A V-belt is the cheapest component on most drives and the one that stops the machine. That is why V-belt section selection wrapped versus raw edge cogged comes up so often in a sourcing enquiry, and why the answer a supplier gives you tells you a lot about whether they make belts or resell them. The section is geometry; it decides which pulley groove the belt fits. The construction — wrapped or raw-edge cogged — is what happens inside that geometry, and it decides how the belt behaves on small pulleys, on overload, and when the drive is not perfectly aligned. Get the section wrong and the belt does not fit. Get the construction wrong and the belt fits, runs, and wears out in a quarter of the time.
This article works through the section families and how to read them, what wrapped and raw-edge cogged constructions actually change, the drive data that lets a transmission belting maker quote a belt rather than a price, and what to ask before an order.
Sections: the geometry you have to get right
V-belt sections are defined by a nominal top width and a nominal height, and by the groove angle of the pulley they run in. The families you will meet in a quote:
| Family | Sections | Nominal top width (mm, approx.) | Where you will meet it |
|---|---|---|---|
| Classical (ISO 4184 / ARPM IP-20) | Z (or 10), A, B, C, D, E | 10 / 13 / 17 / 22 / 32 / 38 | Older machinery, fans, pumps, agricultural drives, general OEM equipment |
| Narrow / wedge, metric (ISO 4184 / DIN 7753) | SPZ, SPA, SPB, SPC | 10 / 13 / 17 / 22 | European-designed machinery, compressors, higher-power drives in less space |
| Narrow / wedge, inch (ARPM IP-22) | 3V, 5V, 8V | 3/8 in / 5/8 in / 1 in | US-designed machinery, HVAC, industrial fans |
| Double-V (hexagonal) | AA, BB, CC | as classical | Serpentine drives where the back of the belt also drives |
Treat the widths as approximate and confirm against the current edition of whichever standard your drawing cites. Three details inside that table matter more than the table itself:
- Narrow sections are not just thinner classical sections. SPA and A share a nominal top width of about 13 mm, but SPA is deeper and carries the tension cord higher in the section, which is why a narrow belt transmits more power for its width. They do not run in the same groove: an SPA belt in an A groove rides wrong and wears its flanks off. If a supplier tells you they are interchangeable, stop the conversation.
- The section letter does not fix the length. Length is designated separately — classical belts usually by inside length or by a nominal number, narrow belts commonly by datum or effective length. The same physical belt can carry two different length numbers depending on the system. Ask which the supplier's number refers to, and ask them to state it on the belt marking.
- Groove angle changes with pulley diameter. Pulley groove angles for the same section step through a range (commonly 34° to 38°) as pulley diameter increases, because the belt's sides deflect inward as it bends around a small pulley. The belt is designed to match that. A belt from a maker who does not know this is built to one angle and will fit some of your pulleys well and others badly.
Wrapped vs raw-edge cogged: what changes inside the belt
Every V-belt has the same anatomy: a tension member (cord, usually polyester, sometimes aramid) carrying the load, a compression section of rubber under the cord that takes the bending, and something on the outside. The something is where the two constructions diverge.
Wrapped belts have the whole section enclosed in a rubberised fabric jacket. The jacket gives a smooth, fabric-faced flank that runs on the pulley groove, protects the rubber core, and — the point most buyers miss — allows the belt to slip under sudden overload without tearing. That slip is a feature on drives that see shock loads or jams: the belt squeals and the motor keeps turning instead of stalling or shearing a key. The cost is stiffness. The jacket resists bending, so a wrapped belt runs hotter on small pulleys and has a larger minimum pulley diameter for the same section.
Raw-edge cogged belts have no jacket on the flanks. The rubber compound itself runs on the groove, which means the compound has to be the wear surface — usually a harder, fibre-loaded compound rather than the softer core of a wrapped belt. The underside is moulded with transverse notches (the cogs) that let the compression section bend without the rubber buckling. The result is a belt that bends round smaller pulleys with less internal heating, grips the groove more firmly because the raw rubber flank has a higher coefficient of friction than fabric, and transmits somewhat more power in the same section.
Where each construction is the right answer:
| Drive condition | Wrapped | Raw-edge cogged |
|---|---|---|
| Small pulley diameters, tight centre distance | Runs hot, shorter life | Designed for it |
| Shock loads, jams, drives that must slip to protect the machine | Slips on the jacket, survives | Grips; may transfer the shock to the machine |
| Pulleys with worn or rough grooves | Jacket tolerates it | Raw flank wears fast on a rough groove |
| Misaligned or poorly tensioned drive | More forgiving | Less forgiving; edge wear shows quickly |
| Continuous high-load, well-maintained drive | Adequate | Higher efficiency, cooler running |
| Dusty or wet environments | Jacket sheds contamination | Notches can pack with dust; check the application |
The detail that separates people who run drives from people who sell belts: the cog is not decorative and its pitch matters. Cog pitch and depth are chosen by the maker against the section and the compound; a cogged belt with notches too deep loses compression rubber and cracks at the notch root, one with notches too shallow does not bend as intended and gains nothing over a plain belt. There is no standard that fixes the cog geometry — it is the maker's design. Ask for it and ask how it was validated.
Matched sets and the length question
Multi-belt drives run two or more belts side by side in a grooved pulley, and they only share the load if the belts are the same length. A belt that is a few millimetres longer than its neighbour carries less tension, slips first, heats first and fails first, and then the remaining belts pick up its share and fail after it.
Matched sets were historically sold by measured length code, so a buyer could pick belts from the same code. Many makers now build to a tighter length tolerance and describe belts as matchable from stock, or supply a set measured together. What you need to know from a supplier is which of those they do:
- A stated length tolerance on a measuring fixture at a specified tension, so you can judge whether belts from the same production lot will match without individual selection. Ask for the tolerance and for the fixture tension it is measured at.
- Whether they will supply sets measured together and marked as a set. On a drive with four or more belts this is worth the small premium.
- Whether a banded (joined) belt is available. A banded belt is several V-sections joined by a common top band, which removes the matching problem entirely and stops belts flipping in the groove on long centre distances or pulsating loads. Not every plant makes them. It is on the list to ask.
None of those is a given. A transmission belting maker who quotes section, length and drive rating to the buyer's drive data — as the Hebei belting works Jiutong Conveyor Belting does, by its own account, asking for pulley diameters, centre distance and duty cycle with the enquiry — is at least quoting the right way round: from the drive to the belt. What that works produces in sections, lengths and banded constructions is not published and should be asked for directly.
The drive data that turns an enquiry into a quotation
A belt maker cannot quote a V-belt from "an A-section belt, 50 inch". They can quote a part number that fits that description, but they cannot tell you whether it will carry your drive. Quoting to the drive means the supplier runs the same arithmetic your machine designer ran: power, speed ratio, pulley diameters, centre distance, service factor, and from those the belt speed, the arc of contact on the small pulley, and the number of belts needed. Send the data and you get a construction and a belt count. Send a part number and you get a price for a belt that may or may not be the right one. This is the same distinction that explains why Chinese quotes for the same spec come back different: the suppliers were not quoting the same thing.
The lines that belong in the enquiry:
- Driver: motor rated power, rated speed, type (electric motor, IC engine, which has a much harsher torque profile).
- Driven machine: what it is, its speed, and its load character — smooth (centrifugal pump, fan), pulsating (reciprocating compressor), shock (crusher, mill). This sets the service factor.
- Pulley pitch (datum) diameters, both pulleys, and the groove profile they are cut to (classical A, narrow SPA, 3V, and so on — read it off the pulley or the drawing, not from the old belt).
- Centre distance, fixed or adjustable, and the take-up range if adjustable.
- Duty cycle: hours per day, starts per hour, ambient temperature, and whether the drive runs through oil mist, dust, moisture or UV.
- Existing belt marking, if replacing — section, length and construction as printed on the old belt, plus how many belts run side by side.
- Constraints: a static-dissipative belt requirement (ISO 1813 is the test for electrical conductivity), a heat-resistant or oil-resistant compound, or a maximum belt count the pulley can take.
The supplier's reply should contain a belt count and a stated design life basis, not just a section and length. If the reply is a single line with a price, the enquiry was read as a part-number order, and you will be paying for the drive calculation later — in belts.
Common questions
Does raw-edge cogged buy anything on a drive that is running fine on wrapped belts?
Usually cooler running and slightly higher efficiency, and on a small-pulley drive, longer life. On a large-pulley, lightly loaded drive with worn grooves, it can buy you faster flank wear and nothing else. If the wrapped belt is lasting its expected life, the case for changing is efficiency and heat, and you should measure the drive temperature before and after rather than take a catalogue's word for it.
Can I swap a classical belt for a narrow belt of the same top width?
Not without changing the pulleys. The groove profile differs. A narrow belt in a classical groove bottoms out or rides high, and either way the flanks do not seat. If the machine is being redesigned, narrow sections let you use smaller pulleys and fewer belts; if you are maintaining an existing drive, buy the section the pulleys were cut for.
Why do belts from two suppliers with the same marking run differently?
Because the marking fixes the section and the nominal length, not the cord material and count, the compound hardness, the cog geometry or the length tolerance. Two "SPB 2500" belts can differ in power rating, stretch and heat resistance. Ask for the cord type, the compound and the rated power per belt at your drive conditions, and compare those.
How much stretch should I expect after installation?
New V-belts seat into the grooves and their cords settle in the first hours of running, so tension drops and has to be reset — the belt maker's installation data usually specifies a re-tension after a run-in period. That drop is a property of cord material and construction, and it is a question for the supplier: what run-in re-tension do they recommend, and what is the expected tension loss after it. A maker who cannot answer has not measured it.
What to ask the supplier next
Write these into the enquiry alongside the drive data above, in the same document format you would use for any spec sheet sent to a Chinese factory:
- Which sections do you produce — classical Z/A/B/C/D, narrow SPZ/SPA/SPB/SPC, inch 3V/5V/8V — and what is the length range per section?
- Is the belt wrapped or raw-edge cogged, and for cogged, what is the cog pitch and depth and how was it validated?
- What is the cord material, and what is the rated power per belt at the pulley diameters and speed given?
- What length tolerance do you hold, at what measuring tension, and will you supply matched sets marked as sets? Are banded belts available?
- What is the recommended installation tension and run-in re-tension, and is a static-dissipative or oil-resistant compound offered?
- Can one belt be ordered as a trial before a drive-set order, and at what lead time?
Jiutong Conveyor Belting lists transmission belting second on its application after fabric conveyor belting, quotes section, length and drive rating to the buyer's drive data, and states a one-piece MOQ with standard samples in five working days — company-stated terms, to be confirmed at quote time. Which sections and constructions come out of the plant is not published on its factory profile, so that is the first question to send.
