BlogApplication notes
Why does my metalworking fluid smell?
- Type
- Application notes
- Published
- 20 November 2026
- Prepared by
- Technical team
- Product group
- Metalworking Fluids (MWF)
Dosing biocide or dumping the whole sump is the usual response to a smelly coolant, but if the cause stays, the fresh fluid goes the same way. What actually makes the odour can be lean concentration, a tramp oil layer, stopped circulation, bacterial growth, entrenched biofilm or fungal mats, and often two or three of these at once.
On the first morning after a long holiday, a machine starts up and a rotten-egg smell rises from the sump and spreads across one end of the shop. On ordinary days a sour smell hangs around the machines too, and operators suggest adding fragrance or dumping the whole sump. Odour is felt first as a nuisance, but it is really a sign that the fluid in the sump has already started to change.
Odour is not the only loss. When microbes grow, components break down, the pH falls, and corrosion protection, lubricity and emulsion stability decline together (Shen et al., 2025). The UK Health and Safety Executive, HSE, warns that fluid deteriorates in use and that breathing mist from contaminated fluid can cause lung diseases such as occupational asthma and hypersensitivity pneumonitis (HSE, 2021). The US National Institute for Occupational Safety and Health, NIOSH, has reviewed the link between metalworking fluid aerosols and respiratory disease (NIOSH, 1998). Dumping a sump over odour does not remove the cause, so the fresh fluid follows the same path and the cost of recharging comes round again.
A water-mix coolant is a concentrate of base oil, emulsifiers, corrosion inhibitors and additives such as amines, diluted in water (Brinksmeier et al., 2015). Working concentrations usually run from 3 to 15%, and in 2018 these products made up 90% of global metalworking fluid demand (Passman & Küenzi, 2020). Mostly water with dissolved organic food, the fluid makes a good home for microbes (Saha & Donofrio, 2012). Bacteria that use oxygen, the aerobic bacteria, break these components down and give off volatile organic compounds, odour molecules that escape into the air (McGuire, 2016), while where oxygen is cut off, bacteria that live without it make rotten-egg hydrogen sulfide (Muyzer & Stams, 2008). The kind of odour and when it is worst are therefore the first clues to the cause.
In such cases the technical team sets up the six hypotheses below and rules them out one by one. The start is always the same: note whether the smell is rotten egg, sour or musty and when it is worst, then on the same day measure concentration, pH, tramp oil and a dipslide, and hold them against the supplier's recommended range and the control values (HSE, 2021).
Hypothesis 1: lean concentration
This hypothesis comes up first for sumps topped up with water alone, for shops that mix concentrate and water without metering, and for machines where the level is kept up but concentration is never read. Open the top-up log on such a machine and the entries may say only "water", week after week. An odour that at first appeared faintly in the afternoon now greets the morning shift as well.
A fluid's ability to hold out against microbes is called bioresistance. It comes from the biocide, where the concentrate contains one, from pH components such as amines and from other additives, all present at the intended concentration (ASTM International, 2024; Passman & Küenzi, 2020). Water-only top-up or unmetered mixing lowers the concentration and dilutes every one of them in the same proportion. Concentration keeps shifting through breakdown, contamination and evaporation (McGuire, 2016), so topping up by feel tends to drift low. A loop forms as well. As microbes grow they break down the pH components (Shen et al., 2025), resistance falls further, and the microbes grow faster still. Biocide is also used up over time, so monitoring and keeping up its level helps hold microbe numbers down (McGuire, 2016).
Read concentration on a refractometer at least weekly and compare it with the supplier's range (HSE, 2021). Heavy contamination can skew refractometer readings, so follow the supplier's correction factor and method. Log how much concentrate and how much water went in, too. Many water-only days, with falling pH and rising dipslide counts during the low-concentration periods, make this the strong hypothesis.
Mix fresh fluid at the supplier's concentration. In use, water evaporates and the fluid gets stronger, so top up with a mix weaker than the recommended concentration, but never with plain water (HSE, 2021). If concentration has already fallen below the range, bring it back with concentrate and recheck pH and dipslides over the next few days. Consider extra biocide only when bacteria still grow with the concentration in range (HSE, 2021).
Hypothesis 2: tramp oil layer
A tramp oil layer builds up on machines with leaking hydraulic lines or generous slideway lubrication, and in broad, quiet sumps whose surface barely moves. A brown band of oil rings the edge of the sump, and when a stick pushes the surface, a rainbow film parts and closes again at once. The smell comes up stronger where the film is pushed aside.
Tramp oil is the hydraulic, slideway or gear oil that leaks in from the machine (HSE, 2021). Even a thin oil film can seal a whole sump off from the air, creating conditions that favour microbial growth, and the oil can also break the emulsion (McGuire, 2016). Beneath the sealed surface, oxygen runs out fast, which favours the anaerobic bacteria that make hydrogen sulfide (McGuire, 2016; Muyzer & Stams, 2008). The oil itself is food as well: bacteria common in metalworking fluids break down many components, including hydrocarbons, the main part of oil (Saha & Donofrio, 2012). Systems with no tramp oil at all still get Monday morning odour, though (McGuire, 2016), so tramp oil is a condition that feeds odour, not the only cause.
Look at the surface film and the band at the edges, and measure tramp oil weekly, keeping it below 2% by a method agreed with the supplier (HSE, 2021). Before dipping a dipslide, stir the sump so oil does not coat the slide (HSE, 2021). Logging whether hydraulic top-ups are rising and how much slideway oil goes in makes the leaks easier to find.
With heavy tramp oil, cut the leaks first, then remove the oil that gets in with skimmers, mobile coalescers or vacuum units (HSE, 2021). Removal works best before the oil has time to emulsify into the fluid. If 2% is often exceeded, skim more often and put seal and line checks on the maintenance plan.
Hypothesis 3: stopped circulation
This is the hypothesis when the odour is worst at the first start after a weekend or long break and fades after a few hours of running. It shows up in plants that switch pumps off completely during shutdowns, on machines that each have their own small sump, and in sumps with chips and sludge on the bottom. When a machine whose pump was switched off on Friday evening starts up on Monday morning, a rotten-egg smell spreads past the guards with the first circulation. By lunchtime it is much weaker.
While the pumps run, air keeps mixing in and some oxygen stays dissolved in the fluid. When circulation stops, aerobic bacteria quickly use up that oxygen, and odour compounds and hydrogen sulfide build up in the still fluid. When the pumps restart, these gases escape all at once before they can be oxidised, which is often called Monday morning odour (McGuire, 2016). Hydrogen sulfide, the source of the rotten-egg smell, is made by sulfate-reducing bacteria (McGuire, 2016). These anaerobic bacteria breathe sulfate, a sulfur compound dissolved in water, instead of oxygen and give off hydrogen sulfide as a result. They are found throughout nature and industrial systems and are known to take part in metal corrosion (Muyzer & Stams, 2008). Anaerobes usually grow where aerobic bacteria have already used up the oxygen (McGuire, 2016).
The key clue is when the smell appears. If the rotten-egg smell is strong at the first start after a stop and fades with running, look at this hypothesis first (McGuire, 2016). Do not trust the strength of the smell. People can smell very low levels of hydrogen sulfide, but at 100 ppm and above the sense of smell can be knocked out (ATSDR, 2016), so a smell that suddenly vanishes does not mean the air is safe. Also look for places air cannot reach, such as chip piles and sludge on the sump floor or an oil film on the surface (HSE, 2021; McGuire, 2016).
The remedy is to keep the fluid from standing still. HSE advises keeping the fluid circulating with pumps so it does not stagnate in sumps and pipes (HSE, 2021). Agree with the supplier how to circulate or aerate at intervals during shutdowns. Once Monday morning odour has appeared, it takes several hours of ventilation and circulation to clear the air (McGuire, 2016). If it keeps recurring, look at tramp oil under hypothesis 2 and bottom sludge under hypothesis 5 together.
Hypothesis 4: bacterial growth
When there is a sour or putrid smell all the time, regardless of stops, look at bacterial growth. It shows up in sumps run on top-up for a long time without a recharge, in systems where concentration and tramp oil control have slipped, and in sumps whose pH log has been sliding over several weeks. Put a sample from the sump in a clear bottle next to fresh fluid: the used fluid is cloudier and off-colour, and it smells sour when the cap comes off.
The odour comes from bacteria feeding. Diluted coolant holds plenty of organic food for bacteria, such as oil, fatty acids, amines and emulsifiers (Saha & Donofrio, 2012), and aerobic bacteria break these down for energy, releasing odour compounds as they go (McGuire, 2016). Most components of the fluid can serve as microbial food, so once bacteria are established their food rarely runs out (Passman & Küenzi, 2020; Saha & Donofrio, 2012). As breakdown proceeds, the pH falls too. Aerobic bacteria consume the amines that hold the pH up, such as triethanolamine, and anaerobic bacteria in oxygen-poor spots release organic acids, sour like vinegar. In a 2025 experiment, coolant inoculated with bacteria dropped from pH 8.81 to 7.15, and corrosion protection and lubricity declined too (Shen et al., 2025). A fluid that has started to smell may therefore be losing pH and corrosion protection at the same time.
Check odour and appearance daily: whether the fluid has turned cloudy or changed colour, and whether there is slimy film on the sump wall (HSE, 2021). Put numbers on it weekly with dipslides. A dipslide is a paddle coated with growth medium, dipped in the fluid and incubated for a few days so the colonies can be counted; 10⁴ per mL or more means bacteria are growing, and 10⁶ or more means control has broken down (HSE, 2021). Measure pH weekly too, and see whether it slides over several weeks with no obvious event (HSE, 2021; Shen et al., 2025).
Even at 10⁴ or more on the dipslide, do not reach for biocide first. Correct concentration, pH, tramp oil, metal contamination, temperature and circulation, and if bacteria keep growing, add biocide only at the supplier's dose, at a point of good mixing (HSE, 2021). Choose the biocide by the steps a standard sets out: target microbes, efficacy, compatibility with the fluid and regulation (ASTM International, 2022). At 10⁶ or more, the rule is to drain and clean the system without delay (HSE, 2021).
Hypothesis 5: entrenched biofilm
The mark of this hypothesis is the same smell coming back soon after a recharge. It develops in pipework with dead ends, branch lines the fluid passes through only now and then, sumps with many baffles, and systems with chips and fines piled on the bottom. Picture a few days of fresh fluid without odour, then one morning the old smell rising again. A torch shone behind the baffles shows black sludge piled up, and the walls are coated with a slick film.
Where fluid barely moves, microbes attach to the walls and form biofilm, a mass of cells living together in a slimy layer they make themselves (HSE, 2021). Chip and fines piles on the sump floor become oxygen-free hiding places as well (HSE, 2021), and biofilm and fungal mats can also clog the system (Saha & Donofrio, 2012). The awkward part is that biocides work poorly on biofilm. In a 2025 study, biocidal agents prevented new biofilm from forming but had little effect on established biofilm (von Känel et al., 2025). In a survey of two US machine shops, microbes quickly recolonised the systems even after the spent fluid was dumped and the systems carefully cleaned, which the researchers put down to incomplete removal of the existing population (Özcan et al., 2019). Replace only the fluid and leave the film on walls and floor, and the surviving microbes can spread into the fresh fluid again.
Checking starts with the pattern of recurrence. If odour and dipslide counts come back within a week or two of a recharge, look here first (Özcan et al., 2019). Then check by eye for slime on sump walls and behind baffles, black sludge on the floor, and clogged filters and strainers (HSE, 2021). Marking the dead ends and unused branches on a map of the pipework makes it clear where work is needed (HSE, 2021).
Remove dead ends, and remove branch lines where practicable; flush the ones that must stay on a regular schedule, such as weekly (HSE, 2021). Clear chips with the machine stopped and protective gloves on, and reduce fines with centrifuges, filters or magnetic separators (HSE, 2021). At a recharge, run the supplier's system cleaner before draining to loosen the biofilm, clean the sump and lines by hand, then fill with fresh fluid. Check dipslides more often in the first few weeks after cleaning to catch any return early.
Hypothesis 6: fungal mats
Suspect fungi when filters and nozzles keep blocking and clumps appear on machine surfaces. They show on surfaces that go wet and dry, such as chip conveyors, in the inside corners of guards, and in systems where the fluid is replaced but machine surfaces are left as they are. Think of a machine with cotton-like or jelly-like clumps under the chip conveyor and in the corners of the guard, where the operator clears a blocked nozzle several times a day. The smell may come across as musty rather than rotten egg.
Fungi are common residents of coolant systems, alongside bacteria. A 2024 study treated fungal contamination of metalworking fluid systems as the norm rather than the exception, and the moulds isolated from fluids in use belonged mostly to the genus Fusarium. Fungi spread thread-like hyphae and grow as mats on machine surfaces and sump walls (Ruiz et al., 2024), and these mats clog filters and lines (Ruiz et al., 2024; Saha & Donofrio, 2012). Fungi often survive on machine surfaces rather than in the fluid, so they spread again after a recharge. In the same study, fungicide effectiveness was limited and depended on the fluid's chemistry (Ruiz et al., 2024).
Search by eye first. Note whether there are cotton-like or jelly-like clumps on sump walls, chip conveyors and the inside of guards, and how often filters and nozzles block. Test with dipslides that show both bacteria and fungi, and when a clump is visible, send the clump itself as a sample. A low bacteria count does not mean fungi are absent.
If it looks like fungi, stop the machine and remove the visible mats by hand. Fungicide alone does little against surface mats (Ruiz et al., 2024), so pair it with system cleaning. When choosing a fungicide, confirm its compatibility with the fluid in use with the supplier and follow the selection steps of the standard (ASTM International, 2022). Where fungi keep returning, deal with the stagnant zones of hypothesis 5 together with surfaces that go wet and dry, such as chip conveyors.
Testing the hypotheses
The cheapest tool for tracing an odour is a record. A few days of noting whether the smell is rotten egg, sour or musty, and whether it is worst on Monday morning or constant, sets the starting point. A rotten-egg smell that is strong only after stops points first to hypothesis 3; a constant sour or putrid smell, to hypothesis 4; a musty smell with visible clumps, to hypothesis 6.
Next come the four numbers taken on the same day: concentration, pH, tramp oil and a dipslide (HSE, 2021). Low concentration means hypothesis 1, so bring it into range before judging anything else. Tramp oil at 2% or more means hypothesis 2: skim it and find the leaks (HSE, 2021). Both can be fixed directly by the numbers, so correcting them first, even when other hypotheses are also present, makes the rest easier to see.
The dipslide shows how far things have gone. At 10⁴ or more, the bacterial growth of hypothesis 4 is under way; at 10⁶ or more, prepare to drain and clean (HSE, 2021). If fungi also show, look at hypothesis 6 alongside. Setting the numbers beside those of another machine on the same concentrate also shows whether the problem lies in the fluid itself or in that machine's sump and pipework.
The last check is to walk the system and watch for recurrence. Look for bottom sludge, wall slime and dead-end pipes, and if the odour returns within a week or two of a recharge, treat it as leftover biofilm and change the cleaning method (Özcan et al., 2019; von Känel et al., 2025). After any fix, keep the numbers in a table or graph to follow the trend (HSE, 2021), and confirm that concentration, pH and dipslides stay in range over several checks in a row.
Hypotheses often overlap. When tramp oil seals out air, low concentration weakens resistance and bacteria hide in bottom sludge, a hydrogen sulfide smell breaks out every weekend. In such cases, correct concentration and tramp oil first, and only then move on to biocide and cleaning (HSE, 2021).
Limits and cautions
Recommended concentration and pH differ by fluid, so the supplier's technical data come first for the numbers (HSE, 2021). Dipslides are a field indicator and do not catch every microbe, so if odour persists with low counts, send samples to a laboratory. Standard test methods also exist for evaluating a fluid's bioresistance (ASTM International, 2024). Dispose of spent fluid under local regulations.
If hydrogen sulfide is suspected, protect people first. At high levels the sense of smell fails, so never judge danger by odour (ATSDR, 2016), and if anyone must enter an enclosed tank or pit, follow gas testing and confined-space procedures first. Where workers develop a cough, breathlessness or hand dermatitis, treat it as a health issue rather than an odour issue and consult an occupational health professional (HSE, 2021; NIOSH, 1998). Handle biocides and cleaners according to the product's safety data sheet, the SDS, with the right protective equipment and only at the supplier's dose (HSE, 2021), and since biocide rules differ by country, confirm the product is allowed (ASTM International, 2022).
Sources
- Agency for Toxic Substances and Disease Registry [ATSDR]. (2016). Toxicological profile for hydrogen sulfide and carbonyl sulfide. U.S. Department of Health and Human Services. https://www.atsdr.cdc.gov/ToxProfiles/tp114.pdf
- ASTM International. (2022). ASTM E2169-22, Standard practice for selecting antimicrobial pesticides for use in water-miscible metalworking fluids. https://store.astm.org/e2169-22.html
- ASTM International. (2024). ASTM E2275-24, Standard practice for evaluating water-miscible metalworking fluid bioresistance and antimicrobial pesticide performance. https://store.astm.org/e2275-24.html
- Brinksmeier, E., Meyer, D., Huesmann-Cordes, A. G., & Herrmann, C. (2015). Metalworking fluids: Mechanisms and performance. CIRP Annals, 64(2), 605–628. https://doi.org/10.1016/j.cirp.2015.05.003
- Health and Safety Executive [HSE]. (2021). MW5, Managing fluid quality: COSHH essentials for machining with metalworking fluids. https://www.hse.gov.uk/pubns/guidance/mw5.pdf
- McGuire, N. (2016, April). Minding the metalworking fluids. Tribology & Lubrication Technology. https://www.stle.org/files/TLTArchives/2016/04_April/Feature.aspx
- Muyzer, G., & Stams, A. J. M. (2008). The ecology and biotechnology of sulphate-reducing bacteria. Nature Reviews Microbiology, 6, 441–454. https://doi.org/10.1038/nrmicro1892
- National Institute for Occupational Safety and Health [NIOSH]. (1998). Criteria for a recommended standard: Occupational exposure to metalworking fluids (DHHS [NIOSH] Publication No. 98-102). https://www.cdc.gov/niosh/docs/98-102/
- Özcan, S. S., Dieser, M., Parker, A. E., Balasubramanian, N., & Foreman, C. M. (2019). Quorum sensing inhibition as a promising method to control biofilm growth in metalworking fluids. Journal of Industrial Microbiology & Biotechnology, 46, 1103–1111. https://doi.org/10.1007/s10295-019-02181-7
- Passman, F. J., & Küenzi, P. (2020). Microbiology in water-miscible metalworking fluids. Tribology Transactions, 63(6), 1147–1171. https://doi.org/10.1080/10402004.2020.1764684
- Ruiz, C., von Känel, G., Burkard, S., & Küenzi, P. (2024). Fusarium spp. in metalworking fluid systems: Companions forever. Pathogens, 13(11), 990. https://doi.org/10.3390/pathogens13110990
- Saha, R., & Donofrio, R. S. (2012). The microbiology of metalworking fluids. Applied Microbiology and Biotechnology, 94, 1119–1130. https://doi.org/10.1007/s00253-012-4055-7
- Shen, Y., Zhang, W., Wu, L., Dong, Y., Guo, G., Dong, L., & Guo, Z. (2025). Microbial proliferation deteriorates the corrosion inhibition capability, lubricity, and stability of cutting fluid. Frontiers in Microbiology, 16, 1522265. https://doi.org/10.3389/fmicb.2025.1522265
- von Känel, G., Steinmann, L. Y., Mauz, B., Lukesch, R., & Küenzi, P. (2025). It is useless to resist: Biofilms in metalworking fluid systems. Life, 15(6), 890. https://doi.org/10.3390/life15060890