Key takeaways
- One titration point on a 3,000 L bath is 0.01 eq/L, about 1.2 kg of caustic equivalent, so a 4-point drop is a 4.8 kg top-up and not a rounding error.
- Titrate the degreaser once per shift, the phosphate free acid every four hours, and the final rinse conductivity continuously against a 30 µS/cm limit.
- Free-to-total alkalinity below about 0.7 means the bath is loaded with oil and should be dumped, not dosed.
- Phosphate coating weight by ISO 3892 is the only measurement that covers the whole line: 3.2 g/m² on a 100 cm² coupon is a 32 mg mass loss.
- Titration cannot see temperature, oil in seams or shadowed coverage, which is why the panel test and the water-break check stay on the schedule.
A pretreatment line does not fail suddenly. Free alkalinity slips two points below set point across a shift, the final rinse climbs from 18 to 42 µS/cm, coating weight falls from 2.1 to 1.2 g/m², and the parts still look right coming out of the dry-off oven.
Pretreatment bath titration exists to catch that drift while it is still a chemical problem. Every measurement on the line answers a different question, and a control plan that runs all of them on the same frequency wastes effort on some and misses others entirely.
The numbers below are the ones that hold a seven-tank line in control: what is titrated, to what endpoint, how often, and what each result catches.
What does pretreatment bath titration actually measure?
Titration measures the concentration of the active chemistry, with an endpoint that depends on which tank is being tested.
Free alkalinity on the degreaser is titrated with 0.1 N hydrochloric acid to the phenolphthalein endpoint at about pH 8.3. It represents the alkali available to saponify oil right now. The result is quoted in points, where one point is one millilitre of 0.1 N titrant consumed by a 10 mL sample.
Total alkalinity continues the same titration to about pH 4, using methyl orange or bromocresol green. The difference between total and free is alkalinity already bound up with soils, and it is the closest thing the line has to an oil-loading meter.
Free acid on a derust or descale tank is titrated with 0.1 N sodium hydroxide, and it tells you the acid still available to dissolve oxide. Total acid includes the acid already spent on metal.
Phosphate free acid and total acid are titrated separately and reported as a ratio. Free acid controls the attack on the steel; total acid tracks the bath's overall strength. A free acid within set point and a total acid climbing usually means iron is building up in the tank.
Accelerator is checked by a spot test or a simple drop count rather than a burette. It is the component that most often runs out quietly, because a bath with correct free acid and no accelerator produces a thin, pale, poorly bonded coat.
How often should each measurement be taken?
| Measurement | Method | Frequency | Acceptance | Drift it catches | Ganesh note |
|---|---|---|---|---|---|
| Degreaser free alkalinity | 0.1 N HCl to pH 8.3, 10 mL sample | Once per shift | Within ±10% of set point | Cleaner strength falling as oil load rises | Set point written on the tank card, not in a folder |
| Degreaser total alkalinity | Continue to pH 4 | Once per shift | Free ÷ total ≥ 0.7 | Oil loading, soap build-up | Below 0.7 the bath is dumped, not dosed |
| Derust free acid | 0.1 N NaOH titration | Twice per shift | Within ±5% of set point | Acid spent on scale, dilution by drag-in | Omitted on tanks not in the route |
| Phosphate free acid | 0.1 N NaOH to the standard endpoint | Every 4 hours | Within ±0.5 point of set point | Loss of attack, poor crystal formation | Ratio to total acid logged together |
| Phosphate total acid | Continue to the second endpoint | Every 4 hours | Ratio drifting by more than 10% | Iron build-up in the bath | High iron gives a heavy, porous coat |
| Accelerator | Spot test or drop count | Once per shift | Within the supplier's colour standard | Thin, pale, weakly bonded coating | The quiet failure on a phosphate tank |
| Rinse pH | Meter, calibrated each shift | Once per shift | 5.5–7.5 on the final rinse | Acid or alkali carry-over | Calibration is the test; a drifting probe reads confidently |
| Rinse conductivity | Inline probe, handheld check | Continuous, with a weekly handheld cross-check | Under 500 / 200 / 30 µS/cm for rinses 1, 2 and final | Drag-out, water quality, exhausted demineraliser | Final rinse is the one that decides blistering |
| Bath temperature | Dial or probe, logged | Twice per shift | Within 5 °C of set point | Cleaning rate, crystal growth | A titration cannot see temperature |
| Phosphate coating weight | Gravimetric per ISO 3892 | Per batch, or every 8 hours | 1.5–4 g/m² zinc, 0.3–1.0 g/m² iron | The whole line at once | Coupon cut from the same material as the parts |
| Water-break | Demineralised water, 30 s observation | Every rack or hourly | Continuous film, no beading | Residual oil the titration cannot see | The cheapest check on the line |
The frequencies are not arbitrary. A degreaser changes slowly and a phosphate changes quickly, because the phosphate is being consumed by every square metre that passes through it and its iron content rises with every part.
What does one titration point mean in kilograms?
The arithmetic is worth doing once, because it turns a number in a log into an action.
One point is 1 mL of 0.1 N titrant per 10 mL of sample, which is 0.1 × 1 ÷ 10 = 0.01 equivalents per litre. In a 3,000 L degreaser tank that is 30 equivalents.
For caustic soda at 40 g per equivalent, 30 equivalents is 1,200 g, or 1.2 kg of sodium hydroxide equivalent per point.
So a bath reading 6 points against a set point of 10 needs 4 × 1.2 = 4.8 kg of caustic equivalent to return to set point. That is a real addition, measured and recorded, not a splash from a jug.
The same arithmetic on a derust tank shows why acid additions are more delicate. One point in the same 3,000 L is 30 equivalents of acid, and at 36.5 g per equivalent for hydrochloric acid that is about 1.1 kg of pure HCl, or roughly 2.9 kg of a 38% solution. Adding that to a tank already near its upper limit pushes free acid out of range and starts attacking the steel rather than the scale.
Two rules follow. Titrate after every addition to confirm it landed, and never dose twice for the same reading because the first addition "did not seem to do anything".
Why is temperature not in the titration?
Because temperature changes the rate of every reaction on the line without changing any concentration.
Alkaline cleaning rate roughly doubles for every 10 °C rise. A degreaser held at 45 °C instead of 55 °C cleans at about half the speed, so a part that dwells for four minutes leaves oil that a six-minute dwell would have removed. The titration reads perfectly. The parts fail the water-break test.
The same applies to the phosphate. Crystal growth is temperature-dependent, and a bath five degrees cold produces a fine, thin coating at a concentration that would give a normal coat when hot. This is why the temperature log and the titration log have to be read together, and why a bath that is chemically correct but cold is still a bath out of control.
Why does conductivity beat pH on the rinse tanks?
pH is a logarithmic, buffered measurement. A rinse carrying 0.05% of drag-out can hold a near-neutral pH and still deposit enough salt on the part to blister a powder film in service. Conductivity responds linearly to dissolved solids and moves as soon as drag-out starts accumulating.
Three limits cover a seven-tank cascade: under 500 µS/cm after the degreaser rinse, under 200 µS/cm before the conversion coat, and under 30 µS/cm on the final demineralised rinse. The final number is the one that shows on the customer's salt spray result, because salts left under the film pull water through it.
Municipal water in most Indian industrial areas runs 200–600 µS/cm, which is why the final rinse is demineralised rather than tapped off the mains, and why a demineraliser that is running on bypass looks exactly like a rinse tank that is working.
What does coating weight tell you that titration does not?
Coating weight is the only measurement that integrates the whole line: concentration, temperature, dwell, activation and rinse quality all show up in one number.
ISO 3892 covers the gravimetric method: weigh a coated coupon, strip the conversion coating with the prescribed reagent, weigh again, and calculate mass per unit area from the mass loss and the coated area. The standard notes that the coupon area has to suit the expected coating mass — roughly 1,400 cm² at 1 g/m² down to 25 cm² at 50 g/m² — or the mass loss is too small to weigh reliably.
A worked example: a 100 cm² coupon loses 32 mg. Mass per unit area is 10 × 32 ÷ 100 = 3.2 g/m², comfortably inside the 1.5–4 g/m² zinc phosphate range. A coupon that loses 12 mg gives 1.2 g/m², which reads as a stain rather than a coating and will fail adhesion before it fails salt spray.
Measurement uncertainty on the method is around 5% under good conditions, so a 3.2 g/m² result is 3.0–3.4 g/m² in practice. That is fine for process control and not fine for arguing about a 0.1 g/m² specification limit.
What does iron build-up in a phosphate bath do?
Total acid climbing while free acid holds steady is the signature of dissolved iron, and it changes the coating in a way that titration alone will not flag as a fault.
Iron dissolved from the parts stays in the bath and is incorporated into the phosphate crystal. At low levels it is normal and even helpful, because it accelerates coating formation on steel. Past a threshold — commonly around 1–2 g/L of dissolved iron for a zinc phosphate — the crystals grow coarse, the coating weight rises above the range and the layer becomes porous and powdery. A powdery coat rubs off on the rack, drops out of the film, and leaves a surface that looks phosphate-coated and behaves like bare steel.
The visible clue is colour and texture: a normal fine zinc phosphate is a light to medium grey with an even crystalline surface, and an iron-rich bath produces a darker, uneven, sometimes bluish coating. The measured clue is coating weight above 4 g/m² with a total acid climbing over consecutive shifts. The action is a partial dump and remake, not more accelerator.
What can titration not see?
Four things, and each has its own check.
Oil trapped in seams. A lap joint or a hem holds oil that the bath cannot reach in a four-minute dwell. The titration is correct, the flat face passes the water-break test, and the seam blisters in the oven. Directed jet cleaning and a longer dwell are the fix.
Shadowed coverage. Rack contact points and the inside of a box section may never see fresh chemistry. A rack mark that is bare steel will rust under the film, and no bath reading predicts it.
Rinse quality after the last tank. Conductivity catches dissolved salts, not the film of dried solids left when a part drains slowly. Drainage time and dry-off temperature belong on the same schedule.
Dry-off. Water left in a seam flashes off in the cure oven and blows a pinhole through the powder. That is a dry-off fault with a perfect titration record behind it.
The panel test closes the loop. A coated test panel per shift, checked for adhesion by cross-cut and, on the qualification batch, by neutral salt spray to ISO 9227, is the evidence that the numbers on the log produced a coating rather than a coincidence.
What Ganesh runs on the pretreatment line
Parts are pretreated on a seven-tank line ahead of an 18 ft × 14 ft electrostatic powder booth, with a 16 ft × 16 ft shot blast chamber upstream for scale and rust and a curing oven downstream. Coating is released on measurement: dry film thickness by gauge and adhesion by cross-hatch on the coated part.
The plants hold ISO 9001:2015 and ISO 14001:2015, and the effluent treatment plant handles the process water from the line, which is also why bath life and dump frequency are managed as an environmental control rather than only a quality one. The surface treatment hub carries the tank-by-tank guides, and the plant quality and sustainability hub covers the control and environmental side of running the line day to day. The paint shop capability page sets out the line and booth scope.
The seven-tank process article covers the sequence and the cascade, and the failure diagnosis guide covers what to check when a batch fails despite a clean log.
What the finish specification should require
- Substrate, including whether any part is galvanised, and the conversion coat type expected.
- Coating weight range in g/m², with the test method named as ISO 3892.
- The adhesion test and class, and the panel it is performed on.
- A corrosion test to ISO 9227 if the service environment demands one, with hours, scribe and acceptance stated.
- Records: titration, temperature, conductivity and coating weight per batch, available with the dispatch documents.
- Rack mark locations that are acceptable, agreed before the first article rather than argued after it.
When titration is the wrong control
Titration is a control for a bath that is in continuous use. On a line that dumps and remakes its tanks every week, or on a small job shop running two racks a day, the daily burette work can cost more than the risk it manages, and a weekly coating weight plus a panel test is the better instrument.
It is also the wrong instrument when the failure is mechanical. A spray nozzle that has blocked, a rack that is not making contact, or a part hanging in a shadow will produce a reject with every bath number in range. Check the physical route before dosing the tank.
And it cannot substitute for the customer's own qualification test. A bath in perfect control still has to prove the coating on the part, and the degreasing guide for mild steel explains what the cleaning stage has to achieve before any of the numbers mean anything.
Send the drawing and the corrosion requirement through the quotation route, and ask for the control plan with the quotation. A supplier who cannot describe the titration schedule is unlikely to have one.
Frequently asked questions
How often should a degreasing bath be titrated?
Once per shift as a minimum, and again after any top-up or a change of part mix. A bath that has been dosed should be re-titrated to confirm the addition actually landed at the set point, because a mis-measured addition is worse than no addition.
What is the difference between free and total alkalinity?
Free alkalinity is the active cleaner strength, titrated to the phenolphthalein endpoint at about pH 8.3. Total alkalinity goes further, to about pH 4, and includes the alkalinity already consumed by oil and soils. The gap between them is a measure of how loaded the bath is.
Why measure rinse conductivity instead of rinse pH?
Conductivity responds linearly to dissolved salts from drag-out and to water quality, while pH is logarithmic and buffered by the residue itself. A final rinse can read a healthy pH and still leave 60 µS/cm of salts on the part, which is what blisters the powder.
What coating weight should a zinc phosphate bath produce?
1.5–4 g/m² for a zinc phosphate on steel and 0.3–1.0 g/m² for an iron phosphate. Above the range the coating is coarse and powdery; below it the surface is only stained and the powder has nothing to key to.
Can titration results be trusted if the bath looks clean?
No. Bath appearance says nothing about concentration, and dark or bright colour is usually a function of dissolved iron rather than strength. The number in the log is the control; the look of the tank is not.
Sources and further reading
- ISO 3892:2000 — Conversion coatings: determination of coating mass per unit area, gravimetric methods · reviewed 3 Nov 2026
- ISO 9227:2022 — Corrosion tests in artificial atmospheres: salt spray tests · reviewed 3 Nov 2026
- BIS — Indian Standards for methods of test for industrial water and coating processes · reviewed 3 Nov 2026
Need pretreatment records with the parts?
Send the substrate, the finish and the corrosion test you have to pass. The paint shop team reviews the tank sequence and the control plan before quoting.
Request a coating quote