DI Water Resistivity Control in the Final Rinse

Resistivity is the resistance of water to carrying an electric current, and in a PCB final rinse it is used as a continuous indicator of how much dissolved ionic material the water is still picking up. Deionised water with nothing dissolved in it reads around 18 megohm-centimetres at 25 degrees Celsius, and every ion that enters the water lowers that figure. The reading is therefore a direct measure of rinse performance rather than of the board.

What the Reading Represents

The figure is a bulk property of the water, not a measurement of the board. It tells the operator how clean the incoming water is and how much contamination the rinse stage has removed, but it cannot say which ions are present or where they came from.

That limit is useful rather than inconvenient. A resistivity monitor responds immediately to a change in rinse loading, which makes it a control instrument, while a laboratory extract test run on a sample of boards is what confirms ionic contamination on the surface itself.

Why the Final Rinse Decides Cleanliness

Each rinse stage dilutes the chemistry carried over from the one before it, and the last stage is the one that leaves its residue on the panel. Water that is nearly pure leaves almost nothing behind when it dries, while water carrying ions deposits them as the droplets evaporate.

The mechanism explains the classic result that a board can be clean after the wash and dirty after the dry. If the final rinse water is loaded, the drying step concentrates its dissolved solids onto the surface, and the spots left behind are a map of where the droplets sat rather than of where the contamination came from.

Targets and Their Meaning

A common requirement is that the final rinse supply is above 1 megohm-centimetre with a target nearer 10 to 18, while the resistivity of the water leaving the rinse stage after a board load is allowed to fall and then must recover to the target before the next panel enters.

The recovery behaviour is more informative than the absolute value. A stage that takes a long time to recover to specification is carrying more ionic load than one that recovers quickly, even when both eventually reach the same reading, and that difference shows up later as a cleanliness failure. For that reason the recovery time is often written into the process sheet as a limit in seconds rather than as a note.

Monitoring and Instrumentation

In-line resistivity cells need a defined sampling point and a temperature compensation reference, because the reading of pure water changes by roughly two percent per degree Celsius. Without compensation, a warm rinse will read lower than a cold one and the trend becomes meaningless.

The probe also has to be located where it sees the water that will actually touch the boards. A cell installed on a supply line upstream of the last spray reads the water quality, while one installed in the sump of the final stage reads the loading that the panels have produced, and both are useful for different decisions.

Rinse Cascade Design

A rinse stage is normally built as a cascade, with the cleanest water entering at the last stage and flowing backwards towards the dirtiest. The arrangement means the panel meets progressively cleaner water, and the last stage it touches is fed by fresh DI water rather than by carry-over from the wash.

The number of stages and the flow rate per stage set how far the dilution can go. Doubling the flow through a single stage does less than adding a second counter-flow stage, and in practice three stages with moderate flow achieve more than two stages pushed hard.

Interaction with Cleaning Chemistry

What the rinse has to remove is decided by the cleaner. A saponifier or a high-alkaline chemistry leaves a larger ionic load than a mild neutral cleaner, so the rinse specification is a function of the chemistry rather than a universal number.

Carry-over is the mechanism that links them. Panels leaving the wash drag a film of chemistry with them, and that film is diluted stage by stage. Where drag-out is large, the first rinse is exhausted quickly and the final stage is asked to do work it was not designed for, which is visible as a long recovery time on the resistivity monitor.

Failure Modes and Their Symptoms

The common failures are a supply that is no longer pure, an exhausted resin bed, a blocked spray nozzle, and a final stage that is bypassed or short-circuited so that wash water reaches the last tank. Each has a different signature in the readings.

A supply problem lowers the reading before any board is processed. A blocked nozzle leaves the bulk reading normal while individual panels come out with high surface ionic contamination. Bypass shows as a reading that never recovers between loads, and a spent resin bed shows as a slow downward drift over days. The quickest check of all is a reading taken before the first load of the day, because a supply fault is present whether or not a panel has been processed.

Verification and Records

The final check is an extract test on the finished assembly, reported as an equivalent of sodium chloride per unit area, with a limit of 1.56 micrograms per square centimetre applied to most assemblies. The extract result is the acceptance evidence and the resistivity record is the evidence that the process that produced it was under control.

Records should carry the resistivity at the start and end of each shift, the recovery time after each load, the DI water supply quality, the resin bed change date and the extract results. A chemistry selection decision made without those records is a guess about how much rinsing the process actually needs. Without those records, a rise in extract results can be attributed to the chemistry, the rinse or the drying step with equal confidence and no evidence.

Water Reuse and Recovery

A rinse line that recovers water still has to deliver the same final quality. Recovery is normally arranged so that the cleanest water, after it has passed the last stage, is returned to the second stage, which keeps the counter-flow principle intact and reduces the volume drawn from the deionising plant without changing what the panel sees at the end.

The risk in recovery is a return path that lets contaminated water move forward. A valve left open, a pump that runs in the wrong direction or a storage tank that is not isolated can all send loaded water into the final stage, and the resistivity monitor will show it as a reading that no longer recovers between loads. Verifying the flow path, and proving the final rinse quality with an extract test rather than with the monitor alone, is what keeps a recovery scheme from quietly becoming a compromise on surface cleanliness.

FAQ

What resistivity should the final rinse hold? A supply above 1 megohm-centimetre is a common minimum, with most lines targeting 10 to 18 megohm-centimetres for water that will dry on the board.

Does a good resistivity reading prove the board is clean? No. It shows that the water is clean and that the stage is recovering, while the surface condition itself is confirmed by an extract test against the 1.56 microgram limit.

Why does the reading drop when boards enter? Because the panels carry chemistry and ions into the tank, which is expected. What matters is how fast the stage recovers to its target before the next load.