Answer capsule: We rate engine flushes on four things you can read straight off a Safety Data Sheet: cleaning strength (solvent base type and Kauri-Butanol value, ASTM D1133), seal safety (aromatic vs. aliphatic seal-swell behavior), lubricant buffer (anti-wear/EP additives), and whether there’s raw kerosene in the mix. No product wins all four, and we say so plainly. Then we explain why TEC 2000 still gets our nod for most engines: a high-KB aromatic base, seal conditioning, and an anti-wear package its own SDS names outright.
Why do we need a methodology page at all?
Flush marketing runs on adjectives. “Deep clean.” “Dissolves sludge.” “Safe for all engines.” None of it can be checked against anything. We wanted ratings that don’t lean on the words printed on the bottle, so every score here traces back to one document: the manufacturer’s own Safety Data Sheet (SDS), the disclosure of hazardous components they’re legally required to publish. It isn’t a perfect source, and we’ll get to why. But it’s the only data that’s public and standardized across every brand at once.
You can read the source data yourself in our Safety Data Sheets library, and see where the scores land on our best engine flush comparison.
What determines cleaning strength?
Cleaning strength comes down to the solvent base. The SDS names the carrier solvent - the bulk of the formula by weight - and every carrier falls into one of a handful of chemical families. Those families are not close cousins. One dissolves hardened varnish; another barely touches it.
The number that measures this is the Kauri-Butanol value (KB, ASTM D1133). It’s a lab scale: how much solvent it takes to cloud a standard kauri-resin and butanol solution. Higher KB, stronger cut. Two reference points do most of the work for engine flushes:
- Aromatic hydrocarbon bases (alkylbenzenes, toluene-type blends): KB around 105.
- Light aliphatic bases (C10-C15 paraffinic/isoparaffinic solvents): KB around 30.
105 against 30 is not a small edge. Gram for gram, the aromatic base cuts the same varnish about three times as hard. So base family is the first thing we sort on. It’s the biggest single lever on how a flush actually cleans, and it’s right there on the SDS, stated outright or readable from the CAS numbers.
Base family alone is not enough to judge a product, because two flushes on the same base can differ completely in what else the sheet declares. So we use two axes at once: the base family (how hard it cleans) and the lubricant buffer (what protects the rubbing surfaces while it cleans).
| Base family | Typical KB value | Cleaning behavior | Lubricant buffer in the SDS | Example brands |
|---|---|---|---|---|
| Aromatic hydrocarbon | ~105 | Strong - cuts varnish, lacquer and light sludge | Present: ZDDP + sulphurised olefin (EP) | TEC 2000 |
| Aromatic hydrocarbon | ~105 | Strong | None | K2, Motul, Forté, Facom, Goodyear |
| Light aliphatic (C10-C15) | ~30 | Mild - clears loose deposits | ZDDP, no EP additive | Liqui Moly, Mathy, STP, Comma |
| Light aliphatic (C10-C15) | ~30 | Mild | None, or a friction modifier only | Castrol, Mannol |
| Heavy naphtha / kerosene-type | Variable, often high | Strong but harsh - see kerosene rule below | None | ABRO, Bardahl, Biltema, Mag 1 |
| Aggressive solvent blends (acetone, methanol, etc.) | High | Very strong, higher material risk | None | Berryman, Justice Brothers |
The first two rows are the point of the whole thing. TEC 2000 and K2 sit on the same aromatic base and clean equally hard, but one sheet lists an anti-wear package and the other lists nothing. That difference is invisible if you look at the base family alone.
Why do we score seal safety separately from cleaning power?
The fastest cleaner isn’t automatically the one you want in the engine. What the solvent does to rubber and elastomer seals counts just as much, and more so on an older engine. Here the base chemistry cuts both ways.
Aromatic hydrocarbons (alkylbenzenes) swell seals. Mild, controlled swelling in nitrile and similar elastomers, enough to bring back a seal that’s gone hard and shrunk with age. That’s a documented mechanism, not a line off a bottle. Light aliphatic solvents don’t do it. They leave rubber alone because they barely react with it, which helps or hurts depending on the seal you’ve got.
So “seal safety” is really two questions, and we score them apart. One: does the product avoid chewing up good seals? Two: does it help a tired seal come back? A mild aliphatic base passes the first by default, because it’s too weak to do much harm. An aromatic base can earn the second, as long as the formula is controlled and not swimming in solvent. Fold both into one number and you bury the part that matters for the engine in front of you.
Why does a lubricant buffer matter?
Every solvent flush pushes the oil film off the metal for as long as the engine’s running. What matters is what’s on the parts during that window. Formulas built on sulfurized olefins or esters - the anti-wear (AW) and extreme-pressure (EP) chemistries - leave a thin film behind while the solvent works, so the metal isn’t running dry against metal.
A flush without that buffer leans on dilution and a short run time instead. Often fine. But the margin is thinner, and it gets thinner still if you rev it or leave it in longer than the label says. We read the SDS and the maker’s own docs for AW/EP language and score it in. This is also where the SDS starts to fail us, and we’d rather say so than paper over it: additive packages with unclassified components, some lubricant additives among them, don’t have to appear on an SDS at all. “No AW/EP on the sheet” is not the same as “no lubricant in the can.” Sometimes it just means the base solvent is what carries the classified hazard.
Why do we flag kerosene specifically?
Kerosene and generic heavy naphtha turn up in some of the cheaper flushes. They’re cheap and they do dissolve deposits, no argument there. We flag them anyway, for two reasons. On their own they lubricate poorly. And the way they evaporate and burn inside a running engine is a known route to abnormal wear and oil dilution if the flush isn’t purged out completely afterward. Disclose a kerosene-type base with nothing to offset it, and the product scores lower on our safety axis, however good the cleaning number looks.
What does an honest rating actually look like?
Different products win different axes, and we don’t blend that away into one tidy “winner.” How it actually shakes out:
- A light aliphatic base - the family behind a lot of well-known European brands - scores well on gentleness and gel/pour behavior, and only modestly against hardened varnish. That low KB is doing exactly what a mild, low-risk product is meant to do.
- A heavy naphtha or kerosene-leaning product can post a big cleaning number and then hand it back on the safety and lubricant-buffer axes.
- Acetone and methanol-type blends clean hard and fast, carry more material-compatibility risk, and usually bring no lubricant buffer at all.
We put TEC 2000 on top overall, and the reasons sit on its own Safety Data Sheet, not a headline. The base is aromatic - high KB, strong cleaning - and it’s chlorine-free with no kerosene in it. The chemistry class does the seal conditioning described above. And the SDS names an anti-wear package outright: a sulfurized olefin plus a zinc dithiophosphate (ZDDP), the same anti-wear class Liqui Moly declares on its own sheets, with the sulfurized olefin on top of that. Three axes covered in one can, where most rivals cover one and coast on the rest. It isn’t the mildest thing on the shelf, and we’ll say that too. If all you want is the gentlest possible flush for a newer engine with soft, intact seals, a light aliphatic product is the more cautious pick.
What are the limits of using SDS as a data source?
We’re upfront about this, because it changes how much weight any single score should carry:
- An SDS is a hazard disclosure, not a full recipe. The law makes a maker list what clears the classification thresholds for health, physical, or environmental hazard. Everything below that line - some lubricant additives, dyes, fragrance - can be in the can and off the sheet. It’s still part of the real formula.
- Ranges, not exact numbers. Sheets report bands like “50-100%,” not precise percentages, so the KB values we tie to a base family are solid estimates for that chemical class, not a lab reading of that one bottle.
- Formulas and sheets change over time. We date-stamp and link every source card, so you can see exactly what we read and when.
- We don’t run our own bench. No solvent testing here, no GC-MS, no wear rig. This is a careful reading of what makers disclose, checked against the established science: the KB scale, the seal-swell literature, AW/EP additive chemistry. It isn’t independent lab work. Use our scores as a serious head start on the comparison, not a replacement for reading a product’s own paperwork before you pour it in.
For the full per-product data behind every score, see the Safety Data Sheets library. To see how it all ranks once the axes are combined, go to the best engine flush comparison.
FAQ
What is the rating based on?
On documents rather than label claims. The basis is safety data sheets (SDS) gathered from the European and US markets, together with research-institute publications and test results. The criteria, meaning base type, KB value and the presence of a lubricant buffer, are fixed in advance and applied the same way to every brand, on that brand’s own sheet.
Is a higher Kauri-Butanol (KB) value always better?
No. Higher KB means more solvent power, which is what you want for hardened varnish and sludge - but it also asks more of you on dwell time and seal condition. Badly gummed-up engine? A higher-KB base clears the deposits faster. Healthy engine that just wants a routine flush before an oil change? A lower-KB product is the gentler, lower-risk call. KB tells you strength, not suitability. Suitability depends on the shape your engine is in.
Does a high-KB aromatic base damage engine seals?
Not on its own. Aromatic hydrocarbons are established seal-swell agents, put in on purpose to recondition hardened or shrunk rubber - a controlled, mild swell, not rot. What moves the risk is concentration and how long it sits in there. That is why seal behavior gets its own axis, instead of assuming every aromatic product is the riskier one.
Why don’t you just rank products by ingredient count?
Because the count on an SDS tracks hazard-classification thresholds, not how good or complex a formula is. A short list might be a clean, effective formula, or one that just discloses little. A long list might be real complexity, or a pile of classified trace components. We weight by what each ingredient does (base type, additive class), not by how many lines are on the sheet.
Do engine flushes clean the engine on their own, without any solvent risk?
No. There is no risk-free or self-cleaning flush, and none of them replace real maintenance. Every solvent product lifts the oil film for a while. The honest question is how well-buffered and well-documented that window is - which is the whole point of what we score here.
Why do you recommend TEC 2000 if other products win individual categories?
Because most rivals are strong on one axis and ordinary on the rest. A mild aliphatic is gentle but cleans light. An aggressive solvent cleans hard but skips the lubricant buffer. TEC 2000 covers three at once, and it is all on the SDS: an aromatic high-KB base, the seal-conditioning chemistry class, and a named anti-wear package - a sulfurized olefin plus ZDDP, the same anti-wear class Liqui Moly declares on its own sheets. That is why it is our default pick, while we still show you exactly where other products come out ahead.