What kind of noise in resistor and double poly cap?

what kind of noise in Res and double poly cap ?

Hi all:

I know there is a thermal noise in the resistor,
are there any other noise type in the resistor?

Besides the resistor, what kind of noise in double poly cap or MIM cap?


Thanks.
mpig

Re: what kind of noise in Res and double poly cap ?

Non-metal resistors suffers also on flicker noise

Re: what kind of noise in Res and double poly cap ?

Capacitors do not generate noise.

Re: what kind of noise in Res and double poly cap ?

Film capacitors only make microphonic noise; electrolytic/tantalum have 1/f noise I think because the dielectric is ionic.

Re: what kind of noise in Res and double poly cap ?

mpig09 said:

Besides the resistor, what kind of noise in double poly cap or MIM cap?

Click to expand...

All electronic components, active and passive, devices produce noise. 1/f noise is the most common and is universal. It has less to do with any particular device and is more like a fundamental source (even at absolute zero).

Re: what kind of noise in Res and double poly cap ?

According to fluctuation-dissipation theorem in physics, thermal noise power spectral density is proportional to the real part of the impedance.
So, capacitors with zero series resistance do not generate thermal noise.
That being said - capacitors do accumulate noise form resistors they are connected to.
This noise is called kTC noise, and is very important in some electronic systems (for example - in image sensors, in switched capacitor filters, etc.).

Resistors can have various noise mechanisms - thermal, 1/f (due to carrier concentration or mobility fluctuation - the exact mechanism of 1/f noise is still not fully clear), noise due to carrier trapping and detrapping (generation-recombination-like noise) at defects and grain boundaries, etc.

Re: what kind of noise in Res and double poly cap ?

Poly resistors develop a lot of the net resistance at
the grain boundaries, rather in clean bulk material.
So the noise is not ideal like in a diffused / I2 one.
Also very dependent on process in the details, grain
size and interface qualities are not really a "control
point", only secondhand via sheet resistance (which
tends to be given sloppy limits if the flow is thought
to be "just digital").

POP caps have this as "access resistance" to the
top and bottom plates and area-for-area, twice the
distance-in-poly as for a MOS cap (which also would
be smaller for same C, so less plate resistance).

Re: what kind of noise in Res and double poly cap ?

Hi timof:

timof said:

According to fluctuation-dissipation theorem in physics, thermal noise power spectral density is proportional to the real part of the impedance.
So, capacitors with zero series resistance do not generate thermal noise.
That being said - capacitors do accumulate noise form resistors they are connected to.
This noise is called kTC noise, and is very important in some electronic systems (for example - in image sensors, in switched capacitor filters, etc.).

Click to expand...

==>do you mean the KTC noise of the capacitor that from the resistors when the capacitor and resistor connect together(parallel or serial)?
Each kind of capacitor type (like MIM Cap, POP cap) has this phenomenon?


mpig

- - - Updated - - -

Hi all:

Is there any noise depend on the area of resistor and capacitor in VLSI?
Like poly resistor, nwell resistor / POP Cap, MIM cap...

mpig

Re: what kind of noise in Res and double poly cap ?

mpig09 said:

Hi timof:

timof said:

According to fluctuation-dissipation theorem in physics, thermal noise power spectral density is proportional to the real part of the impedance.
So, capacitors with zero series resistance do not generate thermal noise.
That being said - capacitors do accumulate noise form resistors they are connected to.
This noise is called kTC noise, and is very important in some electronic systems (for example - in image sensors, in switched capacitor filters, etc.).

Click to expand...

==>do you mean the KTC noise of the capacitor that from the resistors when the capacitor and resistor connect together(parallel or serial)?
Each kind of capacitor type (like MIM Cap, POP cap) has this phenomenon?


mpig

- - - Updated - - -

Hi all:

Is there any noise depend on the area of resistor and capacitor in VLSI?
Like poly resistor, nwell resistor / POP Cap, MIM cap...

mpig

Click to expand...

Thermal noise is applicable only for resistors and the noise power is proportional to R^2 (as expected). The voltage noise Vf/((Hz)^(1/2)) is the 1/f noise (I quote from memory).

Capacitors do not suffer from thermal noise but they are not immune to fluctuation dissipation (even in a vacuum)- I do not have the reference right now though.

Re: what kind of noise in Res and double poly cap ?

mpig09 said:

...................==>do you mean the KTC noise of the capacitor that from the resistors when the capacitor and resistor connect together(parallel or serial)?
Each kind of capacitor type (like MIM Cap, POP cap) has this phenomenon?...........

Click to expand...

Yes, the noise is generated by thermal noise from the resistance in the circuit.
An experiment was done in the early days of testing the theory where they placed a resistor connected to a capacitor separately in an oven.
The KTC noise varied with the temperature of the resistor but not with the temperature of the capacitor, showing that the noise was only due to the resistor.

Re: what kind of noise in Res and double poly cap ?

Thermal noise is generated by any circuit element - not only resistor - that has a non-zero real part of the complex impedance.
Hence, capacitor or inductor with zero series resistance do not generate thermal noise themselves.

Spectral power density of thermal noise generated by resistor R is: P=4kTR.

kTC is a fundamental physical effect, and does not depend on the structure or nature of the capacitor - MIM, MOM, poly-poly, MOS, ...

kTC noise is accumulated on a capacitor when it is connected to a resistor in parallel, or charged from a battery through a resistor.
If capacitor is connected to resistor in series, and they are not connected to anything - there is no noise on capacitor (if its initial charge, before connecting to a resistor in series, is zero, it will stay exactly at zero, doe to charge conservation law).

Re: what kind of noise in Res and double poly cap ?

timof said:

Spectral power density of thermal noise generated by resistor R is: P=4kTR...

Click to expand...

Something is fishy... The quoted formula cannot be true because:

The left hand side is "spectral power density" - energy per unit time in a given freq range (1Hz) and the right hand side does not have freq explicitly. This is therefore white noise- equal power at all frequency.

Because the freq extends to infinity, the total integrated noise power becomes infinite QED

Re: what kind of noise in Res and double poly cap ?

c_mitra - the right hand side units: kTR = [J]*Ohm = [A*V*sec]*[V/A] = [V^2*sec] = [V^2/Hz] - so, Hz units miraculously appear there, if you do some transformations.

Also, if you do Google or Wikipedia search, you will find the formula that I quoted by my memory (that I studied 35 years ago and did not use much since then), for example:

https://en.wikipedia.org/wiki/Johnson–Nyquist_noise

Re: what kind of noise in Res and double poly cap ?

@timof, that is what I am saying; if the left hand side is freq dependent but the right hand side is not, then it must be a const. The spectral power density is independent of frequency.

By the way, please see the talk page of the (wikipedia) article you have quoted. Yes, that was the old formula based on the equipartition principle. Noise is rather complex (even this)

The problem is similar to the ultraviolet catastrophe in connection with the blackbody radiation.

Re: what kind of noise in Res and double poly cap ?

On such occasions, I recall my high school math teacher's joke - "one fool can ask so many questions that even 100 wise man could not answer".

Re: what kind of noise in Res and double poly cap ?

timof said:

Thermal noise is generated by any circuit element - not only resistor - that has a non-zero real part of the complex impedance.
Hence, capacitor or inductor with zero series resistance do not generate thermal noise themselves.

Spectral power density of thermal noise generated by resistor R is: P=4kTR.

kTC is a fundamental physical effect, and does not depend on the structure or nature of the capacitor - MIM, MOM, poly-poly, MOS, ...

kTC noise is accumulated on a capacitor when it is connected to a resistor in parallel, or charged from a battery through a resistor.
If capacitor is connected to resistor in series, and they are not connected to anything - there is no noise on capacitor (if its initial charge, before connecting to a resistor in series, is zero, it will stay exactly at zero, doe to charge conservation law).

Click to expand...

Hi timof ：
kTC noise is accumulated on a capacitor when it is connected to a resistor in parallel, or charged from a battery through a resistor.
If capacitor is connected to resistor in series, and they are not connected to anything - there is no noise on capacitor (if its initial charge, before connecting to a resistor in series, is zero, it will stay exactly at zero, doe to charge conservation law)
==>
1. kTC noise is KT/C noise?
2. Does it mean there is a path between the capacitor and the resistor when they are parallel
to charge or discharge the capacitor, so the thermal noise is accumulated on the capacitor?

Two extra questions:
1. the value of thermal noise is depended on the capacitance or resistance, and we know the difference size(W/L) of the capacitor or resistor can obtain the same value, so does the thermal noise influence the size of the capacitor or resistor?
2. Except the thermal noise, is there any noise generated by the capacitor and resistor?

mpig

Re: what kind of noise in Res and double poly cap ?

1. The noise (fluctuation) of voltage <V^2> is kT/C, and the noise of charge (Q=V*C) is kTC.

2. I don't quite understand this question.
If a resistor is connected in parallel to capacitor, or capacitor is connected to a battery through a resistor - is there a path between C and R? I guess, yes, there is a path, with zero resistance, but how do you use that info?

Extra questions:

1. Thermal noise on resistor depends only on resistor value, not on its W and L.
Thermal noise accumulated on a capacitor also depends only on the value of the capacitor, not on its W and L values.
Other noise sources (like 1/f, RTS,...) may strongly depend on the size of resistor / capacitor.
Some noise sources are caused by defects and carrier capture/release - and since the number of defects depend on the area, the area may have an effect on noise properties.

2. Depending on the material structure and carrier transport details, resistors can display noise mechanisms other than thermal noise - like 1/f noise, etc.
Please also see a post from dick_freebird above, on noise in resistors.
Capacitors themselves do not generate noise (as far as I know), but can accumulate noise from external elements (resistors, transistors,...), or from their series resistance.
An isolated capacitor does not show any noise, as the charge is conserved and V=Q/C (there may be higher order effect, like quantum electrodynamics, fluctuations in vacuum, Casimir effect, etc. - that can be neglected in practical semiconductor applications).

Re: what kind of noise in Res and double poly cap ?

2. I don't quite understand this question.
If a resistor is connected in parallel to capacitor, or capacitor is connected to a battery through a resistor - is there a path between C and R? I guess, yes, there is a path, with zero resistance, but how do you use that info?

Click to expand...

==>Please reference the attached file, that shows my point.

Extra questions:

2. Depending on the material structure and carrier transport details, resistors can display noise mechanisms other than thermal noise - like 1/f noise, etc.

Click to expand...

==>Does it mean the resistor is easy to show two noise mechanisms: thermal noise、1/f noise?

mpig

Re: what kind of noise in Res and double poly cap ?

Yes, when resistor and capacitor are in series, that means the voltage on the capacitor is equal to the voltage on the resistor.
Since voltage on the resistor fluctuates over time, so does the voltage on the capacitor.
When resistor is then disconnected from the capacitor, voltage on the capacitor stays fixed at the voltage that was on resistor/capacitor at the time of the disconnection.
If you repeat connection/disconnection many times, each time the final (fixed) voltage on the capacitor will be different, with the average value of the square of the voltage equal to kT/C.

For some resistors it is easy to show other noise mechanisms (other than thermal noise), for some others (such as an ideal resistor) it is difficult.

Re: what kind of noise in Res and double poly cap ?

Hi timof:

If you repeat connection/disconnection many times, each time the final (fixed) voltage on the capacitor will be different, with the average value of the square of the voltage equal to kT/C.

Click to expand...

==>whether the connection of the resistor and capacitor is parallel or serial, the phenomenon is the same?

mpig