is not affected by anything standing on the surface, and a barrier that rings a property whose drains are still connected to a surcharging sewer builds a bathtub rather than a defence. Below-ground routes need non-return devices or plugs, and a property that will fill from behind regardless is a property where the honest answer is to move what matters upstairs and spend the day on that instead. BS 85500, Flood resistant and resilient construction, and FEMA P-312, Homeowner's Guide to Retrofitting, both frame the decision the same way: keeping water out is one strategy among several, and it is only the right one where the whole envelope can actually be held.
Height, Freeboard, and the Section That Follows From It
Design level is the forecast peak plus freeboard, and freeboard on an emergency barrier is not a refinement. Forecast crests carry real uncertainty, wind across an open flood plain drives a surge and a wash against the face, a passing vehicle in floodwater pushes a wave, and the barrier itself settles by a course thickness or so as the bags bed into one another and into soft ground. Two hundred to three hundred millimetres above the predicted peak is a defensible working allowance for a domestic run, and engineered levees carry freeboard for the same reasons — USACE EM 1110-2-1913, Design and Construction of Levees, treats it as a design quantity rather than a margin of comfort.
The section is then fixed by a rule that flood-fight practice has used for a long time and that the Corps' own flood-fight handbooks set out: build the dike as a pyramid whose base width is about three times its finished height. The reason is only partly stability. Half a metre of water pushes about 1.2 kilonewtons against each metre of barrier, and 900 millimetres pushes about 4 — modest forces against a wall of wet sand weighing the better part of a tonne per metre run, and easily resisted by friction alone. What the width is really buying is seepage path. Water finds its way between and beneath bags, and the longer the route it has to take through the stack and under the toe, the slower it arrives and the less likely it is to start moving soil with it.
Nine courses, around 900 millimetres, is where hand-built sandbag work stops being sensible. Above that the bag count per metre has grown past the point where a small crew can build any useful length of it, the base has to be nearly three metres wide, and the failure mode shifts from seepage to underseepage and piping, which is an engineering problem rather than a shovelling one. Anything taller belongs to a designed temporary defence, or to one of the demountable barrier systems tested to PAS 1188, Flood protection products, rather than to bags.
One more constraint applies when the line runs tight against the building. Stacking bags against a wall turns that wall into the dam, and masonry and framed walls take hydrostatic load and the resulting seepage badly. FEMA's retrofitting guidance limits dry floodproofing to shallow depths for exactly this reason. Standing the barrier a metre or two off the wall where the ground allows is better on every count: the wall is not loaded, the seepage that gets through has somewhere to collect, and there is room to work a pump.
| Finished height | Courses | Bag positions across the base | Bags per foot of run | Bags per metre of run | Multiply a single stack by |
|---|---|---|---|---|---|
| 200 mm (8 in) | 2 | 2 | 3 | 10 | 1.5 |
| 400 mm (16 in) | 4 | 5 | 12 | 39 | 3 |
| 600 mm (24 in) | 6 | 7 | 24 | 79 | 4 |
| 900 mm (36 in) | 9 | 11 | 54 | 177 | 6 |
What a sandbag dike is made of
- Polythene on the wet face — the layer that actually resists water, run under the toe on the wet side and drawn up over the crest so no seam faces upstream
- Capping course — the course that sets the finished level, laid centred so the crest is flat enough to walk and to hold the sheet down
- Stepped body courses — each course set back over the one below, which is where the bag count multiplies and why a straight stack undercounts so badly Sandbag Calculator
- Bond course — laid across the line rather than along it and trodden down into the key, so the widest and heaviest row is also the one gripping the ground
- Stripped and keyed strip — turf, mulch and soft mud taken off and a shallow key scraped, because a barrier bedded on grass has a drainage channel under it
Counting the Bags
The calculator below works the honest baseline: one filled bag covers roughly a foot of wall in each course, so a single-width stack is the run in feet multiplied by the number of courses. That figure is exactly right for the two- and three-course work that keeps a puddle out of a garage, and it is the number every wider section is built from, so it is the one to get on paper first.
For anything that needs a pyramid, take that baseline and multiply it by the figure in the last column of the table above — or, if you would rather see it built up, run the calculator once per lane and add the results, entering the length of each lane as it shortens towards the crest. Both routes arrive at the same place, and the second has the advantage of showing you where the bags actually go, which matters when the sand runs low and somebody has to decide which lane to drop.
Worked through on the job in front of us: 38 metres of line, 600 millimetres finished, six courses. The single-stack baseline is 124.7 feet times six, which the calculator rounds up to 749 bags. The section multiplier at that height is four, so the real requirement is around 3,000. Add five to ten per cent for bags that split on the pallet, that are dropped and burst, and that get used up ballasting the sheet and building the two end returns, and the number to work to is 3,200. Those two figures — 749 and 3,200 — are worth writing next to each other on the same piece of paper, because the gap between them is the entire subject of this article.
Three thousand two hundred is also the moment to test the plan against the clock rather than against the budget. If the crew cannot produce that many bags in the hours left, the answer is not to start and hope. It is to shorten the line to the openings that actually matter, drop the height by a course and accept a wet floor rather than a wet room, or spend the money on a hired demountable barrier while there is still time for it to arrive.
Put in the length you walked and the number of courses your design level needs, and read the single-width baseline — then multiply it by the section factor from the table above before anybody phones a supplier.
The total length of the barrier you're building.
How many bags high the barrier will be stacked.
Sandbags needed
98 sandbags
This is a rough planning estimate — taller barriers need a wider, pyramid-shaped base (each course set back slightly) for stability, which uses more bags per unit of length than a simple straight stack.
- Barrier length
- 49 linear ft
They open the calculator with your figures already in it
Sandbag Calculator: 98 sandbags — shown in imperial, US market. The link sets both, so the result they see is the one on your screen.
Estimated cost — your price
This site holds no price list for this material — local prices vary too much to publish honestly. Enter your supplier's price and the result is costed with it.
What this calculation does not cover
- Counts a single-width stack — one bag per foot of run in every course. A barrier that actually has to hold water is built as a stepped pyramid with a base several bags wide, so the real bag count is a multiple of this figure, and the multiple grows with height.
- Takes the length you enter as the finished barrier. It adds nothing for the end returns that have to climb onto ground standing above the water level, or for the extra run around bays, driveways and gates — and the ends are where barriers of this kind usually fail, not the crest.
- This is a bag count, not a flood defence design. It says nothing about seepage through and under the stack, piping in the ground beneath the toe, or the hydrostatic load a barrier stacked hard against a wall puts into that wall, and it cannot address water arriving through drains, air bricks or other below-ground routes that no surface barrier blocks.
- No waste allowance is included. Bags split on the pallet, burst when they are dropped, and get spent ballasting sheeting and holding laps down; none of that is in the number.
- Courses are entered directly. Nothing relates them to the water level you are protecting against or to freeboard above it, and nothing here gives the sand volume, the delivery weight or the crew-hours to fill this many bags — on a long barrier the fill rate, not the material, is usually what decides whether the job finishes.
The Sand Behind the Bags
A bag filled the way it should be — half to two thirds, so it slumps and moulds against its neighbours — holds somewhere around 8 to 11 litres of sand, roughly a third of a cubic foot, and finishes at 13 to 18 kilograms. That gives a planning rate of about 100 bags to the cubic metre and about 80 to the cubic yard. The 3,200 bags on this job therefore need in the region of 32 cubic metres, call it 42 cubic yards, and at a damp bulk density in the 1,600 to 1,700 kilograms per cubic metre band that is a little over 50 tonnes of sand.
Which sand matters more than people expect. A sharp or concrete sand — the grading described by ASTM C33, Standard Specification for Concrete Aggregates, or by BS EN 12620, Aggregates for concrete — has enough coarse fraction to stay in a woven bag and enough fines to pack. A fine blinding or dune sand pours out through the weave and out of the folded neck, and a heap of it turns to slurry the moment the barrier starts leaking. Slightly damp is ideal; sand that has been rained on all week is heavier to shift for the same volume, and sand delivered frozen in lumps cannot be filled from at all until it thaws. Bulk density is a measured property rather than a habit — ASTM C29, Standard Test Method for Bulk Density and Voids in Aggregate, is the method behind whatever figure your supplier quotes — so ask for the number and use theirs rather than the one on this page.
Order in the unit the yard sells in. Loose sand is sold by weight almost everywhere, which means the conversation is about tonnes and the volume is your problem, and it is worth doing that conversion out loud on the phone so a misunderstanding does not arrive on a lorry. Where access is the binding constraint rather than quantity, half-cubic-metre bulk bags at around 800 kilograms are the alternative: 50 tonnes is about sixty-five of them, they can be craned over a fence or set down individually along the line, and they cost considerably more per tonne than the same sand tipped loose.
Loads are where volume and weight part company. The load count below comes from volume, and it is a floor rather than an answer, because a full bed of damp sand will often reach a truck's legal weight before it reaches the top of its sides. In North America gross weights are governed by the federal limits and bridge formula in 23 CFR Part 658 together with the state's own rules, and in the United Kingdom by the Road Vehicles (Authorised Weight) Regulations 1998 — so a hauler quoting a smaller payload than the bed suggests is not being difficult. Settle the practical questions with the same call: whether the vehicle can reach the line at all on saturated ground, whether there is 4 metres of clear height to raise a body, and where a 13-tonne heap can sit without blocking the road the crew are using. Tipping the sand in two or three places along the run rather than one costs an extra movement and saves hours of barrowing.
| Bags | Sand volume | Cubic yards | Approximate weight, damp | Half-cubic-metre bulk bags |
|---|---|---|---|---|
| 500 | 5 m³ | 6.5 yd³ | 8.3 t | 10 |
| 1,000 | 10 m³ | 13 yd³ | 16.5 t | 20 |
| 2,000 | 20 m³ | 26 yd³ | 33 t | 40 |
| 3,200 | 32 m³ | 42 yd³ | 53 t | 64 |
Enter the sand volume the bag count implies and pick the bed your hauler actually runs — then treat the answer as the minimum number of movements, and confirm the payload by weight before the first one is booked.
The total material volume to haul.
The hauling capacity of the truck being used.
Truck loads needed
5 x 12 yd³ loads
Trucks are also limited by legal weight, not just volume — dense materials like gravel or wet soil may reach a truck's weight limit before filling its full volume capacity, requiring more, partial loads.
They open the calculator with your figures already in it
Dump Truck Load Calculator: 5 x 12 yd³ loads — shown in imperial, US market. The link sets both, so the result they see is the one on your screen.
Estimated cost — your price
This site holds no price list for this material — local prices vary too much to publish honestly. Enter your supplier's price and the result is costed with it.
What this calculation does not cover
- This counts volume only. It does not check payload, axle loading or gross vehicle weight, and a body filled with gravel, sand, wet clay or broken concrete normally reaches its legal weight limit before it is full — so dense material takes more trips than the volume count shows. It is not a legal load check; what the truck may legally carry is the hauler's call.
- The figure you enter is treated as loose, as-loaded volume. Excavated soil and demolition debris swell once they are dug out, and delivered fill loses volume once it is compacted in place, so a bank (in-ground) or compacted-in-place quantity entered here undercounts loads. Convert to a loose volume before using this.
- The three bed sizes are generic fleet sizes, not your truck. Rated capacity moves with body length, sideboard height and whether the quoted figure is struck (level) or heaped, and liners, tailgate gear and a worn body all change what actually goes in. Confirm the number with the hauler or rental yard.
- Nothing here covers getting the truck to the material. Site access and turning room, overhead clearance, tipping space, posted road and bridge weight limits, seasonal load restrictions and permits all change how many trips are actually possible, and none of them are modelled.
- One material, one truck type, and no cost. Loads that cannot be mixed — segregated waste streams, separate stockpiles — need their own counts, and haulage rates, tipping fees and disposal charges (often billed by weight or per ticket rather than per load) sit outside this calculation.
How Fast a Crew Actually Produces Filled Bags
Filling is a three-person operation: one holding the bag open with the cuff rolled back, two shovelling in turn. A team that has found its rhythm produces somewhere between fifty and a hundred filled bags an hour, and that band is wide because it depends almost entirely on how far the shovel travels between the heap and the bag. Time your own crew's first quarter of an hour and use their rate rather than this one — it is the only measurement in the whole job that costs nothing to take and changes every subsequent decision. On the 3,200-bag example, even the optimistic end of that band is over thirty team-hours: four teams working a long day, or twelve people, before anyone has carried a bag to the line.
That is the number that justifies a machine. A filling frame or a stand that holds four to six bags open at once roughly doubles a hand crew, costs almost nothing, and can be knocked together from timber offcuts. A hopper loaded by a mini excavator or a loader bucket changes the order of magnitude entirely, and if one is available within reach it is worth an hour of the morning to fetch it. So are pre-filled bags: many local authorities, builders' merchants and plant hire yards hold stock during a flood warning, and a pallet of ready-filled bags collected at eight o'clock is worth more than any amount of afternoon effort.
Fill fraction is a safety decision as much as a construction one. A bag filled to the brim weighs upwards of 25 kilograms, will not conform to the bag beneath it, and leaves a void at every joint; a bag at half to two thirds sits at 13 to 18 kilograms and moulds. Repeated lifting at that weight, at ground level, twisting, for hours, is exactly the exposure the NIOSH Applications Manual for the Revised Lifting Equation is written about, and the practical translation is short chains with a person every few metres rather than long carries, alternating who fills and who places, and a genuine break every hour. Gloves and eye protection are the minimum — sand in the eyes ends someone's day quickly — and the personal protective equipment requirements of 29 CFR 1926 Subpart E are the reference on a job with any commercial involvement. Do not tie the bag: the neck is folded under the bag as it is laid, so the weight of the bag closes it, and a tied bag holds a stiff sausage shape that will not seal against anything.
Laying the Courses
Preparation is the ten minutes most often skipped and the one that decides whether the barrier leaks along its whole length or only in places. Turf, leaf litter, mulch and soft mud all form a continuous drainage layer under a barrier laid straight onto them, and a scraped, keyed strip removes it. On a hard surface — a drive, a path, a road — there is nothing to key into, so the sheet does the whole job and the bags simply hold it down.
After that the work is bricklaying with a very forgiving unit. Bags go down across the line rather than along it in the bond course, each one trodden flat before the next arrives, and every joint is broken by the course above in the same way a stretcher bond breaks a brick wall. A bag placed and left unstamped is a bag with a void under it.
- Strip the line and scrape a shallow key, 50 to 100 millimetres deep and a bag-length wide, along the whole run including the end returns.
- Roll the polythene out on the wet side, run it into the key so the toe of the barrier lands on top of it, and leave enough sheet to reach over the crest later. Sheeting for this work is the class specified by ASTM D4397, Standard Specification for Polyethylene Sheeting for Construction, Industrial, and Agricultural Applications; 250 micron, six mil, is the usual weight.
- Lay the bond course across the line, folded neck tucked under and facing the water, and stamp each bag flat.
- Build each following course set back over the one below on both faces, staggering the joints, and stamp as you go — the stepped face is what turns a stack into a section.
- Keep the ends climbing: the last two or three metres of each return rise with the ground until the crest is above the design level on natural ground, not on made-up material.
- Lay the capping course centred on the crest, flat enough that somebody can walk it with a torch during the night.
- Draw the sheet up the wet face and over the crest, and hold it there with a single row of bags laid on the dry side of the cap so no seam or free edge faces upstream.
- Walk the finished line looking along it rather than at it — a course that has drifted, a low spot, or a gap where the run meets a wall shows up in a sight line and not in a close inspection.
It Will Leak, So Build the Inside of It Too
A sandbag barrier is a delay, not a seal. Water arrives through the stack, under the toe and around anything the sheet could not be dressed against, and the volume it brings is entirely manageable if it has somewhere to go and something to move it. That means a low point inside the line — a natural hollow, a scraped sump, or a lined pit — with a pump in it and a discharge run that ends outside the barrier and far enough downstream that it is not simply recirculating. Petrol pumps do not care about the power cut that a flood usually brings with it; a mains pump on the same supply as everything else is a plan with one obvious failure.
The polythene is what does most of the actual watertightness, and where it is placed decides whether it earns its keep. It belongs on the wet face and under the toe, so that water pressure holds it against the barrier rather than lifting it, and it should be a single width along the run wherever possible. Where sheets must be joined, lap them generously with the upstream sheet over the downstream one, and hold the lap down under bags rather than trusting tape. A sheet dressed up a wall and over a threshold does more for a doorway than three extra courses.
Then somebody watches it. Rising water at the toe on the dry side, a patch of ground going soft, or water arriving clear and then turning cloudy are all worth acting on immediately, because cloudy water means soil is moving and that is the beginning of a piping failure rather than a leak. The response is bags on the inside toe to weigh the ground down and lengthen the path, never digging on the dry side to let it drain. Two-hourly walks with a torch through the night, a marked stick in the water to see whether the level is still climbing, and a decision agreed in advance about the point at which everyone leaves regardless of what the barrier is doing.
After the Peak
Sand that has been in floodwater is not the sand you bought. Floodwater carries sewage, fuel, agricultural run-off and whatever was in the sheds it passed through, and the bags at the wet toe have been filtering it for a day. Treat the material as contaminated: it does not go on a vegetable garden, it does not become a children's sandpit, and where it is going to a waste facility the classification rules apply — in the United Kingdom that is Environment Agency Technical Guidance WM3, Guidance on the classification and assessment of waste, and elsewhere the local waste authority's equivalent. Bags from the dry back face that never saw water are usually fine to keep, and separating the two as the barrier comes down costs nothing.
Keep the empties and the survey. Woven polypropylene bags lose strength quickly in sunlight, so unfilled stock stored in the dark stays serviceable for years while the same bags left on a pallet outside are brittle by the following season. The more valuable thing to store is the piece of paper: the line you walked, the length, the design level and the bag count that came out of it. Next time the gauge rises, that sheet turns twenty minutes of measuring and arithmetic into a phone call, and on this job the twenty minutes is the part you cannot buy back.
Fix these before anybody picks up a shovel
Six figures decide whether the day works, and all six can be settled in the first half hour. The stack opens on the 38-metre, six-course line worked through above; change the length and the courses to your own and the rest follows.
- Design level: forecast peak plus 200 to 300 mm of freeboard — Wind wash, vehicle wake, forecast error and the barrier's own settlement all eat into the margin. Set it once, mark it on a stake at each end, and build to the marks rather than to a tape reading taken later.
- Length walked to high ground at both ends, not the frontage — Returns, bays, driveways and the climb onto natural ground above the design level. Thirty metres of elevation is routinely thirty-eight metres of barrier, and the extra costs the same per metre as the rest.
- Section: base width about three times the finished height — Width buys seepage path more than it buys stability. At 600 mm finished that is seven bag positions across the base and roughly four times the single-stack bag count.
- Bag count with 5 to 10 per cent on top — Splits, bursts, the ballast row over the sheet and the two end returns. The waste allowance is small and the consequence of omitting it lands at the worst possible hour.
- Sand ordered by weight, delivered in more than one place — About 100 bags to the cubic metre, 80 to the cubic yard, and 1,600 to 1,700 kg per cubic metre damp. Confirm the yard's own bulk density, and split the tip points along the run to cut the barrow distance.
- Crew rate measured in the first fifteen minutes — Fifty to a hundred filled bags an hour per three-person team is the planning band. Measure the real one early — it is what tells you whether to fetch a hopper, buy pre-filled bags, or shorten the line while there is still time.
Opens the calculators above on one screen with the dimensions from this article already filled in. Quantities only — this site publishes no price list, because local prices vary too much to publish honestly.