??? 12/13/06 01:21 Read: times |
#129371 - It's not that simple ... Responding to: ???'s previous message |
There are two phases to either of which you could be referring. Neither of these is targeted at ensuring that a particular design works, but, rather is intended simply to catch manufacturing faults.
First, there's random-sample testing, in which a few samples are extracted from the outgoing line and fully "tested," meaning that they're cycled over their entire set of specified environmental and other stress levels, including temperature extremes and variation, electrical noise, EMI, supply variation, shock (hence the name, shake-and-bake) and all manner of within-specified-limits stimuli. Second, there's outgoing inspection, in which the items are physically inspected, and checked for completeness. Sometimes they're checked for basic functionality. The latter depends on the standards of the manufacturing company. This step is often omitted in products shipped to major retailers, e.g. Wal-Mart, etc. so that they're not forced to pay for something that their return policy will cover anyway. (no, that's not a good thing!) On some products, there's separate pre-assembly checkout of the electronics and mechanical components, e.g. in hard disks, where the HDA's are run on an ougoing inspection line before being combined with their electronics. In such cases, parameters such as number of heads, number of sectors, or number of cylinders may be changed in order to accomodate the performance of the individual unit. They may even have their servos rewritten in order to allow for lower data density, which impacts data transfer rate. All this can be done automatically before the HDA is assembled by attaching the PCB to make a whole HDD. The electronics are, of course, subjected to a comparable set of testing and verification measures to ensure that their functional capability is verified. RE |