Process testing in a cleaning laboratory is a typical and useful step in the development of an industrial cleaning process. In most cases, test cleaning is performed in laboratories maintained by the suppliers of cleaning equipment and/or cleaning chemistries. The goal of lab testing, of course, is to simulate a cleaning process to determine its effectiveness and, in some cases, to verify that the chosen cleaning method will not have any detrimental effect on the item(s) being cleaned. For practical reasons, tests are usually limited in scale with just a few parts representing as many as thousands of parts in the eventual full scale implementation. Although it is a valuable tool, laboratory testing of cleaning processes does have some serious limitations which, if not considered, may lead to a process which, although laboratory verified, is not applicable on a large scale.
A previous blog discussed the difficulty in obtaining representative parts for lab testing. But an even bigger and often overlooked problem is developing a cleaning process and the necessary equipment to sustain cleaning over the long term. The cleaning of a few parts in a laboratory setting using freshly mixed chemistry and fresh, uncontaminated rinses seldom represents what the process will look like once thousands of parts have been cleaned in a production setting. Chemistry, by its very purpose, becomes contaminated with both soluble and non-soluble contaminates which have been removed from parts in the cleaning process. Active components of the chemistry become depleted over time. Rinses become contaminated with chemistry and contaminants carried over from the cleaning process. In order to be effective over the long haul, a process must include provisions to maintain the parameters required for good cleaning. The challenge is to anticipate what parameters may degrade in the long term and make provisions to prevent or counteract these changes.
In some cases the cleaning test lab personnel will know what might be a “gotcha” from experience with earlier, similar applications. There is a great benefit to having experience on your side when establishing a new cleaning process. But there are also some quantitative ways to anticipate long term effects. The “dirt in = dirt out” concept may be useful. If you know that a manufacturing process consumes two barrels of coolant per week, for example, the cleaning process must be able (in the worst case) to tolerate the introduction of two barrels of coolant a week. Granted, some of the two barrels will be lost to evaporation and other losses but if one anticipates two barrels a week, the process should be safe. Another useful technique is to measure the amount of contamination on a typical part or group of parts and use a multiplier to determine the total amount of contamination the cleaning process will have to tolerate over time. This technique can be applied with both liquid and solid contaminants to determine the required chemical dosing and assist in sizing filter retention and holding capacity to assure a reasonable service life between filter changes.
I do not have the time or space here to list all of the considerations involved in the development of a cleaning process and the specification of equipment required to perform it. I do implore the reader, however, to know that these considerations exist an must be a part of the overall process development procedure. A little foresight and ingenuity in the testing stage goes a long way toward preventing shortcomings when the process is scaled to full production.
– FJF –