If you use smd resistors in your audio projects, do you have a preference for thick film or thin film. And if you have a preference, why?
At critical places thin film: less 1/f noise, better accuracy, less thermal distortion due to lower temperature coefficient.
Interesting article, even though VentureCraft's engineering director clearly hasn't a clue what he is talking about:
""By dramatically reducing thermal noise, Vishay's MELF devices are very effective solutions in high-resolution audio products," said Mr. Masaaki Nemoto, engineering director at VentureCraft."
The 1/f noise can be reduced by using thin-film technology, as was done in these MELF devices, but the only way(*) to reduce thermal noise is to cool down the resistors, for example with liquid nitrogen.
(*): Assuming a normal passive resistor, otherwise you can emulate a low-thermal-noise resistor with a feedback amplifier with series and shunt feedback.
""By dramatically reducing thermal noise, Vishay's MELF devices are very effective solutions in high-resolution audio products," said Mr. Masaaki Nemoto, engineering director at VentureCraft."
The 1/f noise can be reduced by using thin-film technology, as was done in these MELF devices, but the only way(*) to reduce thermal noise is to cool down the resistors, for example with liquid nitrogen.
(*): Assuming a normal passive resistor, otherwise you can emulate a low-thermal-noise resistor with a feedback amplifier with series and shunt feedback.
Last edited:
Thin film SMT are metal film and basically fine for any precision use, if suitably sized to avoid thermal issues.
Thick film are ruthenium oxide sintered with glass beads, they are distinctly non-linear compared to metal film.
The smaller an SMT resistor, and the higher the value, the less linear it is - part of this is thermal effects from having exceedingly small volumes of resistance material involved.
Thick film are ruthenium oxide sintered with glass beads, they are distinctly non-linear compared to metal film.
The smaller an SMT resistor, and the higher the value, the less linear it is - part of this is thermal effects from having exceedingly small volumes of resistance material involved.
Hi,
i fully agree to that arguments, since Mrs Jackson (member suzyj) proved that by her measurements. Since that time i only used Minimelfs (sometimes 1206 resistors, nearly same pad spacing and size) in my audio designs.