PVC Resin K Value: What 66, 67 and 68 Change on a Line
The K value comes from a dilute-solution viscosity test, not the powder. What each unit shifts in an extruder, and why the band is wider than the contract.
A K value is not a reading taken from a bag of powder. It is calculated from a measurement made in solution, and the arithmetic between the two is where most of the misunderstanding sits. Two lots can both be described as SG-5 and still behave differently on the same screw, because the band that makes both compliant is four units wide and the line was validated on one of them.
1. The number is derived, not measured
The laboratory dissolves a weighed quantity of resin in cyclohexanone, holds the solution at 25 °C, and times its flow through a capillary viscometer against the pure solvent. What comes out of that step is a viscosity number, expressed in millilitres per gram. The K value is the Fikentscher constant calculated from the relative viscosity of the same solution.
Everything that follows from this matters commercially. Because the figure is calculated from a solution measurement, it moves with solvent quality, bath temperature, concentration and the capillary in use. Two laboratories working to the same method can differ by a few tenths of a unit on the same sample, and that sits inside normal method variation, short of an error.
The method itself changed recently. GB/T 1632.2-2025, the second part of the plastics viscosity standard covering PVC resins, replaced GB/T 3401-2007, which was withdrawn on 1 May 2026. A certificate that names the older edition is not automatically wrong: stock produced and tested before that date carries it legitimately. A lot produced in the second half of 2026 that still quotes the withdrawn edition deserves a question about which laboratory issued the document and when.
2. What the national standard fixes
GB/T 5761-2018 assigns each suspension grade a K band, and pairs the band with the viscosity number and the average degree of polymerisation that go with it. The three columns are the same statement written three ways: K is the conventional figure, the viscosity number is what the bench actually measured, and the degree of polymerisation is what the number means in chain length.
| Grade | K value | Viscosity number (mL/g) | Average degree of polymerisation | Plasticiser absorption (g/100 g) |
|---|---|---|---|---|
| SG-3 | 71 – 72 | 127 – 135 | 1251 – 1370 | ≥ 26 |
| SG-5 | 66 – 68 | 107 – 118 | 981 – 1135 | ≥ 19 |
| SG-7 | 60 – 62 | 87 – 95 | 741 – 845 | ≥ 12 |
| SG-8 | 55 – 59 | 73 – 86 | 650 – 740 | ≥ 12 |
The last column belongs in the same table because it moves with the same variable. A grade that carries longer chains takes more plasticiser before the dry blend goes sticky. The gap between SG-3 at 26 and SG-8 at 12 is the difference between a cable compound and a rigid fitting, and it is not a formulating preference. It is what the molecular weight allows.
3. One K unit on a pipe line
On an extruder, a higher K means a higher melt viscosity at the same temperature. Torque rises, head pressure rises, and output falls unless the temperature profile is adjusted to compensate. What is bought with that cost is melt strength: the ability of the parison, the cone or the thick section to hold its shape while it is still hot.
The practical range for pressure pipe sits at K 66 to 68. Inside that window, the product decides, not preference. Large-diameter pipe and thin-wall sections are usually run at the top of the band, where the extra melt strength carries the wall through calibration. Small-diameter pipe on a high-output line is often run at the bottom, where the lower viscosity buys throughput and the wall thickness still passes.
The same arithmetic explains a familiar complaint. A processor validated on 67 who receives a lot at 67.9 does not see a different grade on the certificate. He sees higher torque, a hotter die, a slower line, and occasionally an unmelted particle in a thick section. Nothing is out of specification. The process window was narrower than the standard.
4. Why the band is wide and the contract is narrow
A national band is an acceptance floor. It answers one question: is this material the grade it claims to be. It does not answer whether the lot will run on a specific line, because that depends on the screw, the die, the stabiliser package and the cooling, none of which the standard knows about.
Where a line has been validated on a narrow window, that window belongs in the purchase contract, not in the specification discussion. An order written as K 66 to 68 accepts a 67.9 lot. An order written as K 67 ± 0.5 does not, and the certificate then has to show the measured figure against that clause. The distinction costs nothing to write and settles the argument before the container is loaded.
5. Two failure directions
| Direction | What appears on the line | Usual cause | What settles it |
|---|---|---|---|
| K too high | Torque and head pressure climb, output drops, unmelted particles in thick sections, surface roughening | Lot at the top of the band, or a band wider than the validated window | Temperature profile adjustment, then a narrower contract window |
| K too low | Dimensions pass, impact and hydrostatic strength fall short, drop test failures | Lot at the bottom of the band, or a grade selected one step low | Grade step up, not a process change |
The second direction is the expensive one, because the pipe leaves the line looking correct. Wall thickness passes, the surface is clean, and the failure appears in a burst or drop test three weeks later, by which time the lot has been consumed.
6. What a buyer can check without a laboratory
Three documents carry the evidence, and none of them requires equipment. The certificate of analysis for the lot should show a measured K value and the method edition it was tested to. The bag should carry the grade designation that matches the certificate. And the last three certificates from the same supplier, read side by side, show the spread that the line will actually see: a supplier who cannot produce three consecutive certificates describes a capability, and not a record.
A single certificate number describes one solution measured at 25 °C. Where the certificate and the line disagree, the line is the authority, and the variable to change is the position inside the band. The full band structure across the four grades sits on the specification table, and the grade that matches a given end product is set out on the SG-5 page and in the grade guide.
SG-5 for the next production run
The grade, the tonnage and the destination port produce a price, a specification sheet and the full document list, usually within 12 hours (GMT+8).