Part of the resulting inaccuracies in the process are caused by the machines being used. In addition to the selection of the machine type, the wear and maintenance condition play a decisive role. Added to this are the inaccuracies of the components and wiring devices used.
Thus, in practice, the placement precision is shown at the position of the connections of the mounted component with respect to the pads. This results in a mixture of the inaccuracy of the component and wiring carrier geometries with the inaccuracy of the placement equipment. The same applies to solder paste printing and dispensing. The production of a PCB is a chain process with interactions between the individual process steps.
This means that the accuracy of the equipment cannot be reliably determined on real products with real components because an inaccurately produced component does not necessarily cause an issue in the placement machine. A separation of the stated inaccuracies is generally not feasible. Furthermore, the use of quality data from the test is difficult and while these are generally countable features, i.e., the module is functional or not functional; a conclusion about the cause can be difficult.
In addition, quality data are only available with a considerable time delay. A meaningful machine capability assessment (MCA) is not effectively possible considering the requirements of mechanical engineering to "measure attributes on the product". The use of the gained knowledge to improve the process in the sense of a quality control loop is limited.
For this reason, the use of dummy products has been established (for example according to VDI/VDE3713 or IPC9850). These dummy products are replicas of printed circuit boards and components which are mostly made of glass. They have typical properties of real products but without their inaccuracies. The idea is to check each equipment individually. The insights gained to improve the accuracy can then be fed back directly. This allows an objective assessment of the machines involved in the process to be carried out.
It is useful to combine this approach with the daily quality (SPC) data. In this way, it is possible to detect an out of control process condition at an early stage and take suitable steps to correct the issue.