DI Water Resistivity: 5 Rinse Control Rules
DI water resistivity is the number that tells a wet process line whether its rinsing is actually working. Deionised water conducts almost nothing, so any fall in resistivity is caused by ions that came off the panel or out of the previous tank, and that makes the reading a direct measure of how much contamination is being carried forward.
The measurement is simple, which is why it is so often trusted without being set up properly. A probe in the wrong place, a rinse with no flow control or a drip time that varies with the operator will produce a stable looking number that says nothing about the panel. Ionic contamination left on the board becomes leakage, corrosion and coating failure later.
What DI Water Resistivity Measures
Resistivity and conductivity are two ways of expressing the same property, and they move in opposite directions. Pure water at twenty five degrees sits near eighteen megohm centimetres, and as dissolved ions increase the resistivity falls. A rinse reading of one megohm centimetre therefore says that the water holds roughly eighteen times more conductive species than pure water does.
What the probe cannot see matters too. Resistivity responds to ionic species, not to organic residues, particles or flux that has not been converted to a soluble form. A line can hold a perfect resistivity reading while panels still carry organic soil, which is why the probe belongs beside process checks rather than in place of them.
Rinse Cascades and Counter-Flow Design
A cascade with two or three tanks in counter-flow uses far less water than a single tank for the same result. Fresh water enters at the last tank, which is the one the panel sees last, and overflows backwards toward the dirty end. The panel therefore meets progressively cleaner water, and the tank with the highest contamination is also the one that receives the least fresh supply.
Flow direction and panel direction are the whole design. If the cascade is plumbed so that fresh water enters at the first tank, the panel meets clean water early and dirty water at the end, which is worse than a single well-stirred tank. Verifying the plumbing is a one-off check that is worth repeating after any maintenance. Our notes on deionised water rinse cover the same layout rules.
Drag-Out: The Load the Rinse Must Remove
Drag-out is the volume of process solution that leaves a tank on the panel, the rack and the rollers. It is the only reason the rinse has to work at all, and it rises with viscosity, with the surface area of the work and with how fast the panel is pulled out of the tank. Slower withdrawal and a longer dwell over the tank both cut drag-out without costing water.
Where a process uses a strong chemistry, a dedicated drag-out tank before the rinse saves money twice. It recovers solution that would otherwise be diluted into the rinse and sent to waste treatment, and it lowers the ionic contamination the rinse has to remove. Chemical suppliers publish recovery figures, and the payback is usually measured in months.
Setting Flow Rate and Drip Time
Flow rate should be set from the drag-out load and the target resistivity, not from the size of the pipe. A useful rule is to supply enough water per square metre of panel treated to dilute the carried chemistry below the limit, and then to verify it with the probe at the end of the shift when the tanks are at their dirtiest.
Drip time is the parameter that operators shorten when the line is busy, and it is the one that most often causes escape. A defined drip of ten to thirty seconds above each tank is cheap to hold and easy to audit, while a rinse that relies on the conveyor speed alone will pass a full shift of panels through with a fraction of the intended contact time.
Resistivity Limits by Process Step
Different steps need different limits. A rinse after an alkaline cleaner can tolerate a lower resistivity than the final rinse before lamination or before a coating, where residual ions sit under a layer that will never be removed. Setting one number for the whole line either wastes water at the rough end or leaves a risk at the critical end.
The final rinse before drying is the one to watch, because anything left on the surface stays there. Where the panel is going into a solder mask cure or a conformal coat, the surface has to be free of salts as well as dry, and the resistivity of the last tank is the evidence that it is. Contamination limits for assemblies are described in cleanliness requirements.
Ionic Contamination and the ROSE Test
Rinse resistivity measures the water, while the ROSE test measures the board. In that test the assembly is washed in a known volume of an alcohol and water mixture, and the change in conductivity of the extract is converted into an equivalent amount of sodium chloride per unit area. It is the standard way to prove that a cleaning process works.
The two measurements answer different questions, and a line should use both. Resistivity catches a rinse that has drifted in real time, and the extraction test confirms what the finished product actually carries. Extraction methods and their equipment are described by ionic contamination testing.
Resin Beds, Probes and Maintenance
Deionised water is only as good as the resin bed that produces it. Resin that is exhausted or channelled passes ions through, and the resistivity at the point of use falls even though the supply system looks healthy. Regeneration cycles should follow throughput, and the point-of-use reading should be compared with the supply reading when a fault is suspected.
Probes drift and foul. A probe coated with a film of salts or organics reads low and a probe with a damaged electrode can read high, and both faults look like a process problem. Calibration with a certified standard solution against a recorded interval, not against a memory of the last value, keeps the number honest.
Records, Trending and Troubleshooting
The useful record is a trend rather than an entry. Resistivity against panels processed, or against shifts, shows a slow decline that a single reading hides, and the decline usually begins before any defect appears. It also shows the effect of a flow change, which is the fastest way to confirm that a fix actually worked.
When a fault does appear, work backwards through drag-out, flow and contact time in that order. Most rinse escapes are hydraulic rather than chemical, and a cleaning validation carried out with assembly cleanliness measurement will confirm whether the surface is genuinely clean or merely wet, and the reference methods come from IPC.
Additional Considerations for This Build
Practical attention to rinse water pays for itself here, because it decides whether the finished board behaves as the drawing intended. Where the requirement is not stated on the fabrication drawing or in the assembly notes, the shop has to assume a default, and that default is rarely the value the design was simulated with. Stating rinse water explicitly, together with the tolerance that applies, removes the assumption and keeps the result predictable from batch to batch.
FAQ
What resistivity should a final rinse hold? The exact figure follows the product and the next process, but a final rinse that falls below a few megohm centimetres is usually carrying enough ionic load to leave residue. The tank should be sampled at the end of a shift, when it is dirtiest.
Does warmer water rinse better? Warmer water dissolves residues faster and lowers drag-out viscosity, so it cleans more effectively at the same flow. It also exhausts resin faster and can raise the ionic mobility that the probe sees, so the change should be verified rather than assumed.
Why does the rinse reading recover after a stoppage? Water continues to flow while no panels are processed, so the tank dilutes back toward the supply value. That recovery is not evidence that the rinse is adequate for a running line, and it is the reason samples should follow production.
