Which type of capacitor has lower 1/f noise: MLCC or film?

  • Thread starter Thread starter eq1
  • Start date Start date
  • Tags Tags
    Noise
Join the discussion
Registration is free. Start your own thread to ask a follow-up.
3 replies · 3K views
eq1
Messages
206
Reaction score
71
I am working on a low noise capacitive feedback TIA where I need to achieve a pretty low voltage/current input referred noise and so I started researching components. For the resistors and transistor I know about 1/f noise, etc. and I'm confident in my choice but it dawned on me, what about the capacitor?

I've read about piezo effects in ceramics, buzz in film, and all the various ways capacitance can change on you but I haven't seen much on how dielectric choice effects inherent noise.

Basically I am trying to pick between a 47pF 0603 film or a NP0 MLCC. It seems like antidotally, in the acoustic band anyway, they both have pros/cons. Input referred noise at the level I'm working at is a pretty labor intensive measurement to take so I'm really hoping to pass on the first try. :)
 
Last edited:
Engineering news on Phys.org
TIA = transimpedance amplifier

Capacitors do not generate 1/f noise in the same way a resistor does.

MLCC = MultiLayer Ceramic Capacitor = skycap; are designed for use on power supplies, when they leak they will self-heal if across a low impedance supply, but they remain leaky if used for low-power signals. While they are leaky they will make noise.

Use silver-mica, or film, not MLCC for signals.
 
  • Like
  • Informative
Likes   Reactions: berkeman, Tom.G and eq1
Baluncore said:
Capacitors do not generate 1/f noise in the same way a resistor does.

On second thought that must be true. If it wasn't we couldn't low pass the white noise spectrum of the resistor, and clearly noise shaping is a thing.

Somehow I am just really surprised by that. I am probably overthinking it. :)
 
It certainly isn't the case that there is NO 1/f noise in capacitors, pretty much everything exhibits 1/f noise at some level, but the 1/f noise should under most circumstances be lower than in a resistor.

Specifically, the dielectrics are intrinsically noisy at some level (even really pure silicone exhibits some dielectric noise) and in some applications this can be problematic, especially for RF/MW applications.

That said, I very much doubt this is something you could need to worry about in this application.
 
  • Informative
Likes   Reactions: eq1