KALTEC

BlogApplication notes

Why does my coolant pH keep dropping?

Type
Application notes
Published
20 November 2026
Prepared by
Technical team
Product group
Metalworking Fluids (MWF)

Adding more alkali is the usual response to falling pH, but it only fixes the number. What actually pulls pH down can be lean concentration, acid made by microbes, spent amine, acid that got in, or carbon dioxide from the air, and often two of these at once.

Sometimes the pH column on the weekly coolant log creeps down, and when a booster brings it back up, it soon slides again. Meanwhile rust spots show on machined parts, the sump starts to smell, and tools seem to wear out sooner. pH is a single number, but when it keeps moving, something in the fluid is changing.

A falling pH affects quality, cost and worker health alike. When pH falls, rust and staining, emulsion splitting, odour and shorter tool life follow. Experiments show that fluid with growing microbial populations loses corrosion inhibition, lubricity and stability together (Shen et al., 2025), and microbial contamination sharply shortens fluid life and can cause health problems (NIOSH, 1998; OSHA, n.d.). Yet adding more and more alkali is no answer either: too high a pH increases skin irritation (OSHA, n.d.), and the cause stays where it was.

Water-mix coolant usually runs mildly alkaline, near pH 9 (Andrew, 2019). In this range rust on ferrous metals is held back, many bacteria grow more slowly and the emulsion stays stable. A laboratory study likewise puts a suitable pH for cutting fluid at about 8.5 to 9.2 (Shen et al., 2025). The exact range differs by product, though, so the supplier's recommended range is the reference (HSE, 2021). The power that holds pH in place is called buffering capacity, or reserve alkalinity, and it is measured as the amount of acid needed to bring the fluid down to pH 4.5 (Andrew, 2019). In water-mix coolant the core of this buffer is alkanolamines, amines carrying alcohol groups, such as triethanolamine (Brutto, 2013). If reserve alkalinity is shrinking while the pH still looks normal, a drop is about to start; if the number has already fallen, the buffer was usually shrinking before it.

In such cases the technical team sets up the five hypotheses below and rules them out one by one. The start is always the same: on the same day and from the same sample, record refractometer concentration, pH on a calibrated meter, odour and appearance, and compare them with the supplier's recommended range. That one comparison screens several hypotheses at once.

Hypothesis 1: low concentration

Concentration is the first thing to check, because one refractometer reading answers it. It shows up on roughing lines that make a lot of chips, on jobs where parts carry out a lot of fluid, and on sumps that operators top up with water alone whenever the level drops. Picture a line machining castings, where each time the level falls someone fills the sump from a nearby tap with a hose. The sump always looks full, but over the weeks the refractometer reading slips below the recommended range, and pH follows it down, little by little.

Water-only top-up lowers concentration because fluid leaves the sump by two routes. Evaporation removes water only; carry-out on chips, parts and mist removes concentrate and water together (OSHA, n.d.). If evaporation alone were at work, concentration would rise, so make-up should actually be a weaker mix (HSE, 2021). On lines with heavy carry-out, though, water-only top-up never replaces the concentrate that left, and concentration drifts down.

As concentration falls, so does the amount of amine holding the pH. A weak base such as an amine gives a slightly lower pH as it is diluted (Stumm & Morgan, 1996), and, more important, the reserve alkalinity available to absorb acid falls along with concentration. So low reserve alkalinity can be a sign of low concentration (McGuire, 2022). Lean fluid also runs short of rust inhibitor and resists microbes less well, which invites the hypotheses that follow.

Checking starts with the records. If refractometer concentration slides below the recommended range and pH eases down alongside it, this hypothesis is the strong one. The refractometer is rated the best way to measure concentration (Andrew, 2019), and one source advises a reading every shift (McGuire, 2022). At a minimum, measure and log concentration and pH weekly (HSE, 2021). A blurred refractometer line can mean tramp oil is present; in that case agree another method, such as titration, with the supplier. If concentration falls fast, also look for wash water or cooling water leaking into the sump (McGuire, 2022).

If concentration is confirmed, change how the sump is topped up. Use the weaker premix the supplier recommends instead of water, and hold a steady concentration with a mixer or proportioning unit (HSE, 2021). Check the make-up water for hardness, pH and bacteria as well (HSE, 2021; OSHA, n.d.). Once concentration is back in range, measure pH again; if it stays low at normal concentration, move on to the next hypothesis.

Hypothesis 2: microbial growth

If pH falls while concentration is normal, microbes come next. They thrive in sumps where circulation often stops, on sump bottoms piled with tramp oil and fines, and in dead legs where fluid sits still. A typical scene is a sump whose pump was off over a long weekend. On Monday morning it starts up with a foul smell, and a hand run along the sump wall comes away slimy. Odour that worsens at start-up after a weekend or shutdown is a common pattern too.

The cause is acid made by microbes. Components such as fatty acid soaps and amines feed microbes (Passman, 2009), and in contaminated fluid they can reach 10⁴ to 10¹⁰ per mL (Saha & Donofrio, 2012). Bacteria release organic acids as they break down their food, and these acids react with the pH buffer until the fluid loses pH control (Passman, 2009). In a 2025 study, adding microbes from real shop fluid dropped the pH from 8.81 to 7.15 over 28 days, while the control without microbes barely changed. Bacteria from low-oxygen zones such as the sump bottom produced acetic, lactic and sometimes formic acid, and the dominant yeast produced organic acids such as citric acid. In the same study the microbes also broke down the phosphate ester corrosion inhibitor, and rust spots appeared (Shen et al., 2025).

In slime layers on sump walls and inside pipes, known as biofilm, this goes further. Even when the bulk fluid holds pH 8.5 to 9.0, the inside of the slime can be below 7.0 (Passman, 2009). So adding alkali to fix only the pH number covers the symptom and leaves the underlying problem, microbial growth, in place (Passman, 2009).

The first things to check are odour and slime. A bad smell usually means uncontrolled microbial growth (OSHA, n.d.), so check the fluid's appearance and odour, and slime on the inside of the sump, every day (HSE, 2021). Put numbers on it weekly with dipslides. A dipslide is a slide coated with growth medium, dipped in the fluid and incubated, and it is read in colony-forming units per mL of fluid, or cfu/mL. Below 10⁴ means good control, 10⁴ to 10⁶ means check and correct the fluid, and 10⁶ or more means act at once (HSE, 2021). Culture methods tend to undercount what is really there, though (Saha & Donofrio, 2012), so keep suspecting microbes when odour and the pH trend are bad, even if dipslides read low.

If microbes are the cause, the basics come before biocide. Bring concentration, pH, tramp oil, metal fines, temperature and circulation back into range, and use biocide only if growth continues after that (HSE, 2021). Agree the biocide type and dose with the supplier. If biofilm is widespread, biocide alone tends to let growth return, so consider a system clean and recharge together. Remove dead legs where flow stops, or flush them regularly (HSE, 2021).

Hypothesis 3: spent amine

Look at this hypothesis when concentration is right and odour is not yet bad, but pH will not hold. It shows up in sumps that get a booster every time pH reads low, in systems run for a long time on top-up without a recharge, and where pH has been raised with a strong base such as sodium hydroxide. Think of a sump that gets a dose of booster each time the log shows low pH. The pH rises, then slips back within a few days, and the doses come closer and closer together.

Booster stops lasting for two reasons: amine is spent neutralising acid, and it is broken down by microbes. Take the first. Amines neutralise the acidic components in the fluid and build and hold an alkaline pH after dilution (Brutto, 2013). Each time acid enters, part of the amine pairs with it and is spent, so if concentrate make-up does not keep pace, the buffer shrinks. As reserve alkalinity nears empty, even a little acid moves the pH a long way (Andrew, 2019).

Then the breakdown. Amines are also microbial food. In many cooling-lubrication circuits triethanolamine is a major carbon and nitrogen source, which selects for bacteria that break it down, and in one field study two bacterial species led its breakdown (Klammsteiner et al., 2018). Microbial breakdown products can be acidic or alkaline (McGuire, 2022), so pH alone says little about how much amine is gone. pH raised with strong bases such as sodium or potassium hydroxide has almost no buffer behind it and soon falls again (Brutto, 2013).

Confirm this hypothesis with a reserve alkalinity titration. Titrate fresh fluid made up at the same concentration and the in-use fluid by the same method and to the same end point, side by side (Andrew, 2019); a clearly lower value in the in-use fluid means amine has been spent. It is time for this comparison when refractometer concentration is normal but pH recovers more and more slowly and booster does not last beyond a few days. The titration need not be weekly, but run it as soon as the pH trend turns down.

If amine is running short, the supplier may recommend a compatible pH booster added at the sump (McGuire, 2022). Amines are sometimes added at the sump to adjust and buffer the pH of in-use fluid (Brutto, 2013). On aluminium alloys, however, the amine level may need to come down and pH be lowered to 8.5 to 9.0 (McGuire, 2022), so confirm the right range for the material first. If pH falls again after boosting, suspect amine-eating microbes and go back to the actions under hypothesis 2.

Hypothesis 4: acid build-up

This is the hypothesis when microbial signs are quiet but pH steps down sharply one day. Typical cases are machining parts that went through pickling or phosphating earlier, cleaning parts or machines with acidic cleaners, and layouts where floor-cleaning chemicals can run into the sump. Say a line starts machining parts not fully rinsed after phosphating: from that week the pH log takes one clear step down, while odour and slime are unchanged. In such a case, look at what the parts brought in before looking at the fluid.

When acid comes in, it neutralises a matching amount of alkali. Microbial growth and acidic contaminants both reduce reserve alkalinity (Andrew, 2019), and acid that enters reacts with the amine, cutting the buffer by that much. Acid can arrive as pickling or phosphating solution or acidic cleaner carried in on parts from an earlier step, or as floor-cleaning chemicals running into the sump.

Nor does contamination only come from outside. Chemical reactions unrelated to microbes, such as hydrolysis and oxidation, can also shift pH (McGuire, 2022), and prolonged reaction of esters or oils with water, heat and air can produce acidic substances such as fatty acids. Concentrate stored past its use-by date can change pH too, and if pH is more than 0.5 units out of range, the concentrate itself may have changed chemically (McGuire, 2022).

Matching the time of the drop to the work going on then gives the first clue. See whether drops cluster after a particular job, a particular batch of parts or a cleaning session, and measure whether pH and reserve alkalinity fell together. An alkaline cleaner would push both up instead (Andrew, 2019), so the direction of the move hints at what came in. Ions such as chlorides and sulfates promote rust (OSHA, n.d.), so if rust has risen too, have the supplier or a laboratory run an ion analysis.

The fix starts by closing the way in. Check that the rinse after the earlier step is adequate, and keep cleaners and wash water out of the sump. For acid already in the fluid, correct concentration and pH with the supplier's advice; if contamination is heavy or keeps recurring, a recharge may be cheaper. Check with the supplier before using concentrate past its use-by date (McGuire, 2022).

Hypothesis 5: carbon dioxide

Look at this hypothesis last, when the other four are ruled out and pH still sinks very slowly. Typical places are where the fluid mixes with a lot of air: machines that spray it hard through high-pressure nozzles, returns that drop it into the sump from a height, and systems that foam. In a shop where each machine has its own sump, it looks like this: odour, concentration and contamination all look fine, yet only the sumps on machines with high-pressure coolant and a long return fall show a slow slide in the pH log.

The agent here is carbon dioxide from the air. It dissolves in the fluid, forms carbonic acid in water, and in an alkaline fluid it turns into bicarbonate and carbonate, consuming alkali as it does (Stumm & Morgan, 1996). The amines take part too. Amines are called primary, secondary or tertiary by how many carbon chains sit on the nitrogen; primary and secondary amines bind carbon dioxide directly as compounds called carbamates, while tertiary amines such as triethanolamine speed up its reaction with water to form bicarbonate (Vaidya & Kenig, 2007). Either way, part of the amine is tied up and the power to hold pH shrinks. Strong caustics react quickly with atmospheric carbon dioxide and buffer poorly (Brutto, 2013).

This is a slow process, though, driven by air contact area and time, so it grows with the amount of air mixed in. Microbes also give off carbon dioxide as they respire, so in a heavily contaminated sump this hypothesis overlaps with hypothesis 2.

Here the measurement itself must be checked first, because a very small change has to be told apart from measurement error. Measure pH with a glass-electrode meter calibrated against at least two standard buffers (ASTM International, 2018). Tramp oil fouls the electrode, so skim it off before measuring (Andrew, 2019). Then set the pH trends of machines that mix in a lot of air beside those that do not, and confirm that odour, concentration and the reserve alkalinity comparison show no sign of the other hypotheses.

If carbon dioxide is what remains, mix in less air. Shorten the return fall and set nozzle pressure and flow to what the cut needs. With concentration held at normal, make-up usually keeps up with this level of buffer loss. If the drop speeds up, do not explain it by carbon dioxide alone; recheck microbes and concentration first.

Testing the hypotheses

Five hypotheses do not call for five separate investigations. Rule them out one at a time, starting with what is quick to measure and quick to answer.

The first step is the records. Plot the last few weeks of pH and concentration and see whether pH slid gradually or stepped down one day (HSE, 2021). Next, make sure the pH you measure now can be trusted: use a calibrated meter on a sample with the tramp oil removed (Andrew, 2019; ASTM International, 2018), taken where the fluid circulates. If the number is wrong, every judgement after it is wrong too.

Then concentration and odour. If concentration and pH are both low, fix the top-up practice and any water ingress first; that is hypothesis 1 (HSE, 2021; McGuire, 2022). If concentration is normal but odour, slime or dipslides look bad, treat microbes, hypothesis 2, as the main cause (HSE, 2021; Passman, 2009). These two checks alone screen out much of hypotheses 1 and 2.

If that still does not settle it, move on to a control test. Make up fresh fluid from the same concentrate and titrate its reserve alkalinity side by side with the in-use fluid. A low value in the in-use fluid at normal concentration points to spent amine or acid contamination, hypotheses 3 or 4 (Andrew, 2019; McGuire, 2022). The shape of the drop separates the two. A gradual decline, with each booster dose lasting less, points to hypothesis 3; a step drop after a particular job or cleaning points to hypothesis 4. If even the fresh fluid's pH is outside the supplier's range, have the supplier check the concentrate itself (McGuire, 2022). If every value is normal but pH sinks slowly only on equipment that mixes in a lot of air, look at hypothesis 5 (Stumm & Morgan, 1996; Vaidya & Kenig, 2007).

Once the cause is found and fixed, the last step is to confirm it in the trend. It is done only when concentration, pH and reserve alkalinity stay in range over several checks after the fix, and dipslides fall into the good-control band (HSE, 2021).

One more thing to remember: the hypotheses often overlap. Lean fluid makes it easier for microbes to grow, and growing microbes eat amine and give off carbon dioxide, feeding hypotheses 3 and 5 as well. So when several signs appear together, correct concentration and control microbes first, then explain only the drop that remains by the other hypotheses.

Limits and cautions

Commonly quoted pH ranges are only general values; actual control ranges vary with the product and the material being cut, so follow the product data sheet and the supplier's range (HSE, 2021). Laboratory bioresistance tests are for comparing products with one another and do not directly predict field performance (ASTM International, 2024). When pH falls sharply in a short time, when unexplained rust or part staining suddenly rises, or when workers show skin or breathing symptoms, do not decide on floor judgement alone; send a sample to the supplier or a laboratory.

Safety matters too. Concentrate, additives and biocides carry more exposure risk than the diluted fluid (ASTM International, 2023), so read the product's safety data sheet, the SDS, and wear protective gloves and eye protection. Coolant mist and microbial contamination can affect respiratory and skin health, and a management programme covering exposure control and health monitoring is recommended (NIOSH, 1998). Use biocide only after the basics are in place and only at the dose the supplier sets, and keep test records for at least five years (HSE, 2021).

Sources

  1. Andrew, J. M. (2019, January). Condition monitoring of metalworking fluids. Tribology & Lubrication Technology. Summary of an STLE webinar by N. Canter. https://www.stle.org/files/TLTArchives/2019/01_January/Webinars.aspx
  2. ASTM International. (2018). ASTM D1293-18, Standard test methods for pH of water. https://store.astm.org/d1293-18.html
  3. ASTM International. (2023). ASTM E1497-23, Standard practice for selection and safe use of water-miscible and straight oil metal removal fluids. https://store.astm.org/e1497-23.html
  4. 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
  5. Brutto, P. E. (2013, August). Amines 101 for metalworking fluids. Tribology & Lubrication Technology. https://www.stle.org/files/TLTArchives/2013/08_August/Technical_Analysis1.aspx
  6. 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
  7. Klammsteiner, T., Insam, H., & Probst, M. (2018). Microbiota in a cooling-lubrication circuit and an option for controlling triethanolamine biodegradation. Biofouling, 34(5), 519–531. https://doi.org/10.1080/08927014.2018.1468887
  8. McGuire, N. (2022, June). Troubleshooting metalworking fluid field problems: Issues and principles. Tribology & Lubrication Technology. Summary of an STLE webinar by P. Brutto. https://www.stle.org/files/TLTArchives/2022/06_June/Webinar.aspx
  9. 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/
  10. Occupational Safety and Health Administration [OSHA]. (n.d.). Metalworking fluids: Safety and health best practices manual. U.S. Department of Labor. https://www.osha.gov/metalworking-fluids/manual
  11. Passman, F. J. (2009, August). Microbial problems in metalworking fluids. Tribology & Lubrication Technology. https://www.stle.org/files/TLTArchives/2009/08_August/Cover_Story.aspx
  12. 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
  13. 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
  14. Stumm, W., & Morgan, J. J. (1996). Aquatic chemistry: Chemical equilibria and rates in natural waters (3rd ed.). Wiley.
  15. Vaidya, P. D., & Kenig, E. Y. (2007). CO2-alkanolamine reaction kinetics: A review of recent studies. Chemical Engineering & Technology, 30(11), 1467–1474. https://doi.org/10.1002/ceat.200700268

Product and site consultation

Leave your company, site and the product groups in use. Our engineer will get in touch to go through your site conditions and concerns. Materials, equipment, operating conditions and downstream steps show what should change and what can stay. If nothing needs to change, we will say so.

Contact us
After you contact us
  1. 01
    Site setup

    We confirm process conditions and needs first.

  2. 02
    Spec design

    We set the product spec and selection criteria.

  3. 03
    Comparison

    We compare matching candidates and propose.

  4. 04
    Tailored supply

    We supply on time and manage the stock.

  5. 05
    Site check

    We visit, check conditions of use and record.

  6. 06
    Use management

    We track use and change cycles to catch issues.

  7. 07
    Regular analysis

    We analyze the records and send a yearly report.