|Ask the Experts|
October 15, 2020
Spotting After DI Water Cleaning
We have a close loop 3 Bed DI System fitted with a resistivity monitor. When the beds are new we generally generate water at approximately 18 Mega Ohms. This system was performing well until recently. We had been running 2-3 months between bed changes, now we are changing the beds out every 2 weeks.
We are seeing visible spotting on surfaces yet the puzzling thing is that the water resistivity reads 18 Mega Ohms. This spotting immediately goes away after a bed change. We've double checked the resistivity meters. Do you have any idea what may be causing this?
|Expert Panel Responses|
This appears to be some minerals which are still in the water. We had an issue similar when working for another company and we found that the town switched its water supply during the year, from a lake reservoir to wells and it created havoc in the manufacturing process until we figured this out. So please check with your town water department to make sure of the water supply consistency.
I would then have the water tested for mineral content before and after the DI columns and have this done on a regular basis. Many times the town will do this work for you, so check.
Vice President, Technical Director
It sounds like there's some organic fouling of the IX resin. Is there a carbon pre-filter prior to the IX beds to remove organics?
Have you analyzed the feed and outlet water to/from the IX beds for total carbon content?
Has there been a process change that matches the decline in bed life?
The spotting could conceivably be due to fine, insoluble material suspended in the water. Such material would not necessarily contribute to reduction in resistivity. I don't consider this to be a likely scenario, however.
I assume that the reason that you are changing beds every two weeks is to get rid of the spotting. Is the resistivity monitor also showing a drop? If not, check the resistivity meter. It's very possible that it is not functioning properly. You should always have a second, hand-held meter to verify the meter on the system.
You don't say what the source of the input water supply is; it is possible that there has been a change in the quality of the water coming in,namely a rise in the amount of mineral content. This can happen when a utility changes sources (different well?), or changes their treatment strategy.
Have your water treatment supplier test your incoming water. They should be able to tell you, based on the test results, how often you would be expected to need to change beds. If this does not match up with what you are seeing, then further troubleshooting of the system is needed.
Have you changed flux?There is something obviously fouling the beds if they are going bad in 2 weeks. Having sold recirc systems for Separation Tech, Trek & Stoelting over the years, experience would point to either a change in the flux or something has contaminated the system.
I remember a situation many years ago where the stainless piping became contaminated between the remote, DI recirc system, & the cleaners. The system was completely purged, flushed with peroxide & the problem went away.
The client needs to let us know if they started to use any cleaning agent or additives (such defoamer etc.) in their wash process. Also, calibration of the built in resistivity sensor needs to be verified with a handheld resistivity-meter just to make sure the sensor is working fine.
At a certain frequency level, the original acquired canisters (for containing resin and carbon) need to be disposed of, and replaced with new canisters (never been used) for the regenerated resin and carbon materials.
The aged canisters could cause problems in the long run. Also, the regenerated resin beads should also be disposed of, and the client should start with fresh, never been processed, resin beads.
They don't have to be concerned with carbon, because each time new,unprocessed carbon is provided to the end user. Please ensure to have a first particle filter, then carbon bed placed in front of the mixed beds (anion+cation).
Application Technology Manager
DI resins will only remove ionizable material and it looks like you are building a concentration of non-ionic compounds in the rinse.
If this is the problem it can be solved by adding a activated carbon inline before the DI tanks to absorb these compounds.
Austin American Technology
You can have very clean, properly de-ionized water coming in and cleaning the soils and fluxes from your CCAs, but a key to proper cleaning is not only to have the right solvent or DI water, but a clean rinse and dry as well. Some drag-out from the cleaning process is always present in the water still left on the CCA after cleaning.
In fact, the better your cleaning process is, the more suspended solids are in the remaining water.If you allow that remaining water to evaporate from your assembly, those suspended solids, oils, and other impurities are allowed to remain on the board as "high-water" marks. If you then bake the board dry, those high water residues are then baked on and can be very difficult to remove.
So you should always be sure a proper blow-off of the remaining water takes place, whether air knives or simple ionizing blow off guns are used. The blow off will take all of the suspended solids away, and then if you bake away any remaining moisture, you will be left with a clean, dry, pristine CCA with very little ionic residues left.
The various scenarios given above are all possible. However, at a different company, a 3 bed DI system (Strong base, cation, mixed bed polisher) was used and a similarly drastic reduction of the tank life observed.Mark Rasmussen, Itron
It was determined that the strong base and cation tanks had been installed in the wrong order once and the positions were copied during subsequent replacements. Tank life returned to normal after the proper order was restored.
|Submit A Comment|
Free Newsletter Subscription
Circuitnet is built for professionals who bear the responsibility of looking ahead, imagining the future, and preparing for it.
Insert Your Email Address