Penetration grades — a needle at one temperature
A standard needle loaded with 100 g is allowed to sink into a conditioned sample for 5 seconds at 25 °C, and the depth reached is recorded in tenths of a millimetre. Land between 60 and 70 dmm and the material is a 60/70. The method is ASTM D5, and identically EN 1426 and IS 1203. A lower number means a harder binder, which is the first thing that catches people out.
This is the commercial language of the Middle East, South Asia and most of Africa, and it is the family with the most grade pages here: 20/30, 30/40, 30/45, 35/50, 40/50, 50/70, 60/70, 70/100, 80/100, 85/100, 120/150 and 200/300. The American and European standards draw the bands differently — ASTM D946 names 40/50, 60/70, 85/100, 120/150 and 200/300, while the grades named in EN 12591 include 20/30, 30/45, 35/50, 40/60, 50/70, 70/100, 100/150, 160/220 and 250/330. That is why 50/70, 35/50 and 70/100 sit in the same index as 60/70 and 85/100: they are not duplicates, they are the counterpart grades from the other standard. A specification written against EN 12591 does not automatically accept a grade named under ASTM D946, and an engineer is entitled to say so.
The philosophy is worth stating plainly, because it is the source of everything the system cannot do. Penetration grading takes one measurement, at one temperature, on unaged binder. It is fast, cheap, reproducible and understood everywhere. It says nothing whatever about how the binder flows at 60 °C in a summer pavement, nothing about how it cracks at −20 °C in a winter one, and nothing about how much of its consistency it will lose during mixing and service. The standards close part of that gap with separate lines for softening point, ductility, solubility, flash point and retained penetration after the thin film oven test, but the grade name carries only the needle.
Viscosity grades — flow at service temperature
Viscosity grading replaces the bench-temperature needle reading with absolute viscosity at 60 °C, measured in poise by a vacuum capillary viscometer under IS 1206 (Part 2) or ASTM D2171. The reason is straightforward: 60 °C is roughly what the top of a bituminous layer reaches on a hot afternoon, so the measurement is taken where the binder actually has to work. The grade number is the nominal viscosity in hundreds of poise, and the acceptance band runs about ±20 % around it — 800 to 1200 poise for VG-10, 2400 to 3600 poise for VG-30.
Four grades are published: VG-10, VG-20, VG-30 and VG-40. VG-10 is the spray and seal grade and the usual feedstock for emulsion and cutback plants; VG-30 is the dense-graded highway grade for hot climates; VG-40 is for slow, standing or channelised loading. India moved from penetration to viscosity grading with IS 73:2013, and the standard also specifies ageing on RTFOT residue rather than TFOT — a detail worth knowing because a certificate quoting a TFOT result against an IS 73 specification has not tested what the standard asks for.
Two features distinguish the philosophy from penetration grading. The first is that the measurement is taken at a service temperature rather than a bench temperature, which makes the grade directly relevant to rutting. The second is that IS 73:2013 brings ageing inside the grade by capping the viscosity ratio at 60 °C after RTFOT at 4.0 — the ratio of aged viscosity to original viscosity. That single line is the only place in either of the first two systems where the grade itself constrains how the binder ages, and it is the bridge towards performance grading. Note also what the narrow bands imply for storage: a VG-30 parcel that ships at 3200 poise has 400 poise of headroom before it leaves its own grade, and hot storage spends that headroom.
Performance grades — rheology across the service range, on aged material
A grade written PG 64-22 is the only classification whose name is a pair of temperatures the pavement will actually reach: it was proved at a 64 °C high pavement design temperature and at −22 °C low. Both ends move in 6 °C steps, which is why the published grades run 52, 58, 64, 70, 76 and 82 at the high end and −10, −16, −22, −28 and lower at the low end. Ten grades are published here, from PG 52-10 through PG 58-22, PG 64-22 and PG 70-10 to PG 76-16.
Getting to a PG designation requires a test sequence rather than a single reading, and the sequence is the philosophy. Under AASHTO M320 or ASTM D6373 the binder is tested in three states. Original binder gives rotational viscosity by ASTM D4402, capped at 3.0 Pa·s at 135 °C so that the binder can still be pumped and mixed, and a dynamic shear rheometer result under ASTM D7175 of at least 1.00 kPa for the complex modulus divided by the sine of the phase angle. RTFO-aged residue, produced under ASTM D2872 at 163 °C for 85 minutes, must reach at least 2.20 kPa on the same rutting parameter — the short-term ageing of mixing and laying is simulated before the binder is judged. PAV-aged residue under ASTM D6521, which subjects the RTFO residue to 20 hours of pressurised air, is then tested by DSR for fatigue and by bending beam rheometer under ASTM D6648, where creep stiffness must not exceed 300 MPa and the m-value must be at least 0.300.
The consequence is that a PG grade contains information about ageing behaviour that no penetration or viscosity number carries, because most of the criteria are applied to material that has already been aged in the laboratory. That is exactly why a PG designation cannot be inferred from a penetration certificate: the tests that define it were never run.
One further layer sits on top. AASHTO M332 and the multiple stress creep recovery test under AASHTO T350 replaced the older habit of bumping the grade upward for heavy or slow traffic. MSCR is run on RTFO residue at the high grade temperature and reports the non-recoverable creep compliance Jnr at 3.2 kPa, with traffic designations set at 4.5 kPa−1 or less for S (standard), 2.0 or less for H (heavy), 1.0 or less for V (very heavy) and 0.5 or less for E (extremely heavy), together with a cap on the Jnr difference between stress levels. If your project document names an M332 traffic designation, an M320 certificate does not answer it, and no amount of correspondence will make it do so. The full test-by-test treatment is on the performance grading guide.
The Chinese climate-zone class — a selection rule, not a measurement
A fourth system is in daily use and it is not a variant of the other three. In China a grade is not chosen from a catalogue at all. The site is first classified with a three-part climate-zone code under JTG F40, in which the first digit is the summer division taken from the hottest month's mean maximum air temperature, the second is the winter division taken from the extreme minimum air temperature, and the third is the rainfall division. A code such as 1-4-1 or 2-2-2 is what a Chinese engineer works from, and only after the code is fixed does a grade designation follow.
The designations themselves are penetration bands with a quality class attached. GB/T 15180 names the AH series — AH-30, AH-50, AH-70, AH-90, AH-110 and AH-130 — where the number is the nominal penetration at the centre of the band, so AH-70 covers 60 to 80 dmm at 25 °C. JTG F40 writes the same bands as 30#, 50#, 70#, 90#, 110# and 130#, then divides each into A, B and C classes carrying progressively lighter requirements, with the class rather than the number deciding what traffic the binder may carry.
Three consequences follow for anyone quoting into that market. The bands are not the ASTM or EN bands: AH-70 is 60 to 80 dmm, not 60 to 70, so a compliant 60/70 sits comfortably inside the Chinese band while a compliant Chinese 70# may sit outside the ASTM one. The class letter is an entirely separate acceptance dimension with no counterpart in ASTM D946 at all. And because selection runs from the climate code, a Chinese enquiry that names only a number has not yet told you which class or which zone it is for — that is a question to ask before pricing, not after. The three pages that carry this in full are the climate zones and their boundaries, JTG F40 and GB/T 15180.
Why a buyer cannot convert between them by arithmetic
People try, constantly, and the reason it fails is not that the conversion factors are hard to find. It is that the four systems measure different quantities, at different temperatures, on differently aged material, and two of them are not measurements at all in the same sense.
The physical obstacle is temperature susceptibility. Two binders can have identical penetration at 25 °C and different softening points, because the rate at which consistency changes with temperature is set by crude source and processing route rather than by the grade name. Different susceptibility means different viscosity at 60 °C and different stiffness at −20 °C from the same needle reading. The number that expresses this is the penetration index computed from penetration and softening point together: straight-run paving binders typically land between about −1 and +1, and blown grades far higher. A conversion from penetration to viscosity is therefore only ever valid within a family of binders that share a susceptibility, and it carries scatter of the order of a grade band.
Layered on top of that is the ageing problem. Penetration grading measures unaged binder. The critical PG criteria are applied to RTFO and PAV residue. There is no arithmetic that manufactures an ageing result from a measurement that was taken before ageing, which is why a penetration certificate can never evidence a PG grade no matter how many correlations are consulted. And the Chinese system adds a dimension that is not a number at all: the A, B and C class is a bundle of separate requirements, so even a perfect band match leaves the class unanswered.
The practical rule is short. An equivalence table is a shortlisting tool; a test report is a compliance statement. Use a cross-reference to work out which grade to ask a supplier for, then require a certificate reporting the tests the destination standard actually names. Where you must offer a grade from a different system than the one specified, get the engineer's written approval before the cargo is produced, not after it arrives. The mappings, including the ones that look right and are not, are set out in full on the grade equivalence page; nothing on this hub reproduces them.
What none of the four grade names tells you
Four things matter commercially and are absent from every grade designation in every system. Adhesion belongs to the binder and the aggregate together, not to the binder alone, so no certificate can predict it and no grade name implies it — see adhesion and anti-stripping. Wax content is separately limited in the Chinese market and is invisible in a penetration band; see wax content. Flash point varies with grade and with standard, and counter-intuitively the softest grades carry the lowest floors — ASTM D946 requires a minimum of 232 °C for 60/70 but only 177 °C for 200/300. And whether the binder will still be in grade when it arrives depends on how it was held hot, which is a logistics question rather than a specification one.