Among the different techniques for Deterministic Leak Test procedures, one of the most interesting and effective is the HVLD High Voltage Leak Test.
This method is known and applied for many years, at least from the early 70ies of last century, but originally it has been developed for application in the Food and Beverage industry, and many applications were for inline or online testing of large containers, typically bottles, to have early warning in case that filling or sealing process was not working properly. The high speed of the lines and the size of the tested container did not allow for very high accuracy and resolution, but it was more than acceptable for need of the application.
Later, around the end of same decade, the first application in pharma industry, for ampoules, was introduced. Since then, several manufacturers have been offering inline machines for 100% testing of inline heat-sealed containers, such as glass ampoules or BFS Blow Fill and Seal. The current state of the art mandates to 100 % testing for this class of containers, while sampling and offline testing is accepted for filled and stoppered containers.
The principle, see Figure 1, is that, when the container is positioned between electrodes where High Voltage is applied, a weak current can be observed flowing across the sealed container.
Where: HV is the High Voltage Generator
C1 and C2 are the non-conductive container walls on the two sides, plus air gaps.
R is the resistance of the product
When there is a hole on the wall, the capacitance in that point disappears and the point becomes conductive. The missing capacitance results in increased flow of current, which is detected.
It is required that the product inside the container is liquid and it presents a minimal level of conductivity (1-5 micro Siemens/cm or more); the current peak on a bad sample is more evident when the defect is facing one of the electrodes. The laboratory application of this method allows to get results also for extremely small defects (below 1 micron nominal size, down to 0.2), it is anyway fast (seconds) and it is deterministic in the sense that you can set a minimum and maximum level of signal, defined in the phase of method development, to pass or fail your sample without subjective decision.
Even though you can normally observe that bigger leak generates higher signal, there is no theoretical fundament to correlate the signal level to a defined defect nominal size.
Normally, the container is exposed to HV in a way that can better result in current peak in case of defect being present: thus, for containers having approximately cylindrical form, the sample is measured laying horizontal and in rotation on longitudinal axis: this is to expose all surface to electrodes in the same way and to make sure that liquid homogeneously wets all surfaces.
For containers where rotation is not a practical option (IV bags, BFS blisters, …) other solutions are chosen to obtain the most effective exposure to HV.
To summarize the pros and cons of the HVLD method:
- Container must be non-conductive (no metal surface, no Al composite multilayer)
- Product inside must be liquid and have at least 1 micro Siemens conductivity, no oil based jelly product.
Due to flammability, alcohol-based product cannot be tested. - Containers filled with powder only (lyo) cannot be tested.
- The air gap between electrodes and container surface, must be very small, thus a proper design of sample holder and electrodes must be used, customized for type of container.
- IV bags or other flexible containers can be tested (providing there is no metal layer in the film)
- No sample preparation is required, very simple test operation
- HVLD method is deterministic and non-destructive
- Inline HVLD instruments are available for 100% testing.
- The use of very short measuring time and not too high Voltage minimizes the risk of product damage.
- HVLD in lab version allows to study which area of the container is subject to present defects.
- A well designed HVLD lab unit, that can also be used at-line, can be available at a price around 90.000 US$, which makes it competitive with the best Vacuum Decay
This makes HVLD most likely the preferred method, together with Vacuum Decay, for a CCIT modern deterministic lab instrument.
