Useful appnotes for Cap/Resistor selection

[Teem Kaa]

Hey guys!

I'm in a process of selecting a nomenclature for my audio projects, and currently looking for a good appnotes describing pros and cons for a different types of caps and resistors.

I remember I've seen quite few appnotes from Vishay, I guess for that, but can't find them again. Maybe anyone can suggest some?

I just came to the forum, so maybe it's somewhere already? Quick search didn't find any particular.

I'm particularly interested in notes describing pros and cons of SMD details: 0603 vs 0805 vs 1206.
Best one would be with suggestions for a particular manufacturers series?

Thanks!

 

[Osvaldo de Banfield]

The different SMD size varies mainly with power, and the voltage it can support. seriously doubt that the size can affect sound.

 

[Teem Kaa]

I remember note stating that resistor size makes difference for noise figure of resistors.

Trying to find it at the moment.

 

[Osvaldo de Banfield]

In my knowledge (always limited, of course) the noise depends upon the resistance value, not on its size. It is obvious that a larger size will collect more noise from its larger capacitance, but doubt it will be noisier per se.

 

[Conrad Hoffman]

This is a good one- http://www.analog.com/media/en/tech...018494695982855668424783486554001060AN348.pdf
Then be sure to look up the references at the end.

 

[Teem Kaa]

Thanks Conrad!

I've found good one from Panasonic regarding resistor noise.
https://industrial.panasonic.com/cdbs/www-data/pdf/RDM0000/DMM0000COL11.pdf

Take a look for at fig. 14.

It appears that current noise is related to size and it's better to take a larger scale chip resistor.

 

[VS]

In my knowledge (always limited, of course) the noise depends upon the resistance value, not on its size. It is obvious that a larger size will collect more noise from its larger capacitance, but doubt it will be noisier per se.

Unfortunately some statements above are not right. Size does matter and capacitance effectively decrease noise (En=k*T/C).

All resistors exhibits thermal noise in accordance of equation En=sqrt(4*k*T*R*B). Noise increases as function of square root of resistance (as well as temperature and measuring bandwidth). Size does matter also. Larger sizes exhibits less flicker noise (1/f). Flicker noise dominates overall noise at low frequencies with a corner frequency of tens do hundreds of Hz. It dependes upon materials purity and technological aspects in the resistors manufacture. Moreover, resistor type also has a great influence. The less noisy resistors are wirewound then thin film, metal film, thick film, carbon film, and carbon composite as being the worst.

Regards,
VS

 

[DF96]

Capacitance is unlikely to affect noise in the audio band.

 

[VS]

Capacitance shapes noise of thermal sources

 

[wintermute]

I'm particularly interested in notes describing pros and cons of SMD details: 0603 vs 0805 vs 1206.
Thanks!

You don't need an apnote for that....

1206 quite easy to hand solder
0805 bloody difficult to hand solder
0603 I'm not even going to try to hand solder

For really high frequency stuff (probably not an issue in the audio band) ESL will vary with size...

Tony.

 

[AIM65]

I found this Maxim's note very usefull, I now only choose smd cap correctly documented. For example, Murata provide the cap derating curge vs bias voltage.

https://www.maximintegrated.com/en/app-notes/index.mvp/id/5527

 

[VS]

You don't need an apnote for that....

1206 quite easy to hand solder
0805 bloody difficult to hand solder
0603 I'm not even going to try to hand solder

For really high frequency stuff (probably not an issue in the audio band) ESL will vary with size...

Tony.

Agree! Also, lower size = lower parasitic elements. The only thing that should concern you is not to use thick film on the signal paths (and feedback elements) if low noise is required. In those particular situations use thin film instead.

 

[DF96]

Capacitance shapes noise of thermal sources

Yes, but not in the audio band - unless the capacitance or resistance is very high.

 

[VS]

Yes, but not in the audio band - unless the capacitance or resistance is very high.

That's right under the assumption that any RC pair have a noise BW (fcn=(pi/2)*fc) much higher that the audio BW.

Regards,
VS