Dear Koh: My opinion is that first you should distinguish between active components (like oscillators, but also EMC sensitive parts like microprocessors, memories, ASIC's and all fast logic IC's) and passive components (like ferrites). For the first type, in any case you cannot be enough confident about the goodness of second source components without testing. My experience tells me that even same vendor, "same-all" active components can have significant spreads in EMC behavior, because of the parasitic (and often uncontrolled) parameters involved. That is especially true for radiated emissions. I remember having sometimes nasty surprises when testing products with "identical" components that were just from different mfg batches. One typical example is the edge rate (rise/fall times) that is usually specified relative to maximum values but not at all for minimum values. So even if the edge rates in the data sheets are identical, you can get different EMI behaviors (even from the same vendor). As for passive components (like ferrite beads, resistors, capacitors etc. used to filter noise at the PCB level) I think you can go with just looking at the data sheet, but I would recommend that you test if there is even a small difference in impedance curves (that could trigger unexpected resonances or change the emissions profile). Whenever the purchase dept guys approach me with second (or third) source EMI-sensitive components (like those that you mention) I follow these steps: 1. First screening based on data sheet (example impedance curves for ferrites). If they at least don't match those of the already qualified/tested components they are discarded. If they look the same or even better (example: higher impedance for ferrites or higher resonance frequency for capacitors) I go on to step 2. 2. Test
Just my opinion... Hope this helps. Paolo -----Messaggio originale----- Da: Koh Nai Ghee [SMTP:[email protected]] Inviato: martedì 19 settembre 2000 22.48 A: EMC-PSTC Oggetto: Component Qualification Group, Can anyone advice on any guide for checking/qualifying second source component, such as crystal, oscillator & ferrite bead, mounted on product is still meeting the emission limits? We are manufacturer of pheripherals cards. Some card has crystal and/or oscillator for digital circuitry. Ferrite bead are used too. As these component will affect the final product EMI level, therefore some qualification/guide in terms of EMI must be considered when qualifying second source component. These components are used in lots of product that we made, it would be costly to perform scan for every model. Every product are tested and pass EMI with initial source components. Can we just made a comparison of the specification of the first source compnent with this second source? If yes, what are the crucial specs/information that must meet the first source component? For crystal/oscillator, beside the Frequency stability & tolerance specs, what other item is crucial. For Ferrite bead, impedance curve, what other item is crucial. If happens that some specs are slightly out, e.g. 2nd source crystal are having 5PPM more than initla source crystal, Can we make use of the product passing margin, say 3dB, and do some calculation to confidently say that the second source component on the card is still EMI compliance. Likewise for oscillator and ferrite bead. If yes, could you advice on an example of such calculation. Many thanks in advance for the replies. Regards Koh ------------------------------------------- This message is from the IEEE EMC Society Product Safety Technical Committee emc-pstc discussion list. To cancel your subscription, send mail to: [email protected] with the single line: unsubscribe emc-pstc For help, send mail to the list administrators: Jim Bacher: [email protected] Michael Garretson: [email protected] For policy questions, send mail to: Richard Nute: [email protected] ------------------------------------------- This message is from the IEEE EMC Society Product Safety Technical Committee emc-pstc discussion list. To cancel your subscription, send mail to: [email protected] with the single line: unsubscribe emc-pstc For help, send mail to the list administrators: Jim Bacher: [email protected] Michael Garretson: [email protected] For policy questions, send mail to: Richard Nute: [email protected]
