3½ in code floor, 4 in as built, 5 in where a second mat fits, and +⅜ / −¼ in of tolerance.
← Back to the calculatorEvery page on this query gives a table of thicknesses by use. That table is real and it is below. What it does not tell you is that thickness is decided by three separate constraints and you take the largest: the code floor, the depth a second mat needs, and the span rule. On a shed floor the code floor wins. On a 4 in slab with two mats the geometry wins, and it wins by a full inch.
Then there is the number nobody publishes: what the pour actually measures. ACI 117 allows +⅜ in and −¼ in, so 4¾ in against a 4 in spec fails on the high side — and it fails by costing 19% more concrete, not by being weak.
Two families of number are in this table and they are not the same kind of thing. The slab-on-ground rows come from a code that sets a floor — IRC R506.1 — and then a design standard, ACI 332 or ACI 360, that sets the actual thickness from the load. The suspended rows come from ACI 318, where the thickness is a deflection rule and has nothing to do with what the slab carries.
| What it carries | Thickness | Where the number comes from |
|---|---|---|
| Patio, shed floor, foot traffic only | 4 in | IRC R506.1 floor is 3½ in; 4 in is what gets drawn |
| Basement or crawlspace floor | 4 in | IRC R506.1 |
| Walkway, sidewalk | 4 in | IRC R506.1; local sidewalk specs often say 4 in over 4 in of base |
| Garage floor, car or pickup | 4 in | IRC R506.1 floor, then ACI 332 for the vehicle |
| Driveway, residential car | 5 in | ACI 332, design for the axle load |
| Driveway, truck, RV, bin lorry | 6 in | ACI 332 |
| Shop floor, loaded racking, forklift | 6 to 8 in | ACI 360, design for the load |
| Suspended one-way slab | l/20 to l/28 | ACI 318-19 Table 7.3.1.1 |
| Suspended two-way, no interior beams | ≥ 5 in, then l/33 | ACI 318-19 8.3.1.1 |
| Suspended two-way, with beams | ≥ 3½ in | ACI 318-19 8.3.1.1 |
The blue row is where the two families collide. A garage floor is 4 in on the IRC floor and 4 in under most residential ACI 332 details, and it is the one slab where people routinely try to put a second mat in. That is the check in section 03, and it does not pass.
Thickness is not a number you look up. It is the largest of three numbers, and knowing which one is in charge tells you what to do when a slab comes out wrong.
| Check | Formula | Governs when |
|---|---|---|
| Code floor | 3½ in on ground (IRC R506.1) | Nothing structural is happening — patio, shed, basement |
| Two mats fit | 2c + 2db + 1 in (ACI 318-19 25.2.2) | Any slab with top and bottom steel, which is most 4 in slabs people ask about |
| Span rule | l/20 to l/28, or l/33 two-way (ACI 318-19) | Suspended slab, any span over about 6 ft |
| Load design | ACI 332 / ACI 360, from axle or rack load | Vehicle, forklift, loaded racking, heavy storage |
The useful consequence: on a slab on ground the thickness almost never comes from strength. A 4 in slab of 4000 psi concrete carrying a car is not close to failing in flexure; it fails at the joint, at the edge, or on a soft spot in the base. Thickness on a slab on ground is a cover and constructability number, which is why the code floor is 3½ in and not 6 in. On a suspended slab the reverse is true: thickness is a stiffness rule and the steel follows it.
1. h_code = 3.5 in (slab on ground, IRC R506.1) 2. h_mats = 2c + 2db + 1 in (ACI 318-19 25.2.2) 3. h_span = l / 20 to l / 28 (ACI 318-19 Table 7.3.1.1) 4. h = max(h_code, h_mats, h_span, h_design)
The clear gap between two layers is what is left of the depth after both covers and both bar diameters, and ACI 318-19 25.2.2 puts a 1 in floor under it. Turn that around and you get the depth at which a second mat starts to exist.
| Bar | Cover 1½ in | Cover 2 in |
|---|---|---|
| #3 | 4.75 in | 5.75 in |
| #4 | 5.00 in | 6.00 in |
| #5 | 5.25 in | 6.25 in |
| #6 | 5.50 in | 6.50 in |
The blue row is the number to remember: two #4 mats at 1½ in of cover need 5 in. Not 4½, not 4¾ — five, because the 1 in gap is a hard floor and not something a chair can buy back. A 4 in slab drawn with a top mat is a drawing error, and it shows up in the field as two mats sitting on top of each other in the middle of the slab, which is worse than one mat in the right place.
The same arithmetic with 2 in of cover, which is what an exterior slab exposed to weather needs for a #6 and larger bar, pushes the minimum to 6½ in. That is why a driveway with two mats is 6 in and not 5.
Thickness is the first term in four downstream numbers. All four move together, and the last one is the only one people forget to re-check when the thickness changes.
1. d = h − c − db/2 (effective depth, in) 2. V = L × W × h ÷ 324 (yd³, L and W in ft, h in in) 3. bags = V × 45 (80 lb) or V × 60 (60 lb) 4. n_bars = L × 12 ÷ s + 1 per direction 5. W_reb = (n_x × W + n_y × L) × lb/ft
Worked on a 24 ft by 24 ft slab at 4 in, #4 at 18 in on centre both ways, 1½ in of cover:
d = 4 − 1.5 − 0.25 = 2.25 in V = 24 × 24 × 4 ÷ 324 = 7.11 yd³ (7.47 with 5% waste) bags = 7.11 × 45 = 320 bags of 80 lb n_bars = 24 × 12 ÷ 18 + 1 = 17 per direction W_reb = (17 × 24 × 2) × 0.668 = 545 lb of #4
| Thickness | yd³ per 1000 ft² | 80 lb bags | 60 lb bags |
|---|---|---|---|
| 3.5 in | 10.80 | 486 | 648 |
| 4 in | 12.35 | 556 | 741 |
| 4.5 in | 13.89 | 625 | 833 |
| 5 in | 15.43 | 694 | 926 |
| 5.5 in | 16.98 | 764 | 1019 |
| 6 in | 18.52 | 833 | 1111 |
| 8 in | 24.69 | 1111 | 1481 |
The 4 in row and the 5 in row differ by 1.78 yd³ on the same footprint, which is 80 bags or a short load. That is the price of the second mat, and it is the whole argument in section 03: you are not choosing between a thin slab and a thick one, you are choosing between one mat at 4 in and two at 5 in.
ACI 117 gives slab thickness a tolerance of +⅜ in and −¼ in, and ACI 302.1R repeats it. That window is asymmetric on purpose: too thin loses cover and section, too thick only costs concrete, so the code is generous upward and tight downward.
| Measured | Deviation from 4 in | ACI 117 verdict |
|---|---|---|
| 3½ in | −½ in | out — 2× the low limit |
| 3¾ in | −¼ in | at the limit, accept |
| 4 in | nominal | what the drawing says |
| 4⅜ in | +⅜ in | at the limit, accept |
| 4½ in | +½ in | out — 1.3× the high limit |
| 4¾ in | +¾ in | out — 2× the high limit, and 19% more concrete |
Two rows in that table are the ones that get argued about on site. 4¾ in is out of tolerance on the high side by 2× the limit, and the concrete to build it was never ordered: on 1000 ft² at 4 in, 12.35 yd³ becomes 14.66, so 2.3 yd³ — at $150 a yard, about $350 that appears on nobody's estimate. And 3¾ in is exactly at the low limit, which means it passes, which surprises people who assume a quarter inch off a 4 in slab is a defect.
The reason the low side is only ¼ in is cover: on a 4 in slab with #4 at 1½ in of cover the bar centre sits 1.75 in down, and ¼ in less depth is 14% less effective depth. That is a real loss of capacity on a suspended slab and a rounding error on a slab on ground, which is why the tolerance bites differently depending on which of the three checks from section 02 was in charge.
Four inches for anything that only carries people, five inches for a car, six inches for a truck or an RV, and a suspended slab is not picked from a chart at all: it is the span divided by 20 to 28, per ACI 318-19 Table 7.3.1.1. The code floor for a slab on ground is 3½ in, IRC R506.1, and almost nobody builds to it because 4 in is the smallest depth that a crew can place and finish reliably.
3½ in, IRC R506.1. That is the floor for a slab that carries nothing but its own weight and foot traffic. Two things push it up: a vehicle, which hands the design to ACI 332, and a second layer of rebar, which needs 2c + 2db + 1 in of depth — 5 in for two #4 mats at 1½ in of cover, before any load is considered.
For a car, on a good base, in a warm climate, yes, and it is what a lot of residential driveways are. It is below what ACI 332 recommends for anything heavier, and it leaves no depth for a second mat. The thickness is not the weak point on a 4 in driveway — the base and the joint spacing are, which is why a 4 in driveway with 10 ft panels outlasts a 6 in one poured in one continuous strip.
No. Two #4 mats at 1½ in of cover need 1.5 + 0.5 + 1.0 + 0.5 + 1.5 = 5 in, and the 1 in in the middle is the ACI 318-19 25.2.2 minimum clear gap between layers, not a comfort allowance. At 4 in the arithmetic gives zero clearance, which means the bars touch, which means the top mat sits at the depth where the bottom mat should be. Either go to 5 in or use one mat.
7.1 cubic yards. 24 × 24 × 4 ÷ 324 = 7.11 yd³, and at 5% waste, 7.5 yd³. In bags that is 320 bags of 80 lb before waste, which is why anything over about 2 yd³ gets delivered rather than mixed. One extra inch of thickness on the same slab is 1.8 yd³ more, or 80 more bags.
Plus ⅜ in and minus ¼ in, ACI 117, repeated in ACI 302.1R. On a 4 in spec that means 3¾ to 4⅜ in passes. The high side is the one nobody checks: a slab that measures 4¾ in is 19% more concrete than the drawing called for, and on 1000 ft² that is 2.3 yd³ nobody ordered.
Drill a ¼ in hole and hook a bent wire or a piece of coat hanger through it until it catches the underside, then measure the wire. Three or four holes on a 400 ft² slab, at the joints and in the middle of a panel, because the thickness varies more across a slab than it does from slab to slab. Ultrasound and ground-penetrating radar will do it without a hole, and both need a calibration point from a hole anyway.
Span over 20 for a simply supported one-way slab, over 24 with one end continuous, over 28 with both ends continuous, over 10 for a cantilever — ACI 318-19 Table 7.3.1.1, for 60 ksi steel and normal weight concrete. A 16 ft simple span is 9.6 in, so 10 in. That table only applies when the slab does not support partitions likely to be damaged by deflection; otherwise the thickness comes out of a deflection calculation, which is usually thicker.
Sometimes, and never by much. Thickness buys effective depth, and more depth means a smaller steel area for the same moment — but the minimum reinforcement for temperature and shrinkage is fixed at 0.0018 of the gross section, ACI 318-19 24.4.3.2, so it grows with thickness. Going from 4 to 6 in cuts the steel the moment needs and raises the steel the code requires. On a slab on ground the second effect wins, which is why a 6 in slab carries more steel than a 4 in one.
Not the structural thickness. A vapor retarder under a slab on ground is IRC R506.2.3 and it does not add depth, but it does change how the slab is placed — concrete over a membrane has to be pulled, not dumped and spread, or the membrane rucks and the slab thickness varies by half an inch across a pour. Insulation under the slab changes the frost depth question, IRC R403.1.8, not the thickness.