Can You Mix Two Hydraulic Oil Brands If They Share the Same ISO VG Grade?

Last verified: 2026-07-21

Sometimes, yes. But the number that decides most buyers' minds — the ISO VG grade — is the weakest link in the chain of things that actually have to match before two hydraulic oils can safely share a system, and it is almost always the only one anyone checks before topping up a reservoir. A shared grade tells you the two fluids sit in the same viscosity band at one temperature. It tells you nothing about the additive chemistry, the base oil, the seals your machine depends on, or what happens when the two liquids meet inside a fine filter. Matching grade lowers risk. It does not remove it, and treating it as proof of safety is the mistake this article exists to stop.

What does an ISO VG number actually promise?

ISO 3448, Industrial liquid lubricants — ISO viscosity classification, sets out 20 grades running from ISO VG 2 to ISO VG 3200. Each grade is defined by exactly one thing: the mid-point kinematic viscosity at 40 °C, held within a ±10% tolerance band. ISO VG 46 means 41.4–50.6 mm²/s at 40 °C. Nothing more.

The standard is silent on base oil type, API Group, additive package, anti-wear or EP performance, oxidation stability, demulsibility, foam behaviour, filterability, and viscosity index. It says nothing about viscosity at 100 °C either. It is a single-temperature bucket, not a formulation spec.

Here is the surprise most buyers never encounter until it costs them: a Group I solvent-refined oil and a severely hydrocracked Group III oil can both, legitimately, carry the ISO VG 46 stamp. Group is set by saturates content, sulfur content and viscosity index — API 1509 Annex E — not by target viscosity. Two oils can hit the identical number on a datasheet while being built from chemically distant stock.

An ISO VG grade is a viscosity measurement, not a compatibility statement.

Same grade, different oil — how far apart can two products really be?

Even at one shared VG, two hydraulic oils can differ in saturates, aromatics and sulfur content; in oxidative and thermal stability; in aniline point, which governs how the oil behaves against seal rubber; in how well the base stock dissolves the additive package it carries; and in whether the formulator needed a viscosity-index improver polymer to hit the target at all, or reached it with a naturally high-VI base stock.

None of that shows up on a grade label. All of it can show up in your filter housing.

Where those differences come from — and how the ISO 6743-4 family codes map onto real products — is covered in the guide to hydraulic fluid types, which is worth reading first if the HM, HV and HLP codes are not yet second nature.

Miscible or compatible: are these the same question?

No, and the gap between the two words is where most mixing problems live.

Miscible is a physical fact: the two liquids form one homogeneous phase, with no separation and no visible second phase. It is necessary. It is nowhere near sufficient.

Compatible is a performance judgment: the blend still meets the requirements of the application — function, performance, service life — not merely staying visually clear. Industry technical literature states the relationship in two rules worth writing on a whiteboard: what is not miscible is incompatible, but what is miscible is not automatically compatible. A blend can look flawless in a jar and still be quietly wrong for the pump.

Which standard was actually built to answer this?

ASTM D7752, Standard Practice for Evaluating Compatibility of Mixtures of Hydraulic Fluids, exists for exactly the question in this article's title. Its scope covers mixtures of hydraulic fluids as defined by ASTM D6158, DIN 51524, ISO 11158 and ISO 15380 — the mainstream mineral-oil and environmentally-acceptable-fluid families. It can be used to evaluate two new, unused fluids, or the effect of adding a replacement fluid to fluid already in service.

The practice prescribes three mixture ratios: 2:98, 10:90 and 50:50. It invokes the ISO 13357-1 Stage II filterability test — the procedure built for oils in the presence of water — because hydraulic systems live or die by fine filtration, and incompatibility characteristically announces itself as premature filter plugging long before anything else goes visibly wrong.

Read the limit carefully, because it is where most confidence goes too far: the practice does not evaluate anti-wear performance, load-carrying or EP capacity, or mechanical shear stability of the mixture. A blend can pass D7752's filterability check and still have unverified anti-wear performance. That needs separate testing.

Confirm the current edition of both D7752 and ISO 13357-1 before you cite either in a technical exchange with a supplier — do not assume the edition on an old datasheet is still current.

D7752 sits at the top of a chain, not on its own. ISO 3448 sets the viscosity grade. ISO 6743-4 classifies hydraulic fluid families by additive content — the letter codes HH, HL, HM, HV, HG — without setting performance thresholds. ISO 11158 and DIN 51524 set the performance minimums per category. ASTM D6158 and ISO 15380 name the mineral-oil and environmentally-acceptable-fluid analogues that D7752's own scope references. And a set of individual property tests — ASTM D892, D1401, D471, ISO 9120 — defines what "incompatible" actually looks like when something fails.

Where does an incompatible blend actually go wrong?

Additive antagonism is the primary documented failure mode. When acidic and alkaline additive components from two different formulations meet, they can react to form a soap — often a calcium soap — or another insoluble species: a grease-like gel or flocculent precipitate that interferes with flow and blocks fine filtration. This is the mechanism commonly given in supplier technical bulletins, and it is exactly what the D7752 filterability test is designed to catch.

A second mechanism is quieter: competitive surface adsorption. Detergents, dispersants, anti-wear agents and friction modifiers all compete for the same adsorption sites on the metal surface, and the more surface-active species wins. Detergents and dispersants are documented to adversely affect ZDDP's anti-wear and antioxidant performance through exactly this competition. A formulator balances that fight deliberately inside one package. Combine two independently formulated packages in an unplanned ratio, and that balance is gone.

Anti-foam agents add a third failure route, because they work as fine dispersions rather than true solutions — discrete droplets that destabilise foam bubble walls. Mixing two independently optimised anti-foam systems does not simply average their performance; droplet size, distribution and a carefully titrated treat rate can be disrupted. Measured by ASTM D892 at 24 °C and 93.5 °C, and worth taking seriously: foam risks inadequate lubrication, cavitation and overflow loss.

Demulsibility is a fourth. ASTM D1401 mixes 40 mL oil with 40 mL water and watches until the emulsion band shrinks to 3 mL or less — at 54 °C for oils of 28.8–90 cSt at 40 °C, 82 °C above that. A blend with degraded demulsibility holds water in the oil phase instead of shedding it, raising corrosion and hydrolysis risk.

Air release is a fifth. Both ISO 9120 and ASTM D3427 measure how long entrained air takes to fall to 0.2% by volume after air is blown through the oil at a set temperature. Two oils with different air-release chemistry are not guaranteed to keep either parent's performance once blended.

Field evidence backs all of this up. Mineral HLP 46 fluids of different additive systems — zinc-containing versus zinc-free — that are perfectly miscible with each other have still shown increased foaming, deposit formation or restricted filterability once mixed. In one documented case, an engine oil containing zinc dithiophosphates used to flush a gear system caused foaming problems in the transmission oil afterward — two mineral oils, one avoidable surprise.

Does zinc versus zinc-free settle it?

Not on its own.

ZDDP — zinc dialkyldithiophosphate — forms a glassy zinc-polyphosphate boundary film on metal surfaces and works as a secondary antioxidant. Zinc-free chemistry emerged for real reasons documented in SAE Technical Paper 982000: increasingly compact, high-pressure, high-temperature hydraulic systems were degrading ZDDP thermally into insoluble sludge. Other drivers behind ashless formulations include hydrolytic stability in wet systems, filterability performance in fine-filtration circuits, and yellow-metal attack associated with zinc-based additives in some applications.

Named formulators quoted in trade press give the honest, adjudicated answer: mixing zinc and zinc-free packages can form a gel-like precipitate that plugs filters and cuts hydraulic flow — but this is a potential consequence, not a certainty. Some zinc and zinc-free pairings mix without incident.

Here is the part worth underlining twice: two zinc-containing fluids can be incompatible with each other, and so can two zinc-free fluids, if their underlying chemistries diverge. Zinc-free is an absence, not a formula. Ashless anti-wear packages vary by supplier — amine phosphates, sulfurized esters, borate esters, and others — so "both zinc-free" is not a compatibility guarantee any more than "both ISO VG 46" is. The same coverage that raises this point explicitly declines to offer a universal rule, and recommends following OEM guidance and flushing when switching.

If you encounter a claim that modern zinc-free oils are broadly cross-compatible with zinc-containing oils already in service, treat it as a claim about one specific product, not a category-wide fact — and verify it against that supplier's own documentation.

What happens to the seals?

ASTM D471 evaluates how rubber and rubber-like compounds withstand a liquid over accelerated immersion: swelling, hardness change, tensile-strength loss, dimensional change. It is the mechanism, not a jar test, that predicts what your O-rings actually experience.

The driver is polarity, indicated by aniline point. A lower aniline point means a more polar oil, and more polar oils swell rubber more aggressively. A higher aniline point means less swell — or shrinkage. Group I stocks run more polar, with lower aniline points, than Group III or IV stocks. Switch or blend across base-oil families, and you have a documented, mechanistically explained route to a seal swell or shrink shift — separate from, and additional to, any additive antagonism happening at the same time.

Shrinkage is the direction that should worry you more. A little swell is often tolerated, sometimes even useful — it compensates for compression set and helps hold sealing-lip interference. Shrinkage does the opposite: it reduces interfacial pressure and opens a leak path that gets worse with time.

NBR (nitrile) is the default hydraulic seal elastomer for mineral oils; its compatibility with higher-VI Group III/IV or ester-containing fluids needs per-formulation checking. FKM (fluorocarbon) is broadly compatible across mineral, PAO and ester stocks and is the most chemically resilient common choice. HNBR follows NBR's pattern and also needs per-formulation verification. EPDM is fundamentally incompatible with petroleum-based hydraulic fluids — it swells rapidly and fails — and belongs on water-glycol and phosphate-ester fire-resistant fluid systems instead. Polyurethane is generally tolerant of mineral oils but should be checked against ester or PAG-family fluids case by case.

No hydraulic-specific seal-immersion standard distinct from ASTM D471 is confirmed to exist. The test that matters uses the actual elastomer compound in your machine's seals, immersed in the actual candidate fluid — not a generic reference rubber.

Why can the same ISO VG grade shear apart over time?

Two routes reach one grade. A straight high-VI base stock can get there with little or no viscosity-index-improver polymer. Or a lower-natural-VI base stock can be boosted with VI-improver polymer to the same target number.

Polymer chains shear-degrade in service. Hydraulic systems route fluid repeatedly through high-shear elements — pumps, especially vane and gear types, and relief or flow-control valves — and that shear breaks the chains, producing permanent viscosity loss, distinct from ordinary temporary shear-thinning. A straight-graded oil has essentially no polymer to break and does not show this failure mode. A VI-improved oil can drift out of its own grade band over months of service.

ASTM D6278 and ASTM D7109 measure shear stability using a European diesel injector apparatus — D6278 reports percent viscosity loss at 100 °C after 30 cycles; D7109 evaluates at both 30 and 90 cycles, with 90 the more severe test. Neither method is confirmed to be calibrated against real hydraulic vane-pump service specifically — treat them as generic polymer-shear bench indicators.

One more currency check worth having: ASTM D2882, the constant-volume vane-pump wear test, was withdrawn in 2003 and succeeded by ASTM D7043 and the international equivalent ISO 20763. Any document still citing D2882 for present-day qualification is citing a standard that no longer exists.

If a straight-graded oil is topped up with a VI-improved oil, the blend's shear behaviour is no longer predictable from either parent's datasheet. Only a shear-stability test of the actual blend tells you what it will do.

Can a cleanliness code tell you this instead?

No. ISO 4406 reports a particle-count cleanliness code — for example 18/16/13 — quantifying particles at ≥4 µm, ≥6 µm and ≥14 µm per millilitre. It is a cleanliness standard, not a chemistry or compatibility standard.

A fluid can read perfectly clean by ISO 4406 immediately after two incompatible products are combined. A particle counter cannot tell normal wear debris from freshly forming additive-reaction gel. By the time it can, the filter-plugging mechanism is already underway. What that code does and does not tell a buyer is set out in the guide to ISO 4406 cleanliness codes.

Is a shared specification enough on its own?

No — and this is the point most procurement processes get backwards.

A shared specification, say DIN 51524-2 HLP and ISO 11158 HM at the same VG, guarantees that both fluids clear the same minimum performance thresholds: the viscosity band under ISO 3448, and the minimum anti-wear, oxidation and rust-protection pass criteria under DIN 51524-2 or ISO 11158. That is real, and it is worth something.

It does not guarantee identical base oil or API Group, identical viscosity index, identical additive chemistry, identical anti-foam or demulsibility packages, or that the two fluids remain both miscible and compatible once blended. Two products can each independently pass DIN 51524-2 or ISO 11158 while being built on fundamentally different, mutually antagonistic additive chemistries.

Here is the clinching argument, and it is not a marketing line — it is why the standard exists: ASTM D7752 was created as a separate layer on top of D6158, DIN 51524, ISO 11158 and ISO 15380 precisely because the industry recognised that shared-specification compliance was not proof of mixing safety. If matching the spec were enough, D7752 would not need to exist.

There is a second-order trap here too. ISO 11158's current edition is ISO 11158:2023, approved on 13 October 2023, and it superseded a 2009 edition that had itself superseded a 1997 one. Three editions circulate in field documentation, so "meets ISO 11158" on a datasheet is not self-verifying — an RFQ should require the edition year explicitly.

Matching a declared specification is a materially lower-risk substitution than an arbitrary, unknown-brand top-up. It bounds viscosity and minimum performance. It is not, by itself, evidence that the two products can share a reservoir.

Does a small top-up carry the same risk as a full blend?

Not automatically — and the standard itself treats them as different questions. D7752 tests at 2:98, 10:90 and 50:50 because a trace top-up, a partial blend, and a genuine 50/50 mix are three separate risk tiers, not one.

That does not make a small top-up risk-free. Hoses, cylinders and low points in a circuit hold fluid that a simple reservoir drain never removes, which is exactly why proper flushing procedures call for opening secondary and low-point drains, not just the sump plug.

Can a jar test or a familiar-sounding standard settle this for good?

Neither one, by itself.

A jar or thermal-cycling test — mix, cycle hot and cold, look for separation, colour change or particulate — is a legitimate first-pass visual miscibility screen. Trade literature is candid about where it sits: a fast option to reach for when there is no time for anything more rigorous, and no substitute for a controlled laboratory result. A clear jar proves only that the two fluids are miscible and did not grossly react under that screen. It cannot detect tribofilm antagonism, because no metal surface is present in the jar; it cannot detect filterability loss, foam-stability shift, demulsibility shift, air-release shift, or seal response, because each of those needs its own dedicated test. A blend can stay perfectly clear in a jar and still fail the filterability test outright.

It is also worth retiring one specific misreading: ASTM D6922 does not apply here. Its title is Standard Test Method for Determination of Homogeneity and Miscibility in Automotive Engine Oils, and its scope is exactly that — automotive engine oils, motivated by the fact that a vehicle operator often does not know what oil is already in the crankcase. It is not a hydraulic-fluid standard and should never be cited as one; the correct reference is ASTM D7752.

And retire the loose phrase "the additives cancel each other out" too. It is not symmetric cancellation. It is competitive surface adsorption — one additive loses its effective treat rate to a more surface-active one — or, in some acid-base cases, an outright precipitation reaction.

Before you switch or top up, what do you actually need on paper?

Two separate parties can give you real answers here — the supplier, and an independent accredited lab. Altonex Global is the B2B trade platform where you find and contact the supplier and route the technical question; it does not test, certify, inspect or approve any fluid itself.

From the supplier, ask for, in writing:

• The exact standard and edition year the product is built to (DIN 51524-2, ISO 11158 — the year matters; three editions of ISO 11158 currently circulate)
• The API base-oil Group, and ideally the aniline point or saturates/aromatics figures
• The additive chemistry family — zinc-containing or ashless/zinc-free
• Any existing ISO 13357 filterability, ASTM D892 foam, or ASTM D1401 demulsibility data — often standard rows on a finished hydraulic-oil TDS
• Whether the supplier holds a compatibility statement or a D7752-style result covering the specific incoming product, not a generic claim
• A batch-specific certificate of analysis, not a typical-values sheet

From an independent accredited lab, run on the actual blend at the intended ratio — never on the two products tested separately:

• A visual/thermal-cycling screen, as a first pass only
• ISO 13357 filterability
• ASTM D892 foam
• ASTM D1401 demulsibility
• ASTM D471 seal immersion, using the actual seal compound from the machine
• Elemental analysis to confirm additive elements have not depleted or precipitated
• Where a mixed VI-improver route is involved, a viscosity check before and after a shear pass

If a supplier claims cross-compatibility for a specific product, ask for that supplier's own documentation supporting it. Do not extend the claim to the category. And check your equipment OEM's own warranty terms and fluid guidance before any switch — that guidance sits with the OEM, outside this article and outside the platform.

So, can you mix them?

Sometimes. A shared ISO VG grade and a shared declared specification narrow the risk — they do not close it. The only way to know, for a specific pair of products at a specific ratio, is the documentation above and, where the stakes justify it, a compatibility test run on the actual blend.

Take the supplier question set into your next enquiry. Suppliers listing hydraulic fluids on the platform can be asked all six questions through a single request for quotation before anything is poured — which is a great deal cheaper than finding the answer in a filter housing.

Frequently asked questions