Two products share the word — one is the connection, the other is a coating that costs you embedment
← Back to the calculatorThe adhesive is the one you want when the bar goes into cured concrete. A #4 bar takes a 5/8 in hole, 16.5 in deep, 1.82 in³ of adhesive per hole, and a 10 oz cartridge fills nine of them.
The coated bar is a different purchase altogether: a factory coating to ASTM A775 on bar that goes into chloride exposure. It is not a glue and it does not improve grip — the coating factor raises development length by 20% or 50% depending on cover and spacing, so a #4 dowel's embedment goes from 16 in to 20 in or 24 in. The rest of this page keeps the two apart, because a quote, a drawing and a search box all use the same word for both.
A cartridge of anchoring adhesive and a bundle of green bar are sold by different people, specified in different sections and tested to different standards. They share a word and nothing else. Reading the two columns side by side is the fastest way to see which one a project actually needs.
| Injected adhesive | Epoxy-coated bar | |
|---|---|---|
| What it is | Two-part epoxy or hybrid mortar in a cartridge, mixed in a nozzle and injected into a drilled hole | Deformed or plain bar with a shop-applied epoxy film, electrostatic spray, to ASTM A775 |
| What it does | Bonds the bar to the hole wall and transfers load in shear over the embedment | Keeps chloride and moisture off the steel surface |
| Where it belongs | Setting a bar into cured concrete — a dowel into a wall, a footing, an existing slab edge | Bridge decks, marine and parking structures, de-icing salt exposure — anywhere the bar is at risk of corrosion |
| Effect on the arithmetic | Sets the hole diameter, the drilled depth and the cartridge count | Raises development length by 20% or 50%, so it changes the embedment the drawing has to show |
| Governing document | ICC-ES evaluation report for that adhesive, under AC308; design to ACI 318-19 Chapter 17 | ASTM A775 for shop-applied coating, ASTM A934 for prefabricated; bond factor in ACI 318-19 Table 25.4.2.5 |
| What it does not do | Does not protect the bar outside the hole | Does not glue anything, and does not improve bond — it costs bond |
Swipe the table sideways for more columns →
The load path is short and it does not start at the glue. Bar, then adhesive, then hole wall, then concrete. The adhesive's only job is to fill the annulus completely and stay there; the surface that carries the load is the bonded area of the bar, and that is a plain cylinder:
That number is why the glue brand is the last variable, not the first. Doubling the embedment doubles the bonded area; changing the cartridge changes the mixing ratio and the cure time, not the area. It is also why a dusty hole fails: the load has to cross the wall, and dust is a layer between the adhesive and the wall.
| Bar | db (in) | Bonded area per inch of embedment (in²) | Embedment (in) | Bonded area per dowel (in²) |
|---|---|---|---|---|
| #3 | 0.375 | 1.18 | 12 | 14.1 |
| #4 | 0.500 | 1.57 | 16 | 25.1 |
| #5 | 0.625 | 1.96 | 19 | 37.3 |
| #6 | 0.750 | 2.36 | 23 | 54.2 |
| #7 | 0.875 | 2.75 | 34 | 93.5 |
| #8 | 1.000 | 3.14 | 38 | 119.4 |
Swipe the table sideways for more columns →
Scale that to a joint and the cartridge count falls into place. A 20 ft joint at 16 in on centre is floor(240 ÷ 16) + 1 = 16 dowels:
The hole table, the bit sizes and the cleaning sequence are already worked out per bar size on the rebar dowel page, and this page does not repeat them. What matters here is the arithmetic that separates the two products: adhesive volume is a property of the hole, and it does not change if the bar is coated. A coated #4 in a 5/8 in hole still takes 1.82 in³.
ACI 318-19 gives epoxy-coated bar a coating factor that multiplies development length. It is 1.2 when cover is at least 3db and clear spacing at least 6db, and 1.5 in the tighter case — which, in a dowel row at 12 to 16 in on centre with 2 in of cover, is the case you are usually in. Coating is a corrosion decision, and this is its structural price.
| Bar | Uncoated ld (in) | Coated, ψe 1.2 (in) | Extra | Coated, ψe 1.5 (in) | Extra |
|---|---|---|---|---|---|
| #3 | 12 | 15 | +3 | 18 | +6 |
| #4 | 16 | 20 | +4 | 24 | +8 |
| #5 | 19 | 23 | +4 | 29 | +10 |
| #6 | 23 | 28 | +5 | 35 | +12 |
| #7 | 34 | 41 | +7 | 51 | +17 |
| #8 | 38 | 46 | +8 | 57 | +19 |
Swipe the table sideways for more columns →
Eight inches per bar sounds like nothing until it is multiplied. The same 20 ft joint, 16 dowels, both coating cases:
There is a second cost that is easier to miss: the extra embedment has to fit. A 20 ft joint detail that was drawn with 16 in of embedment and 12 in of projection becomes a 36 in bar, and a bar that long may no longer come out of the offcut pile — it comes off a new stick. That is an order-quantity change, not a detail change.
The obvious objection is that a joint dowel is not bonded on both sides anyway. A round dowel is sleeved on one half so that half can slide; a plate dowel sits in a pocket former. So where does coating earn its place?
At the joint face. A dowel crosses a joint, which means there is a plane through the slab where the bar is interrupted by a gap, and that gap is the one place chloride, de-icing salt and water reach the steel directly. Corrosion in a dowel does not usually start in the middle of the embedment, where 2 in of concrete covers it; it starts at the joint, where the bar is exposed for the width of the opening. That is what the coating is for, and it is why coated dowels show up on bridge decks and parking structures and almost never on an interior slab on grade.
Which gives the rule this page exists for: the adhesive and the coating solve different problems in the same hole, and neither substitutes for the other. Adhesive bonds the anchored half; the coating protects the exposed face. A coated bar in a drilled hole still needs adhesive, and adhesive in the hole does nothing for the 3/4 in of bar standing in the joint gap.
Coating is a shop process. The bar is blast cleaned, heated, sprayed electrostatically and cured before it leaves the plant, and everything that happens to it afterwards is a chance to lose the film:
Everything on this page is arithmetic: annulus, volume, cartridge count, embedment. None of it is a capacity, and the step where most dowel jobs go wrong is treating a cartridge label as one.
An adhesive anchor's capacity comes from an ICC-ES evaluation report issued for that specific adhesive, under acceptance criteria AC308, and the report tabulates capacity for specific bar sizes, hole diameters, embedments and concrete strengths. ACI 318-19 Chapter 17 governs how those tabulated values are used in design. Three things follow, and they are checkable without doing any engineering:
For a coated bar the equivalent is simpler: the coating changes bond, the bond factor is in the code, and the structural design that produced the embedment already accounts for it — or it does not, and the drawing is short. Either way the number to check is the embedment on the drawing against the table above, not the coating's corrosion performance.
Four questions, in this order, settle it:
No, it makes bond worse. ACI 318-19 gives coated bar a coating factor of 1.2 or 1.5 that multiplies development length, so the same bar needs 20% or 50% more embedment. The coating is there to keep chloride off the steel; it is not a bonding agent, and the smooth, inert film is exactly why the code penalises it.
Yes, and on a bridge deck or a parking structure it is often required — but two things change. The embedment grows by the coating factor, so a #4 goes from 16 in to 19 in or 23 in, and every cut end and field bend has to be patched before the pour. If the dowel also has to slide, the sliding half still needs a sleeve or a pocket; coating does not provide the slip.
It is the annulus between the hole wall and the bar, times the drilled depth. For a #4 bar in a 5/8 in hole drilled 16.5 in deep that is 1.82 in³; #3 is 1.07, #5 is 2.63, #6 is 3.75, #7 is 6.35 and #8 is 8.03. Add about 10% for what stays in the nozzle and on the floor.
A 10 oz cartridge is 18.05 in³. At 1.82 in³ per hole that is nine #4 dowels; #3 gives about 16, #5 about 6, #6 about 4, #7 about 2 and #8 about 2. The count falls fast with bar size because the annulus grows while the cartridge does not.
If the detail is an adhesive anchor, yes — that is the connection. The alternative is a mechanical anchor, which is a different product with its own evaluation report and its own hole tolerance. What you cannot do is substitute construction adhesive from a hardware store: it has no evaluation report for cracked concrete, no tabulated embedment and no design basis.
No. Anchoring adhesive is a structural product with an ICC-ES report under AC308 and tabulated capacities for specific bar sizes, hole diameters and embedments. Construction adhesive is a gap filler for trim and subfloor. They look similar in a cartridge and share nothing else — the failure mode of the wrong one is a dowel that pulls out quietly.
The bond is between the adhesive and the hole wall, so dust is a layer in the load path. The sequence that works is brush, blow, brush, blow — two full cycles — and then look into the hole. A hole that is 16.5 in deep holding an inch of dust is a 15.5 in hole, which is 6% less bonded area and nothing about it looks wrong from the surface.
Only where the specification allows it, and the coating on the outside of the bend will crack. Shop bending before coating is the normal route; field bends are a patching job, and the patch material has to be the one supplied for that coating system. A bend that was not patched is bare steel at the point of highest strain.