Lubinpla

Tramp oil in water-soluble coolant: acceptable levels and limits

Summary

UK HSE guidance MW5 recommends keeping tramp oil in water-mix metalworking fluid below 2 percent and checking it weekly (HSE, 2021). Tramp oil is machine lubricant that gets into the coolant. Treat 2 percent as an action point, not a safe ceiling: the right level also depends on the coolant type, whether soluble oil, semisynthetic or synthetic.

  • US OSHA's best practices manual says there is too much tramp oil when surface oil swished aside closes over the sump again within 5 to 8 seconds (OSHA, best practices manual).

  • In one crankshaft grinding area, haze and odor became noticeable at 6 percent tramp oil; a centrifugal separator kept it below 4 percent (HPAC, 2010).

  • The acid split method, which also estimates tramp oil in soluble oils, takes 6 to 18 hours from sample to result (Master Fluid Solutions, 2025).

  • A floating tramp oil layer cuts the coolant off from oxygen and favors anaerobic bacteria and foul smells (ExxonMobil, 2010).

These levels apply to water-mix coolants in machine sumps and central systems. They do not apply to conventional straight (neat) cutting oils, where compatibility with petroleum lubricating oil is typically not an issue because the chemistry of both oils is very similar; this is not necessarily true for synthetic or vegetable-based oils (Master Fluid Solutions, 2022). A supplier's own tramp oil limit and method for its fluid take precedence.

Tramp oil action levels for water-soluble coolant

Write the action levels into the sump log before the first reading, so that each result triggers a set action instead of a debate. Next to each level, record the sampling point, the method and the standing time. A settle test, a visual check and a laboratory split give different numbers for the same sump. The table below sets what to do at each tramp oil level.

Tramp oil level (source)

What you see

What to do at this value

Below 2 percent (HSE MW5, 2021)

Little or no free oil at quiet time

Keep the weekly check; record it with concentration and pH

2 to 4 percent (Master Fluid Solutions TB033; HPAC, 2010, one shop)

Bacterial activity starts to rise (Master Fluid Solutions)

Skim at quiet time; find the leaking way lube, hydraulic or gear oil

About 6 percent (HPAC, 2010, one shop)

Haze and odor noticeable

Remove with a centrifuge or coalescer; take a dipslide; check concentration by a second method

Surface oil closes over within 5 to 8 seconds of swishing (OSHA, best practices manual)

Continuous floating layer

Skim or pump the surface oil now; then find and fix the leak

The row that matters most is 2 percent. Below it, UK HSE asks only that tramp oil stays at a minimum, checked weekly. Above it, find the source before the oil emulsifies into the fluid. HSE MW5 names mechanical skimmers, mobile oil coalescers and vacuum systems for removing tramp oil (HSE, 2021).

The log pays off later. When a coolant change is judged on tool life or sump life, those records show whether tramp oil, not the fluid, moved the result. Lubinpla's performance evaluation fixes the measurement criteria and sampling plan before any results come in, ties each result to its sample and method, and separates the fluid's contribution from operating conditions such as tramp oil.

How do you measure tramp oil in coolant on the shop floor?

On the shop floor, use a settle test for free oil and a visual check of the sump surface. Free oil is the part that floats as a layer. Oil that has mixed into the fluid, emulsified tramp oil, is a job for a laboratory. UK HSE MW5 tells users to get advice from the fluid supplier on a suitable measurement method (HSE, 2021).

The settle test reads free oil as percent by volume. Draw a sample from a circulating sump into a 100 mL graduated cylinder, let it stand, and read the floating layer: 2 mL of oil is 2 percent (illustrative method). The ExxonMobil TOYODA separability test uses the same principle, letting 90 mL of 5 percent coolant with 10 mL of slideway lubricant rest for 16 hours before reading the oil, cream and coolant layers (ExxonMobil, 2010). Use the same standing time every week so the readings compare.

The settle test misses oil already emulsified into the fluid. For that, ExxonMobil describes laboratory titration compared with fresh coolant to estimate emulsified tramp oil (ExxonMobil, 2010). For soluble oils and high-oil semisynthetics, the supplier's acid split separates product oil from tramp oil, but it uses 50 percent sulfuric or hydrochloric acid (Master Fluid Solutions, 2025). Leave that test to the supplier's laboratory, and handle any reagent under its SDS and COSHH or your local hazardous chemical rules.

Does tramp oil throw off a coolant refractometer reading?

Tramp oil does skew the reading: refractometers tend to read higher than the actual concentration over time, because tramp oil in the fluid also refracts light (Master Fluid Solutions, 2019). A high reading can therefore be oil, not concentrate.

Heavy contamination makes the reading hard to take at all. Master Fluid Solutions adds that in a heavily contaminated system a clear reading is often not possible, so the result becomes an educated guess (Master Fluid Solutions, 2019). ExxonMobil describes the same sign: the refractometer line is normally sharp, but a significant amount of emulsified slideway lubricant obscures it (ExxonMobil, 2010). A blurred line is a tramp oil reading in itself.

In practice the danger is a dosing error. Suppose the refractometer reads high and the operator tops up with water to bring it down. The real concentration then falls below the design range, where corrosion, tool and sump life problems are frequently observed (Master Fluid Solutions, 2019). Before you correct a high refractometer reading, skim the sump and take a fresh sample below the surface. For synthetics and semisynthetics, also check by alkaline titration, which tends to read lower than actual over time instead (Master Fluid Solutions, 2019). For the decision that follows, see whether to correct coolant concentration and pH drift or dump the sump.

Do emulsion and synthetic coolants tolerate tramp oil differently?

Yes: tramp oil should readily split from a synthetic, while a soluble oil shows little splitting and heavy absorption of the oil (Elenteny and Manfreda, 2006). Oil that splits out floats where you can see and remove it; oil that is absorbed stays hidden in the fluid. So the same 2 percent reading means different things in each.

The table below shows how each coolant type handles tramp oil.

Coolant type

How tramp oil behaves (Elenteny and Manfreda, 2006)

What it means for the limit

Synthetic, no mineral oil

Should readily split from solution

Free oil is likely most of the tramp oil; a settle test and skimmer catch it

Semisynthetic, typically under 30 percent mineral oil

Some splitting, some absorption that promotes bacterial activity

Part of the oil hides in the fluid; add a lab check

Soluble oil (emulsion)

Little splitting, heavy absorption; can replace the coolant's oil and destabilize the emulsion

A low free oil reading can understate the problem

For a soluble oil, then, a settle test under 2 percent does not prove the emulsion is clean. Products that contain oil tend to absorb some tramp oil because of the emulsifiers they need, and some recycling systems that remove tramp oil also remove part of the product oil (Elenteny and Manfreda, 2006). Ask the fluid supplier how its product rejects tramp oil and which test it trusts for that fluid type.

If the shop runs several coolant types, keep a separate tramp oil action level for each type rather than one plant-wide number. Note the type on every sump log sheet, so no one ever compares a reading across fluids.

Why do published tramp oil limits differ?

Published tramp oil limits measure different things in different fluids, and only one of them is a written guidance figure. UK HSE MW5 gives a number, below 2 percent, checked weekly, without stating whether that is free oil or total oil (HSE, 2021).

One fluid maker reports very few observable problems below 2 percent, rising bacterial activity above 2 percent, and tool life, finish, mist and residue problems above 4 percent (Master Fluid Solutions, TB033). It adds that the source of the tramp oil can change these percentages substantially.

US OSHA's best practices manual gives no percentage. It uses a visual test instead, with too much tramp oil when the machinist cannot swish the oil aside for more than 5 to 8 seconds before the sump is covered again (OSHA, best practices manual). That test sees free oil only.

The HPAC figures come from a single crankshaft grinding area on a water and soluble-oil coolant, reported by an industrial hygienist: haze and odor at 6 percent, with a centrifuge keeping the level below 4 percent, and the share of cutting oil added to the water then reduced by up to 50 percent without ill effects (HPAC, 2010). The report names no measuring method and no other fluid type. It shows where one system failed, not a general limit.

ASTM E1497-23, the active ASTM practice for selection and safe use of metal removal fluids, covers maintenance among its topics (ASTM, 2023); its clause text is behind a paywall, so no ASTM figure is quoted here. Set your own action level from the HSE figure, then confirm it with the supplier for your fluid type.

Frequently asked questions

Should a tramp oil skimmer run all the time?

A skimmer works best when circulation stops, because tramp oil separates and floats once agitation ceases. US OSHA notes that skimmers and separators are particularly effective when the system pumps are not running, such as on weekends and off-shift (OSHA, best practices manual). Put the skimmer on a timer for those quiet periods.

Where does most tramp oil come from?

Most of it comes from the machine tool itself: way lube applied to the slideways, and hydraulic, gear and spindle oil leaking from seals. US OSHA recommends applying only the required amount of way lube and making sure way lubricators run only when the machine tool runs (OSHA, best practices manual).

Does tramp oil cause skin problems?

Yes, it contributes. US OSHA calls tramp oil one of the main causes of dermatitis, because machine lubricants are not designed for repeated skin contact and the oil carries metallic fines that irritate the skin (OSHA, best practices manual). A low level protects operators as well as the fluid.

Related posts