Verifying a Coolant Shipment on Arrival: What Refractometer, Hydrometer, pH and Reserve-Alkalinity Tests Actually Prove

Last verified: 2026-07-31

Suppose two drums arrive from the same supplier, against the same order, with the same certificate stapled to the delivery note. One tests clean on every instrument the receiving team owns. The other throws a freeze-point reading that looks exactly like a batch someone cut with water on the way — except it isn't diluted at all. It is a correctly made product read on the wrong glycol scale. Nothing on the instrument tells the technician which situation he is in, because a wrong-scale reading and a genuinely off-specification one look identical on the display. That is the real risk in coolant arrival testing: not that the four standard checks are unreliable, but that each one answers a narrower question than the person holding the instrument assumes it does.

What does a refractometer reading on arrival actually confirm?

A handheld refractometer measures how strongly the sample bends light — its refractive index — and converts that number to an approximate freeze point and glycol percentage using a scale built into the instrument. That conversion is governed by ASTM D3321-26, published 9 June 2026, which replaced D3321-19(2023). The revision was not cosmetic: the method was retitled around handheld refractometers, and percent glycol was added as a determined output, where the previous edition's title named the freezing point alone.

The scale is the load-bearing part, not the light. D3321-26's own Note 1 warns that the method "does not cover scales for other fluids that may be present on certain refractometers," and the standard states directly: "Care must be taken to use the correct glycol freezing point scale for the glycol type being measured. Use of the wrong glycol scale can result in freezing point and % glycol levels being incorrect."

On the certificate, this reads as a D3321 result at a stated glycol scale. Without that scale named, the number on its own proves nothing.

Why can a hydrometer disagree with the refractometer on the same drum?

A hydrometer measures relative density — buoyancy — not light. ASTM D1122-22 governs it, and its scope carries a qualifier that is easy to skip past: relative density "may be used to determine the approximate percent glycol, freezing point, and boiling point, provided the glycol type is known." Provided is doing the work in that sentence.

There is a second condition the field reading rarely satisfies on its own: temperature. D1122's Significance section names 25°C, 20°C and 15.6°C as the reference temperatures commonly used, and a reading taken at ambient dock temperature has to be corrected back to one of those before it means anything comparable to a certificate figure. A hydrometer is not self-compensating.

So refractometer and hydrometer can legitimately disagree on the identical sample, for a reason that has nothing to do with either instrument being faulty. Both are correlations, not chemical identifications. Both were built against a reference formulation — glycol plus a typical inhibitor package. A real coolant's organic acids, salts, silicates and borates contribute to refractive index and to density in different proportions than glycol alone does; a shipment with a heavier or lighter inhibitor load than the reference formulation behind the instrument's scale can read differently on the two instruments at a true glycol concentration that is, in fact, identical. Neither standard claims to correct for that.

Correct the hydrometer reading to its reference temperature, and record which reference temperature you used, before comparing it to anything else.

Do pH and reserve alkalinity prove the coolant is still protecting the system?

pH and reserve alkalinity sound like health checks. They are acidity and titration measurements, and their own governing documents say, in writing, that they are not what most buyers reach for them to prove.

ASTM D1287-25, published 5 November 2025, covers pH determination only — its scope is unused engine coolants, insulation and heat transfer fluids, and antirusts, together with used or unused aqueous dilutions of the concentrated products. Nothing about density, and nothing about the inhibitor package's remaining capacity. The standard states plainly that pH "is not significant from the standpoint of predicting service life" and is "not a dependable indication of either existing effectiveness or remaining life."

Reserve alkalinity, governed by ASTM D1121-25 (published 7 November 2025, and retitled to the plural "Test Methods" to reflect a newly added titration procedure alongside the manual one), goes further than a caveat. The standard states that reserve alkalinity "is not a dependable measure of its ability to prevent corrosion, nor can it satisfactorily indicate the additional life of the solution" — and then names the misreading directly, noting the term "is sometimes misused in that its numerical value is said to be directly related to coolant quality, the higher the number, the better the coolant." ASTM Committee D15 on Engine Coolants, the standard adds, "believes there is a need to correct some misconceptions and place the term in its proper perspective."

The two tests a buyer reaches for to prove a coolant is still good are the two whose own standards say they cannot prove it — and on one of them, the committee that wrote the method says so in order to correct a misconception it knows is circulating.

What they do confirm — a snapshot of buffering chemistry at the moment of the titration — is real and worth recording. It is a different fact from "this shipment's corrosion inhibitors are intact," and the certificate should be read that way. Request both figures, but request them as chemistry snapshots, not as a verdict on inhibitor life.

Can any of these tests catch a glycol mix-up?

No, and this is the boundary that matters most for a cross-border consignment. D3321-26 covers the approximate freeze point and % glycol "provided by ethylene and propylene glycol-based coolants" — both bases, each read on its own scale, which is precisely why the standard insists the scale match "the glycol type being measured." Confirming which base actually arrived is a different question, and nothing in the refractometer or hydrometer standards, and nothing in the pH or reserve-alkalinity standards, is sourced as capable of confirming a stated EG/PG ratio or catching a blended shipment mislabelled as single-base.

That gap sits directly on top of an export mislabelling risk. A shipment declared as ethylene glycol that arrives partially propylene glycol, or the reverse, will not announce itself as an anomaly. It will simply produce a number that is off from the expected reading, on an instrument that reports numbers confidently regardless of which glycol scale it is set to. A confident wrong number and a confident right number look the same on the display.

Specify the declared glycol base on the purchase order, and check the instrument's scale setting against it before a single reading is recorded.

Two supplier-side laboratory tests are sometimes cited in the same conversation and do not belong there. ASTM D1384-24 (glassware corrosion) and ASTM D2570-26 (simulated service corrosion, published 21 April 2026) are formulation-qualification tests run during product development, not fast dockside acceptance checks — and D2570 disclaims its own predictive reach directly, stating it "cannot conclusively predict satisfactory corrosion inhibition and service life." Both belong to the supplier's own qualification file, which the certificate of analysis should be able to reference.

What happens when the four readings do not agree with each other?

Disagreement is not automatically a defect. It can mean one of three different things, and only one of them is the buyer's problem to reject over.

Reading a coolant arrival disagreement. Applies to a bulk consignment tested per ASTM D3321 (refractometer), D1122 (hydrometer), D1287 (pH) and D1121 (reserve alkalinity), against the product specification and glycol base named on the purchase order — never against a figure carried over from a different product or a previous shipment.

ASTM D4985-10(2023) shows why the specification has to be named rather than assumed. Its concentrate, at 40 % to 60 % concentration by volume in water, is formulated to protect against freezing to at least −36.4 °C (−33.5 °F) and boiling to at least 108 °C (226 °F). Those figures belong to D4985 — a low-silicate ethylene glycol heavy-duty product requiring a supplemental coolant additive pre-charge. They do not transfer to a D3306 light-duty or a D6210 fully-formulated heavy-duty shipment, and reading a figure across specifications is exactly the kind of error a conflicting-readings dispute produces.

Trace every disagreement back to the declared specification and glycol base before treating it as a quality failure.

Which six items need to be fixed in writing before the shipment sails?

None of the four dockside tests carries a dispute-resolution mechanism of its own. ASTM does publish one separately — D3244-20, Standard Practice for Utilization of Test Data to Determine Conformance with Specifications, which lets a supplier and a receiver combine independently obtained results into an Assigned Test Value and set an Acceptance Limit for accepting or rejecting the product. But it only governs a dispute the two sides have already agreed to settle that way, so naming it is still a commercial act. Six things belong in the purchase agreement, alongside the wider commercial terms covered in the coolant sourcing guide, before the shipment leaves the supplier's dock:

1. The governing product specification — D3306, D6210 or D4985, named once and used consistently, because their published requirements belong to one document each and are not interchangeable.
2. The exact glycol base — ethylene or propylene — since neither refractometer nor hydrometer scale can be chosen correctly without it.
3. The test-method editions both sides will use, since D3321, D1287 and D1121 were all revised in 2025–2026 and a buyer testing to one edition while a supplier certifies to another are not comparing like with like.
4. The sampling practice — ASTM D1176 or an agreed equivalent — because a result from a sample not drawn to an agreed practice has no defensible basis if the numbers are later disputed.
5. Who tests and who breaks a tie — an agreed independent laboratory and a referee procedure, naming ASTM D3244 if that is the basis both sides accept, since none of the four test methods resolves a disagreement on its own.
6. What the supplier's own batch certificate already certifies, so the arrival numbers are compared against a stated baseline rather than a figure the receiving team assumes.

Specify all six at the point of order, and confirm how the supplier documents them — sampling practice, method editions, and the specification the batch was made against — when reading the certificate of analysis that accompanies the shipment. Buyers sourcing through the Altonex Global platform can write these six items into the RFQ before a supplier is engaged, in the coolant and antifreeze category, rather than discovering the gap once drums are already on the dock.

Frequently asked questions