Board testing [Archive] - pcbstandards forums

Smelter

02-14-2007, 12:58 PM

Can anyone guide me the procedures of board testing after the board is manufactured.

1) Steps of board testing after and before assembling of components.

2) Will they use test vias in board for these testsing purposes for checks short and opens.

Perhaps this extract from the CD "Cost Effective Manufacture by Design" issued by The Printed Circuit Interconnection Federation, may answer your questions.

-------

While physical inspection can verify the mechanical; reliability of the assembly, it cannot guarantee that the circuit will function correctly. This can only be done by electrically testing the assembly.

The objectives of testing are:
a) To detect the presence of all possible failure mechanisms.
b) To determine the cause of failure with sufficient accuracy so as to allow an effective repair to be carried out to either the product, the process, or both.

The first objective is called the pass/fail objective. The second is called the diagnostic objective.
In order to reach these objectives a suitable test programme must be written which will run on a specific ATE, and effectively test the board.

The current range of is as follows:
a) Manufacturing Defect Analyzers (MDA) Sometimes called pre-screeners, based on bed-of- nails access to internal nodes followed by simple impendence measurements to determine the presence or absence of shorts and opens, or missing components.
b) In-circuit testing (ICT) As for MDAs but including an ability to backdrive through the nails in order to apply standard test patterns to the input side of a component, with response monitoring on the output side of the component.
c) Functional Board Testing (FBT) Test stimuli are applied to the normal inputs of the board. The output is then monitored.

Each of these techniques have been used for conventional boards and can be applied to surface mounted boards successfully. Some differences are to be expected. These differences will be mainly in the test generation programme since the fault spectrum will change. Broadly speaking, faults can be classified as either related to manufacturing assembly process or to a component malfunction. The change in technology should not bring about any appreciable change in the malfunction category, but there will be changes in the assembly category.
In particular:
a) Open circuits will dominate and be caused by:
- inadequate solder paste (tombstoning)
- misaligned leads
- non-coplanar leads
- wetting or contamination problem.
b) The risk of placing the wrong component onto a particular location increases for hand and semi- automatic assembly because there on some occasions, no markings on the components.
c) The risk of misalignment also increases. The worst case would be a total sideways displacement by one pad.
d) Even if the component is correctly placed it may be of the incorrect polarity. Some polarised SMCs are not clearly marked for polarity.
e) The smaller physical dimensions of the device pinouts may lead to an increased number of pin-to- pin short circuits.

The other difficulty regarding test of SM assemblies is the nail access to all signal nodes. If the driving force behind surface mounted assembly is miniaturisation, then as the density of the board increases, so internal access becomes limited to the point that in-circuit probing becomes impossible. At this point, other options will have to be considered including the use of built-in self-test within the product, no board test other than visual inspection followed by systems test and functional test only.

---

As a foot note, setting up a bed of nails test rig usually requires a commitment to a reasonable volume of boards to justify the expense. Most sizable PCB manufacturers have "flying probe" test rigs which are programmable for each job, rather that being dedicated for testing a specific board.

Hope this helps.